Claude Code

Desktop OS: macOS

Typed value: Yes.

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

• Desktop os: CLI supports macOS, Windows and Linux; Desktop supports macOS/Windows with Linux beta on Ubuntu/Debian. Direct CLI --desktop launch and managed 3P Desktop are separate auth paths.
• Cli: CLI: macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+, x64 or ARM64, 4 GB+ RAM; the built-in sandbox works on macOS, Linux and WSL2 only.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Advanced setup

   Finding: CLI runs on macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+; 4 GB+ RAM; x64 or ARM64; internet required; Bash, Zsh, PowerShell or CMD.

   Locator

   System requirements

   Retrieved

   2026-11-07

2. Desktop application

   Finding: Desktop Code tab has parallel sessions with Git isolation, diff review and comments, app preview, integrated terminal, permission modes, scheduled tasks and Dispatch from phone; needs a paid subscription; Desktop does not accept a Console API key directly.

   Locator

   Sessions; Choose a permission mode; Work in parallel; Dispatch; Coming from the CLI

   Retrieved

   2026-11-07

3. Claude Desktop on Linux (beta)

   Finding: Linux desktop app is beta on Ubuntu 22.04+/Debian 12+; Fedora/Arch should use the CLI; signs in with claude.ai subscription or SSO, not a Console API key.

   Locator

   Requirements; What is not in the Linux beta yet

   Retrieved

   2026-11-07

4. Feature availability

   Finding: Local CLI features work across supported providers. Desktop is an exception: gateway/3P deployments can run Code on Bedrock, Google Agent Platform, Foundry, self-hosted gateways and the Claude API with Console key. Subscription-only features have separate plan restrictions.

   Locator

   Availability by model provider; Features that require a Claude subscription

   Retrieved

   2026-11-07

Desktop OS: Windows

Typed value: Yes.

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

• Desktop os: CLI supports macOS, Windows and Linux; Desktop supports macOS/Windows with Linux beta on Ubuntu/Debian. Direct CLI --desktop launch and managed 3P Desktop are separate auth paths.
• Cli: CLI: macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+, x64 or ARM64, 4 GB+ RAM; the built-in sandbox works on macOS, Linux and WSL2 only.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Advanced setup

   Finding: CLI runs on macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+; 4 GB+ RAM; x64 or ARM64; internet required; Bash, Zsh, PowerShell or CMD.

   Locator

   System requirements

   Retrieved

   2026-11-07

2. Desktop application

   Finding: Desktop Code tab has parallel sessions with Git isolation, diff review and comments, app preview, integrated terminal, permission modes, scheduled tasks and Dispatch from phone; needs a paid subscription; Desktop does not accept a Console API key directly.

   Locator

   Sessions; Choose a permission mode; Work in parallel; Dispatch; Coming from the CLI

   Retrieved

   2026-11-07

3. Claude Desktop on Linux (beta)

   Finding: Linux desktop app is beta on Ubuntu 22.04+/Debian 12+; Fedora/Arch should use the CLI; signs in with claude.ai subscription or SSO, not a Console API key.

   Locator

   Requirements; What is not in the Linux beta yet

   Retrieved

   2026-11-07

4. Feature availability

   Finding: Local CLI features work across supported providers. Desktop is an exception: gateway/3P deployments can run Code on Bedrock, Google Agent Platform, Foundry, self-hosted gateways and the Claude API with Console key. Subscription-only features have separate plan restrictions.

   Locator

   Availability by model provider; Features that require a Claude subscription

   Retrieved

   2026-11-07

Desktop OS: Linux

Typed value: Yes.

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

• Desktop os: CLI supports macOS, Windows and Linux; Desktop supports macOS/Windows with Linux beta on Ubuntu/Debian. Direct CLI --desktop launch and managed 3P Desktop are separate auth paths.
• Cli: CLI: macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+, x64 or ARM64, 4 GB+ RAM; the built-in sandbox works on macOS, Linux and WSL2 only.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Advanced setup

   Finding: CLI runs on macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+; 4 GB+ RAM; x64 or ARM64; internet required; Bash, Zsh, PowerShell or CMD.

   Locator

   System requirements

   Retrieved

   2026-11-07

2. Desktop application

   Finding: Desktop Code tab has parallel sessions with Git isolation, diff review and comments, app preview, integrated terminal, permission modes, scheduled tasks and Dispatch from phone; needs a paid subscription; Desktop does not accept a Console API key directly.

   Locator

   Sessions; Choose a permission mode; Work in parallel; Dispatch; Coming from the CLI

   Retrieved

   2026-11-07

3. Claude Desktop on Linux (beta)

   Finding: Linux desktop app is beta on Ubuntu 22.04+/Debian 12+; Fedora/Arch should use the CLI; signs in with claude.ai subscription or SSO, not a Console API key.

   Locator

   Requirements; What is not in the Linux beta yet

   Retrieved

   2026-11-07

4. Feature availability

   Finding: Local CLI features work across supported providers. Desktop is an exception: gateway/3P deployments can run Code on Bedrock, Google Agent Platform, Foundry, self-hosted gateways and the Claude API with Console key. Subscription-only features have separate plan restrictions.

   Locator

   Availability by model provider; Features that require a Claude subscription

   Retrieved

   2026-11-07

Web client

Typed value: Yes.

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

• Browser: Cloud sessions run from claude.ai/code on desktop browsers.
• Cross device continuity: Cloud sessions persist across devices; Remote Control continues a local CLI/Desktop/VS Code session from phone or browser while the machine stays on; /teleport pulls a cloud session into the terminal; /desktop continues a CLI session in Desktop. A terminal session cannot be pushed to the cloud (handoff is one-way); a desktop-to-phone move was not tested.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Claude Code overview

   Finding: Documents terminal (macOS/Linux/WSL/Windows), VS Code, desktop (macOS, Windows x64/ARM64, Ubuntu/Debian beta), web and JetBrains surfaces; most surfaces need a Claude subscription or Console account; CLI, VS Code and JetBrains also support third-party providers.

   Locator

   Get started tabs; Terminal install; Desktop app; Web

   Retrieved

   2026-11-07

2. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-11-07

Phone access: Native iOS

Typed value: Yes.

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

• Mobile apps: Claude iOS and Android apps (iPad uses the iOS app) have a Code tab for cloud sessions, projects, Remote Control and Dispatch; code never runs on the phone.
• Task actions: Mobile reaches cloud sessions, Remote Control local sessions, or Desktop Dispatch. Dispatch is on Pro/Max/Ultra; the host must stay awake for local execution.
• Notifications: Push notifications to the phone when Remote Control is active (task finished or decision needed), Dispatch notifications, terminal bell/desktop notification settings in the CLI, and a Notification hook.
• Cross device continuity: Cloud sessions persist across devices; Remote Control continues a local CLI/Desktop/VS Code session from phone or browser while the machine stays on; /teleport pulls a cloud session into the terminal; /desktop continues a CLI session in Desktop. A terminal session cannot be pushed to the cloud (handoff is one-way); a desktop-to-phone move was not tested.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Claude Code on mobile

   Finding: The Claude iOS/Android app is a client for cloud sessions, projects, Remote Control and Dispatch (no code runs on the phone). Push notifications when Remote Control is active; local-only commands such as /plugin and /resume do not work; Bypass permissions cannot be selected from the app.

   Locator

   Work from your phone; Get push notifications; Limitations

   Retrieved

   2026-11-07

Phone access: Native Android

Typed value: Yes.

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

• Mobile apps: Claude iOS and Android apps (iPad uses the iOS app) have a Code tab for cloud sessions, projects, Remote Control and Dispatch; code never runs on the phone.
• Task actions: Mobile reaches cloud sessions, Remote Control local sessions, or Desktop Dispatch. Dispatch is on Pro/Max/Ultra; the host must stay awake for local execution.
• Notifications: Push notifications to the phone when Remote Control is active (task finished or decision needed), Dispatch notifications, terminal bell/desktop notification settings in the CLI, and a Notification hook.
• Cross device continuity: Cloud sessions persist across devices; Remote Control continues a local CLI/Desktop/VS Code session from phone or browser while the machine stays on; /teleport pulls a cloud session into the terminal; /desktop continues a CLI session in Desktop. A terminal session cannot be pushed to the cloud (handoff is one-way); a desktop-to-phone move was not tested.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Claude Code on mobile

   Finding: The Claude iOS/Android app is a client for cloud sessions, projects, Remote Control and Dispatch (no code runs on the phone). Push notifications when Remote Control is active; local-only commands such as /plugin and /resume do not work; Bypass permissions cannot be selected from the app.

   Locator

   Work from your phone; Get push notifications; Limitations

   Retrieved

   2026-11-07

Phone access: Phone browser

Typed value: Yes.

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

• Mobile browser: A phone browser can open the Remote Control session at claude.ai/code and send instructions. An eligible subscription and an awake local host are required.

Checked

2026-11-13 · checked 2 days ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Continue local sessions from any device with Remote Control

   Finding: Remote Control connects claude.ai/code or the mobile apps to a local Claude session. Users can send messages from the terminal, browser and phone; the local host must remain running. Eligible Claude subscriptions are required.

   Locator

   Introduction; Connect from another device; Requirements

   Retrieved

   2026-11-13

CLI

Typed value: Yes.

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

• Desktop os: CLI supports macOS, Windows and Linux; Desktop supports macOS/Windows with Linux beta on Ubuntu/Debian. Direct CLI --desktop launch and managed 3P Desktop are separate auth paths.
• Cli: CLI: macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+, x64 or ARM64, 4 GB+ RAM; the built-in sandbox works on macOS, Linux and WSL2 only.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Advanced setup

   Finding: CLI runs on macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+; 4 GB+ RAM; x64 or ARM64; internet required; Bash, Zsh, PowerShell or CMD.

   Locator

   System requirements

   Retrieved

   2026-11-07

2. Configure the sandboxed Bash tool

   Finding: OS-enforced sandbox for shell commands on macOS, Linux and WSL2 (off by default; native Windows runs unsandboxed); built on open source sandbox-runtime; reads allowed broadly by default, writes limited to working directory, network via domain-allowlist proxy; file tools, MCP servers and hooks run outside it.

   Locator

   What the sandbox restricts; What runs outside the sandbox

   Retrieved

   2026-11-07

Coding agents: Claude Code

Typed value: Yes.

Claude Code embeds its own harness and runs Claude models only. Providers: Anthropic (subscription or Console), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, and gateways that expose an Anthropic-format API. Anthropic does not support routing to non-Claude models and the docs state that workers in every multi-agent mode are Claude sessions.

• Harnesses: One harness: Claude Code itself. Subagents, agent view, teams, workflows and projects all run Claude sessions.
• Model providers: Anthropic API (claude.ai or Console login), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, Claude apps gateway and compatible LLM gateways. Claude models only.
• Custom adapters: Gateways that expose a supported Anthropic API format are supported; other tools can be added as MCP servers. No adapter mechanism for running a different agent harness inside Claude Code was found in the pages checked.
• Session integration: Claude Code is the harness; it can be driven programmatically through claude -p, the Python/TypeScript Agent SDK, or as a stdio MCP server (claude mcp serve).
• Mixed harness operation: Docs state "In every approach the workers are Claude sessions. To involve a different tool, expose it to Claude as an MCP server", so mixed-harness operation is not offered natively.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Scope is the Claude Code harness; unofficial third-party routing to other model families is not covered by the documentation and was not tested.

1. Authentication

   Finding: Account types: Claude Pro/Max, Teams/Enterprise, Claude Console, Bedrock/Vertex(Agent Platform)/Foundry, Claude apps gateway. Multiple accounts via separate CLAUDE_CONFIG_DIR directories. ANTHROPIC_API_KEY approval skips login.

   Locator

   Log in to Claude Code; Log in with multiple accounts

   Retrieved

   2026-11-07

2. Run Claude Code programmatically

   Finding: claude -p runs non-interactively with exit codes, --output-format json/stream-json, --bare, --allowedTools; Agent SDK for Python/TypeScript.

   Locator

   Basic usage; Bare mode; Structured output

   Retrieved

   2026-11-07

3. Agent SDK overview

   Finding: Python and TypeScript libraries run the Claude Code agent loop with built-in tools, hooks, subagents, MCP, permissions and sessions; non-Python/TS languages can drive the CLI as a subprocess.

   Locator

   Compare the Agent SDK to other Claude tools; Capabilities

   Retrieved

   2026-11-07

4. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

5. Connect Claude Code to tools via MCP

   Finding: MCP client supports http, sse, stdio and websocket servers with OAuth 2.0, local/project/user scopes and claude.ai connectors; claude mcp serve exposes Claude Code as a stdio MCP server; channels push events.

   Locator

   Installing MCP servers; Authenticate with remote MCP servers; Use Claude Code as an MCP server; Scopes

   Retrieved

   2026-11-07

6. Feature availability

   Finding: Local CLI features work across supported providers. Desktop is an exception: gateway/3P deployments can run Code on Bedrock, Google Agent Platform, Foundry, self-hosted gateways and the Claude API with Console key. Subscription-only features have separate plan restrictions.

   Locator

   Availability by model provider; Features that require a Claude subscription

   Retrieved

   2026-11-07

7. Other LLM gateways

   Finding: Any gateway exposing a supported Anthropic API format works; Anthropic does not support routing Claude Code to non-Claude models through any gateway.

   Locator

   What a gateway provides; Roll out a gateway

   Retrieved

   2026-11-07

Coding agents: Codex

Typed value: No.

Claude Code embeds its own harness and runs Claude models only. Providers: Anthropic (subscription or Console), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, and gateways that expose an Anthropic-format API. Anthropic does not support routing to non-Claude models and the docs state that workers in every multi-agent mode are Claude sessions.

• Harnesses: One harness: Claude Code itself. Subagents, agent view, teams, workflows and projects all run Claude sessions.
• Custom adapters: Gateways that expose a supported Anthropic API format are supported; other tools can be added as MCP servers. No adapter mechanism for running a different agent harness inside Claude Code was found in the pages checked.
• Mixed harness operation: Docs state "In every approach the workers are Claude sessions. To involve a different tool, expose it to Claude as an MCP server", so mixed-harness operation is not offered natively.

Checked

2026-11-13 · checked 2 days ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Scope is the Claude Code harness; unofficial third-party routing to other model families is not covered by the documentation and was not tested.

1. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

Coding agents: Cursor Agent

Typed value: No.

Claude Code embeds its own harness and runs Claude models only. Providers: Anthropic (subscription or Console), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, and gateways that expose an Anthropic-format API. Anthropic does not support routing to non-Claude models and the docs state that workers in every multi-agent mode are Claude sessions.

• Harnesses: One harness: Claude Code itself. Subagents, agent view, teams, workflows and projects all run Claude sessions.
• Custom adapters: Gateways that expose a supported Anthropic API format are supported; other tools can be added as MCP servers. No adapter mechanism for running a different agent harness inside Claude Code was found in the pages checked.
• Mixed harness operation: Docs state "In every approach the workers are Claude sessions. To involve a different tool, expose it to Claude as an MCP server", so mixed-harness operation is not offered natively.

Checked

2026-11-13 · checked 2 days ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Scope is the Claude Code harness; unofficial third-party routing to other model families is not covered by the documentation and was not tested.

1. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

Coding agents: Gemini CLI

Typed value: No.

Claude Code embeds its own harness and runs Claude models only. Providers: Anthropic (subscription or Console), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, and gateways that expose an Anthropic-format API. Anthropic does not support routing to non-Claude models and the docs state that workers in every multi-agent mode are Claude sessions.

• Harnesses: One harness: Claude Code itself. Subagents, agent view, teams, workflows and projects all run Claude sessions.
• Custom adapters: Gateways that expose a supported Anthropic API format are supported; other tools can be added as MCP servers. No adapter mechanism for running a different agent harness inside Claude Code was found in the pages checked.
• Mixed harness operation: Docs state "In every approach the workers are Claude sessions. To involve a different tool, expose it to Claude as an MCP server", so mixed-harness operation is not offered natively.

Checked

2026-11-13 · checked 2 days ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Scope is the Claude Code harness; unofficial third-party routing to other model families is not covered by the documentation and was not tested.

1. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

Coding agents: OpenCode

Typed value: No.

Claude Code embeds its own harness and runs Claude models only. Providers: Anthropic (subscription or Console), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, and gateways that expose an Anthropic-format API. Anthropic does not support routing to non-Claude models and the docs state that workers in every multi-agent mode are Claude sessions.

• Harnesses: One harness: Claude Code itself. Subagents, agent view, teams, workflows and projects all run Claude sessions.
• Custom adapters: Gateways that expose a supported Anthropic API format are supported; other tools can be added as MCP servers. No adapter mechanism for running a different agent harness inside Claude Code was found in the pages checked.
• Mixed harness operation: Docs state "In every approach the workers are Claude sessions. To involve a different tool, expose it to Claude as an MCP server", so mixed-harness operation is not offered natively.

Checked

2026-11-13 · checked 2 days ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Scope is the Claude Code harness; unofficial third-party routing to other model families is not covered by the documentation and was not tested.

1. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

Coding agents: Pi

Typed value: No.

Claude Code embeds its own harness and runs Claude models only. Providers: Anthropic (subscription or Console), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, and gateways that expose an Anthropic-format API. Anthropic does not support routing to non-Claude models and the docs state that workers in every multi-agent mode are Claude sessions.

• Harnesses: One harness: Claude Code itself. Subagents, agent view, teams, workflows and projects all run Claude sessions.
• Custom adapters: Gateways that expose a supported Anthropic API format are supported; other tools can be added as MCP servers. No adapter mechanism for running a different agent harness inside Claude Code was found in the pages checked.
• Mixed harness operation: Docs state "In every approach the workers are Claude sessions. To involve a different tool, expose it to Claude as an MCP server", so mixed-harness operation is not offered natively.

Checked

2026-11-13 · checked 2 days ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Scope is the Claude Code harness; unofficial third-party routing to other model families is not covered by the documentation and was not tested.

1. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

Coding agents: Grok

Typed value: No.

Claude Code embeds its own harness and runs Claude models only. Providers: Anthropic (subscription or Console), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, and gateways that expose an Anthropic-format API. Anthropic does not support routing to non-Claude models and the docs state that workers in every multi-agent mode are Claude sessions.

• Harnesses: One harness: Claude Code itself. Subagents, agent view, teams, workflows and projects all run Claude sessions.
• Custom adapters: Gateways that expose a supported Anthropic API format are supported; other tools can be added as MCP servers. No adapter mechanism for running a different agent harness inside Claude Code was found in the pages checked.
• Mixed harness operation: Docs state "In every approach the workers are Claude sessions. To involve a different tool, expose it to Claude as an MCP server", so mixed-harness operation is not offered natively.

Checked

2026-11-13 · checked 2 days ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Scope is the Claude Code harness; unofficial third-party routing to other model families is not covered by the documentation and was not tested.

1. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

Coding agents: Aider

Typed value: No.

Claude Code embeds its own harness and runs Claude models only. Providers: Anthropic (subscription or Console), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, and gateways that expose an Anthropic-format API. Anthropic does not support routing to non-Claude models and the docs state that workers in every multi-agent mode are Claude sessions.

• Harnesses: One harness: Claude Code itself. Subagents, agent view, teams, workflows and projects all run Claude sessions.
• Custom adapters: Gateways that expose a supported Anthropic API format are supported; other tools can be added as MCP servers. No adapter mechanism for running a different agent harness inside Claude Code was found in the pages checked.
• Mixed harness operation: Docs state "In every approach the workers are Claude sessions. To involve a different tool, expose it to Claude as an MCP server", so mixed-harness operation is not offered natively.

Checked

2026-11-13 · checked 2 days ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Scope is the Claude Code harness; unofficial third-party routing to other model families is not covered by the documentation and was not tested.

1. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

Coding agents: Antigravity

Typed value: No.

Claude Code embeds its own harness and runs Claude models only. Providers: Anthropic (subscription or Console), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, and gateways that expose an Anthropic-format API. Anthropic does not support routing to non-Claude models and the docs state that workers in every multi-agent mode are Claude sessions.

• Harnesses: One harness: Claude Code itself. Subagents, agent view, teams, workflows and projects all run Claude sessions.
• Custom adapters: Gateways that expose a supported Anthropic API format are supported; other tools can be added as MCP servers. No adapter mechanism for running a different agent harness inside Claude Code was found in the pages checked.
• Mixed harness operation: Docs state "In every approach the workers are Claude sessions. To involve a different tool, expose it to Claude as an MCP server", so mixed-harness operation is not offered natively.

Checked

2026-11-13 · checked 2 days ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Scope is the Claude Code harness; unofficial third-party routing to other model families is not covered by the documentation and was not tested.

1. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

Use my subscription

Typed value: Yes.

Pro, Max, Ultra, Team and Enterprise include Claude Code. Console/API and third-party cloud-provider routes are separate. Free excludes Code; platform API credits bundled with Max/Ultra cannot pay Code usage.

• Agent subscriptions: Pro/Max/Ultra/Team/Enterprise include Code, sharing subscription usage with other Claude clients.
• Vendor credits: Paid usage credits can extend limits; bundled Max/Ultra Claude Platform API credits cannot pay Claude Code or usage credits.
• Account switching: Separate CLAUDE_CONFIG_DIR directories keep settings, session history and claude.ai logins separate, so several accounts can stay signed in; /login and /logout switch accounts. Two Console sign-ins without API keys are not isolated by config directory because their credentials are stored outside it (authentication page).

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Pro, Max, Ultra, Team, Enterprise, Console API, third-party providers

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Local-model statement is the vendor support position; unofficial base-URL workarounds were not tested.
• No trial or promotional pricing beyond the public pricing page was checked.
• Two Console sign-ins without API keys are not isolated by config directory (authentication page); the account_switching atom is partial for that case.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

2. Use Claude Code with your Pro or Max plan

   Finding: Pro, Max and Ultra include Code in a shared subscription quota; Max and Ultra platform API credits cannot pay Code usage. ANTHROPIC_API_KEY causes API billing instead of subscription. No workload usage test.

   Locator

   Connect; usage limits; Claude Code billing

   Retrieved

   2026-11-07

Where agents run: On your computer

Typed value: Yes.

Claude Code runs locally by default; cloud sessions run on Anthropic VMs (4 vCPU, 16 GB, 30 GB) with no separate compute charge; cloud sessions are available on Pro, Max, Ultra and Team plans, and for Enterprise users with premium seats or Chat + Claude Code seats, optionally on customer-run self-hosted runners (Team/Enterprise beta); Remote Control lets other devices drive a local session.

• Local: The CLI runs on the user machine.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-10

Plan

Cloud: Pro, Max, Ultra, Team, Enterprise; self-hosted: Team, Enterprise (public beta)

Surface

Claude Code CLI; Claude Code web/mobile

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Region and exact idle-expiry durations were not found.

1. Claude Code overview

   Finding: Documents terminal (macOS/Linux/WSL/Windows), VS Code, desktop (macOS, Windows x64/ARM64, Ubuntu/Debian beta), web and JetBrains surfaces; most surfaces need a Claude subscription or Console account; CLI, VS Code and JetBrains also support third-party providers.

   Locator

   Get started tabs; Terminal install; Desktop app; Web

   Retrieved

   2026-11-07

Where agents run: On your own server

Typed value: Yes.

Claude Code runs locally by default; cloud sessions run on Anthropic VMs (4 vCPU, 16 GB, 30 GB) with no separate compute charge; cloud sessions are available on Pro, Max, Ultra and Team plans, and for Enterprise users with premium seats or Chat + Claude Code seats, optionally on customer-run self-hosted runners (Team/Enterprise beta); Remote Control lets other devices drive a local session.

• Byo remote: Remote Control on an own always-on machine; Desktop supports configured SSH sessions.
• Self hosted: Self-hosted environments run cloud sessions on customer runners (public beta, Team/Enterprise, not with ZDR/HIPAA).

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-10

Plan

Cloud: Pro, Max, Ultra, Team, Enterprise; self-hosted: Team, Enterprise (public beta)

Surface

Claude Code CLI; Claude Code web/mobile

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Region and exact idle-expiry durations were not found.

1. Self-hosted environments

   Finding: Public beta on Team and Enterprise (off by default); runners on customer infrastructure execute cloud sessions; unavailable with ZDR or HIPAA; model inference still via Anthropic API unless Bedrock/Agent Platform configured; session content goes to Anthropic.

   Locator

   How self-hosted environments work; Availability and limitations

   Retrieved

   2026-11-07

Where agents run: In the vendor's cloud

Typed value: Yes.

Claude Code runs locally by default; cloud sessions run on Anthropic VMs (4 vCPU, 16 GB, 30 GB) with no separate compute charge; cloud sessions are available on Pro, Max, Ultra and Team plans, and for Enterprise users with premium seats or Chat + Claude Code seats, optionally on customer-run self-hosted runners (Team/Enterprise beta); Remote Control lets other devices drive a local session.

• Managed cloud: Cloud sessions run in Anthropic-managed Ubuntu 24.04 VMs.
• Hybrid: Projects threads can run on the user computer via Remote Control; claude --cloud and /teleport move work between terminal and cloud (one direction each).
• Provisioning: Fresh VM per session with a cached setup-script snapshot.
• Payer: Cloud VM use has no separate compute charge; it consumes the account subscription limits.
• Resources: Anthropic-hosted environments: 4 vCPUs, 16 GB RAM, 30 GB disk.
• Expiry: Idle VMs pause after a few minutes and can later be reclaimed; reopening provisions a fresh VM with history restored.
• Client disconnect: Cloud sessions continue after disconnect; Remote Control and local sessions need the machine on.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-10

Plan

Cloud: Pro, Max, Ultra, Team, Enterprise; self-hosted: Team, Enterprise (public beta)

Surface

Claude Code CLI; Claude Code web/mobile

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Region and exact idle-expiry durations were not found.

1. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-11-07

2. Configure cloud environments

   Finding: Anthropic-hosted sessions use fresh Ubuntu 24.04 x86_64 VMs with approximate ceilings of 4 vCPUs, 16 GB RAM and 30 GB disk. Setup under roughly five minutes is cached as a filesystem snapshot, expiring after roughly seven days. Foreground Bash defaults to two minutes, up to ten; timed-out commands move to background except sleep-prefixed commands. Unattended background commands have a separate time limit. Idle VMs pause and can be reclaimed. Network secrets are available on Pro, Max, Ultra, Team and Enterprise, with role/configuration requirements.

   Locator

   Installed tools; Time limits; Setup scripts; Environment caching; Network access

   Retrieved

   2026-11-07

Workspace isolation: Git worktree

Typed value: Yes.

Isolation options: git worktrees per session, an OS-level sandbox for shell commands (macOS/Linux/WSL2, off by default), dev containers or the sandbox runtime for the whole process, and a fresh VM per cloud session. The CLI by default runs with the user permissions and file tools outside the sandbox.

• Worktree: claude --worktree, subagent isolation: worktree, and background sessions moving into worktrees before editing.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Current first-party documentation read 2026-11-07; changelog through v2.1.319

• No escape or boundary test was run.

1. Run parallel sessions with worktrees

   Finding: The --worktree or -w flag starts a session in a separate Git working directory, by default .claude/worktrees/<name>, on a new worktree-<name> branch. Subagents can use isolation: worktree. Worktrees share repository history and Git metadata; this documents file separation, not a security sandbox. Source check only; no session was run.

   Locator

   Start Claude in a worktree; Isolate subagents with worktrees; What worktrees share with the main checkout

   Retrieved

   2026-11-07

Workspace isolation: Container

Typed value: Yes.

Isolation options: git worktrees per session, an OS-level sandbox for shell commands (macOS/Linux/WSL2, off by default), dev containers or the sandbox runtime for the whole process, and a fresh VM per cloud session. The CLI by default runs with the user permissions and file tools outside the sandbox.

• Container: Dev containers and Docker are documented isolation options but are user-configured.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Current first-party documentation read 2026-11-07; changelog through v2.1.319

• No escape or boundary test was run.

1. Choose a sandbox environment

   Finding: Compares built-in sandbox, sandbox runtime, dev containers, containers and VMs for isolating Claude Code.

   Locator

   Sandbox runtime; dev containers; VMs

   Retrieved

   2026-11-07

Workspace isolation: VM or microVM

Typed value: Yes.

Isolation options: git worktrees per session, an OS-level sandbox for shell commands (macOS/Linux/WSL2, off by default), dev containers or the sandbox runtime for the whole process, and a fresh VM per cloud session. The CLI by default runs with the user permissions and file tools outside the sandbox.

• Vm: Each Anthropic-hosted cloud session gets a fresh VM.
• Network: Sandboxed commands reach hosts only through a domain-allowlist proxy; cloud environments have network access levels.
• Secrets: Sandbox inherits environment secrets by default; credential protection settings exist; cloud network secrets are available on Pro, Max, Ultra, Team and Enterprise, subject to the documented role and configuration requirements.
• Code separation: Self-hosted runners lock to a single session owner, running only that owner’s sessions until the runner exits. Cloud sessions separately receive fresh VMs; no boundary test ran.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Current first-party documentation read 2026-11-07; changelog through v2.1.319

• No escape or boundary test was run.

1. Configure cloud environments

   Finding: Anthropic-hosted sessions use fresh Ubuntu 24.04 x86_64 VMs with approximate ceilings of 4 vCPUs, 16 GB RAM and 30 GB disk. Setup under roughly five minutes is cached as a filesystem snapshot, expiring after roughly seven days. Foreground Bash defaults to two minutes, up to ten; timed-out commands move to background except sleep-prefixed commands. Unattended background commands have a separate time limit. Idle VMs pause and can be reclaimed. Network secrets are available on Pro, Max, Ultra, Team and Enterprise, with role/configuration requirements.

   Locator

   Installed tools; Time limits; Setup scripts; Environment caching; Network access

   Retrieved

   2026-11-07

Parallel runs

Typed value: Yes.

Parallel work is supported locally (multiple terminals, worktrees, subagents, agent view background sessions, Remote Control server mode with default capacity 32) and in the cloud; every parallel session draws on the same subscription limits.

• Concurrent sessions: Several sessions can run at once in separate terminals or worktrees, as background sessions in agent view, or as cloud sessions; Remote Control server mode defaults to 32 concurrent sessions.
• Concurrent tasks: Background subagents and background Bash commands run alongside the main conversation; agent view dispatches independent sessions.
• Quotas: Parallel sessions share one set of subscription limits; the docs say ten agents use quota roughly ten times as fast; agent view is local and stops if the machine shuts down.
• Status visibility: claude agents shows each session as working, needing input or completed; /tasks lists background work in a session; /workflows lists workflow runs.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Agent view is research preview; no concurrency stress test was run, so observed_concurrency is unknown.

1. Remote Control

   Finding: Drive a local CLI/Desktop/VS Code session from claude.ai/code or the mobile app; machine must stay on; Pro/Max/Team/Enterprise only (no API keys, Bedrock/Vertex/Foundry, gateways); server mode default capacity 32; transcript stored on Anthropic servers while connected.

   Locator

   Requirements; Start a Remote Control session; Limitations; Connection and security

   Retrieved

   2026-10-09

2. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-10-09

3. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-10-09

Coordinator

Typed value: Yes.

Delegation is inspectable through subagents (own model, effort, tools), agent teams, cross-session messaging and dynamic workflows. All workers are Claude sessions; agent teams are experimental and off by default.

• Roles: Lead and teammates in agent teams; custom subagents with their own system prompt and tools.
• Model harness effort: Subagent frontmatter sets model and effort; the harness is always Claude Code.
• Messages: Teammates message each other directly; cross-session messaging passes text between sessions.
• Results return: Subagents return a summary to the spawning conversation; agent-view sessions report only to the user.
• Cross harness: Docs state every approach runs Claude sessions; other tools are only MCP servers.
• Orchestration owner: Claude Code itself (team runtime, subagent runtime, workflow runtime) owns orchestration.
• Activation: Agent teams are disabled by default and need CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1; subagents and fork mode are on in interactive sessions; workflows need enabling on Pro.
• Live progress: Agent view rows, /tasks and /workflows show run state. A background subagent result arrives later as a completion notification, and Claude reports the subagent as still running if asked about progress first.
• Followup: A human can open and prompt any teammate; stopped subagents can be resumed.
• Invocation: Claude spawns subagents with the Agent tool, users @-mention named agents, or start a session with --agent.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Agent teams are experimental; in-process teammates are not restored on resume.

1. Agent teams

   Finding: Experimental, disabled by default (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1). Lead plus independent teammates with own context, direct messaging, shared task list with pending/in-progress/completed states and task dependencies, file-locked claiming; limitations include no in-process teammate resume, one team per session, status lag.

   Locator

   Enablement; shared task list; Assign and claim tasks; Limitations

   Retrieved

   2026-11-07

2. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

3. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-11-07

4. Create custom subagents

   Finding: Subagents have their own context, tools, permissions, model, effort, maxTurns and optional worktree isolation; background subagents run concurrently; results return to the spawning conversation.

   Locator

   Built-in subagents; Supported frontmatter fields; Choose a model; Run subagents in foreground or background

   Retrieved

   2026-11-07

5. Message your other Claude Code sessions

   Finding: Claude can list and message other sessions on the same or another machine or in the cloud; v2.1.224+ on macOS/Linux, v2.1.234+ on Windows; messages are text only.

   Locator

   When to use; Availability

   Retrieved

   2026-11-07

6. Dynamic workflows

   Finding: A JavaScript script Claude writes orchestrates many subagents in the background; on all paid plans, API, Bedrock, Agent Platform and Foundry; /workflows lists runs.

   Locator

   When to use a workflow

   Retrieved

   2026-11-07

Native issue tracking

Typed value: Yes.

Claude Code has persistent native task checklists with optional cross-session lists and experimental team assignment/dependencies. A full issue backlog is not established.

• Create: TaskCreate adds a native checklist task when Task tools are enabled; this is not a full issue backlog.
• Edit: TaskUpdate changes task details and status, dependencies or deletes a task.
• Assign: Experimental team leads can assign tasks and teammates can self-claim unblocked tasks.
• Task session separation: Tasks live in ~/.claude/tasks/<list>/ separately from session transcripts and background process lists; named lists can be shared across sessions.
• Metadata: Task records contain details, status, dependencies and ownership. Labels and priority fields were not established.
• Durable states: Team tasks have pending, in-progress and completed states, and task-list directories remain on disk after sessions end.
• External tracker requirement: Native checklist tracking uses product Task tools and disk storage rather than an external tracker. Complete issue/backlog management remains unestablished.

Checked

2026-11-13 · checked 2 days ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Task tools documented 2026-11-07; model-dependent or opt-in since v2.1.268; team assignment experimental

• Task tools are absent by default on newer interactive models unless enabled; this is a checklist rather than a full issue tracker. Team assignment requires experimental teams.

1. Claude Code tools reference

   Finding: Product Task tools create, read, update and delete tasks. They are model-dependent or opt-in since v2.1.268, and provided across models in cloud/background sessions.

   Locator

   Task tool availability; TaskCreate, TaskGet, TaskList, TaskUpdate

   Retrieved

   2026-11-07

2. Claude Code interactive mode

   Finding: Task checklists are distinct from running shells/subagents, retain state through compaction and resume, and can share a named on-disk task directory across sessions.

   Locator

   Task list

   Retrieved

   2026-11-07

3. Claude Code agent teams

   Finding: Experimental teams use pending, in-progress and completed tasks, dependencies, explicit assignment and file-locked self-claims. Task directories remain after sessions end.

   Locator

   Assign and claim tasks; Architecture; Permissions

   Retrieved

   2026-11-07

Unattended triggers

Typed value: Yes.

Cloud scheduled tasks, formerly routines, support schedule/API/GitHub triggers independently of the laptop. Desktop local schedules and session /loop have separate uptime and expiry rules.

• Schedules: Cloud scheduled tasks are shared across web/Desktop/CLI for one account; /schedule creates them and /routines remains an alias. Local Desktop tasks require app and host awake; /loop is session-scoped.
• Events: Routines trigger on GitHub events and API calls; channels push webhooks and chat messages into an open session.
• Recurring tasks: Routines and Desktop tasks recur indefinitely; /loop tasks expire 7 days after creation.
• Completion conditions: /goal keeps a session working until a model judges the condition met; routines are prompt runs with no documented completion condition.
• Scheduler location: Anthropic cloud (or a self-hosted environment) for routines; the user machine for Desktop tasks and /loop.
• Client independence: Cloud routines run with the laptop closed; /loop and Desktop tasks need Claude Code or the app running.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Scheduled tasks: Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Routines are research preview with experimental API; no scheduled run was executed.

1. Automate work with routines

   Finding: Cloud routines are now named cloud scheduled tasks; shared web/Desktop/CLI account records, schedule/API/GitHub triggers, research preview fire endpoint, and per-account execution. /schedule creates them and /routines remains an alias.

   Locator

   Triggers; Add an API trigger; GitHub event; Usage and limits

   Retrieved

   2026-11-07

2. Run prompts on a schedule

   Finding: /loop and cron tools are session-scoped (1 minute minimum, recurring tasks expire after 7 days, only fire while Claude Code is running and idle, no catch-up for missed fires); cloud routines minimum 1 hour; Desktop tasks need machine on.

   Locator

   Compare scheduling options; Seven-day expiry; Limitations

   Retrieved

   2026-11-07

3. Schedule recurring tasks in Claude Code Desktop

   Finding: Local Desktop scheduled tasks run on the user machine and fire only while the app is open and the computer is awake; remote routines run in the cloud.

   Locator

   Overview; Compare scheduling options

   Retrieved

   2026-11-07

Team roles

Typed value: Yes.

Roles and access are managed at the Claude account level: Owners and admins configure cloud environments and settings; Team adds SSO and central billing; Enterprise adds role-based access, SCIM, audit logs and spend limits.

• Organization roles: Owners configure shared cloud environments and defaults; admins govern settings; Enterprise lists role-based access with fine-grained permissioning.
• Membership: Admins invite members; Console logins require an admin invitation.
• Sso: SSO is listed for Team and Enterprise; Claude apps gateway sign-in uses corporate SSO.
• Provisioning: SCIM is listed for Enterprise and configured at the Claude account level; seat assignment is separate.
• Plan scope: Pricing page separates Team (SSO, central billing) and Enterprise (RBAC, SCIM, audit logs, spend limits, IP allowlisting).

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Project-level and task-level roles were not found; role definitions were not read in detail.

1. Set up Claude Code for your organization

   Finding: Managed settings delivered via claude.ai admin console, MDM or file; SSO, SCIM and seat assignment at the Claude account level; Owners govern cloud environments and the default environment; org effort limits enforced server-side.

   Locator

   API providers; managed settings; usage monitoring; cloud sessions admin

   Retrieved

   2026-10-09

2. Claude pricing

   Finding: Free $0 (Claude Code: No); Pro $17/month annual or $20 monthly (Claude Code: Yes); Max from $100/month (5x or 20x Pro usage); Team standard seat $20 annual/$25 monthly, premium seat $100 annual/$125 monthly (Claude Code included); Enterprise US$20/seat/month plus usage at API rates; usage limits apply; prices exclude tax; model training opt-out on consumer plans, none by default on Team/Enterprise.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-10-09

SSO

Typed value: Yes.

Roles and access are managed at the Claude account level: Owners and admins configure cloud environments and settings; Team adds SSO and central billing; Enterprise adds role-based access, SCIM, audit logs and spend limits.

• Organization roles: Owners configure shared cloud environments and defaults; admins govern settings; Enterprise lists role-based access with fine-grained permissioning.
• Membership: Admins invite members; Console logins require an admin invitation.
• Sso: SSO is listed for Team and Enterprise; Claude apps gateway sign-in uses corporate SSO.
• Provisioning: SCIM is listed for Enterprise and configured at the Claude account level; seat assignment is separate.
• Plan scope: Pricing page separates Team (SSO, central billing) and Enterprise (RBAC, SCIM, audit logs, spend limits, IP allowlisting).

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Project-level and task-level roles were not found; role definitions were not read in detail.

1. Authentication

   Finding: Account types: Claude Pro/Max, Teams/Enterprise, Claude Console, Bedrock/Vertex(Agent Platform)/Foundry, Claude apps gateway. Multiple accounts via separate CLAUDE_CONFIG_DIR directories. ANTHROPIC_API_KEY approval skips login.

   Locator

   Log in to Claude Code; Log in with multiple accounts

   Retrieved

   2026-10-09

2. Claude pricing

   Finding: Free $0 (Claude Code: No); Pro $17/month annual or $20 monthly (Claude Code: Yes); Max from $100/month (5x or 20x Pro usage); Team standard seat $20 annual/$25 monthly, premium seat $100 annual/$125 monthly (Claude Code included); Enterprise US$20/seat/month plus usage at API rates; usage limits apply; prices exclude tax; model training opt-out on consumer plans, none by default on Team/Enterprise.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-10-09

Pull request workflow

Typed value: Yes.

The CLI supports git workflows: it can create commits and pull requests on request, run in isolated worktrees, link sessions to PRs, and (cloud) auto-fix CI failures and review comments on GitHub.

• Commits: Claude creates commits when asked.
• Pull requests: Claude creates pull requests when asked.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• GitHub is the supported PR host for cloud features.

1. Common workflows

   Finding: Claude can create commits and pull requests on request; --from-pr finds the session linked to a PR; images can be pasted or referenced by path; Claude reads files fresh on each tool call.

   Locator

   Create pull requests; Work with images

   Retrieved

   2026-10-09

Review interface

Typed value: Yes.

Outputs include diffs (/diff, Desktop diff view), live Artifact pages published to claude.ai, Desktop app preview, and session links to PRs. An artifact gallery lists created artifacts. Retention of test outputs and recordings was not established.

• Diffs: /diff reviews working-tree changes; Desktop and IDE show diffs.
• Previews: Desktop previews the running app in a Browser pane.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Artifacts: Pro, Max, Team, Enterprise

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Artifact availability requires a claude.ai login and plan; no output retention test was run.

1. Desktop application

   Finding: Desktop Code tab has parallel sessions with Git isolation, diff review and comments, app preview, integrated terminal, permission modes, scheduled tasks and Dispatch from phone; needs a paid subscription; Desktop does not accept a Console API key directly.

   Locator

   Sessions; Choose a permission mode; Work in parallel; Dispatch; Coming from the CLI

   Retrieved

   2026-10-09

2. JetBrains IDEs

   Finding: Plugin runs the separately installed claude CLI in the IDE terminal and adds diff viewing, selection and diagnostic sharing; it does not bundle the CLI.

   Locator

   Supported IDEs; Installation; Features

   Retrieved

   2026-10-09

3. Commands

   Finding: Lists /usage, /export, /import (Codex, Gemini CLI, Cursor), /rewind, /branch, /fork, /goal, /loop, /schedule, /teleport, /desktop, /code-review, /security-review, /install-github-app, /web-setup, /tasks, /sandbox, /diff.

   Locator

   All commands

   Retrieved

   2026-10-09

Product license: Apache-2.0

Typed value: No. Proprietary.

Claude Code is proprietary: the public repository LICENSE.md states "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The sandbox runtime (Apache-2.0) and the Python Agent SDK (MIT, per GitHub) are open source components.

• Client license: LICENSE.md: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The GitHub API reports no license for the repository.
• Source visibility: A public GitHub repository exists for issues, changelog, plugins and examples; the CLI source was not inspected.
• Source revision: Latest release is v2.1.295 (2026-10-08); its tag resolves to commit 602df92 (release page). v2.1.294 is the prior release. The stored head commit 71cdddec623889d38af14b7a489670a03186f659 is not shown on the release page and was not re-verified on 2026-10-09; no source file was inspected.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No source-inspected evidence recorded.
• Head commit 71cdddec623889d38af14b7a489670a03186f659 was not re-verified on 2026-10-09; the repository main page did not expose it.

1. anthropics/claude-code repository

   Finding: Releases page marks v2.1.295 (08 Oct 19:48) as Latest, above stored v2.1.294. Issues tab shows 13,889 open (stored 14,591). Main page shows 149.8k stars; no license name or head commit visible.

   Locator

   Repository metadata, LICENSE.md, releases and issues

   Retrieved

   2026-10-09

2. anthropics/claude-code LICENSE.md (main branch)

   Finding: The whole file reads: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." No open-source license grant appears in the file.

   Locator

   Full file text (single line)

   Retrieved

   2026-10-09

Product license: MIT

Typed value: No. Proprietary.

Claude Code is proprietary: the public repository LICENSE.md states "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The sandbox runtime (Apache-2.0) and the Python Agent SDK (MIT, per GitHub) are open source components.

• Client license: LICENSE.md: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The GitHub API reports no license for the repository.
• Source visibility: A public GitHub repository exists for issues, changelog, plugins and examples; the CLI source was not inspected.
• Source revision: Latest release is v2.1.295 (2026-10-08); its tag resolves to commit 602df92 (release page). v2.1.294 is the prior release. The stored head commit 71cdddec623889d38af14b7a489670a03186f659 is not shown on the release page and was not re-verified on 2026-10-09; no source file was inspected.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No source-inspected evidence recorded.
• Head commit 71cdddec623889d38af14b7a489670a03186f659 was not re-verified on 2026-10-09; the repository main page did not expose it.

1. anthropics/claude-code repository

   Finding: Releases page marks v2.1.295 (08 Oct 19:48) as Latest, above stored v2.1.294. Issues tab shows 13,889 open (stored 14,591). Main page shows 149.8k stars; no license name or head commit visible.

   Locator

   Repository metadata, LICENSE.md, releases and issues

   Retrieved

   2026-10-09

2. anthropics/claude-code LICENSE.md (main branch)

   Finding: The whole file reads: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." No open-source license grant appears in the file.

   Locator

   Full file text (single line)

   Retrieved

   2026-10-09

Product license: GPL-3.0

Typed value: No. Proprietary.

Claude Code is proprietary: the public repository LICENSE.md states "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The sandbox runtime (Apache-2.0) and the Python Agent SDK (MIT, per GitHub) are open source components.

• Client license: LICENSE.md: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The GitHub API reports no license for the repository.
• Source visibility: A public GitHub repository exists for issues, changelog, plugins and examples; the CLI source was not inspected.
• Source revision: Latest release is v2.1.295 (2026-10-08); its tag resolves to commit 602df92 (release page). v2.1.294 is the prior release. The stored head commit 71cdddec623889d38af14b7a489670a03186f659 is not shown on the release page and was not re-verified on 2026-10-09; no source file was inspected.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No source-inspected evidence recorded.
• Head commit 71cdddec623889d38af14b7a489670a03186f659 was not re-verified on 2026-10-09; the repository main page did not expose it.

1. anthropics/claude-code repository

   Finding: Releases page marks v2.1.295 (08 Oct 19:48) as Latest, above stored v2.1.294. Issues tab shows 13,889 open (stored 14,591). Main page shows 149.8k stars; no license name or head commit visible.

   Locator

   Repository metadata, LICENSE.md, releases and issues

   Retrieved

   2026-10-09

2. anthropics/claude-code LICENSE.md (main branch)

   Finding: The whole file reads: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." No open-source license grant appears in the file.

   Locator

   Full file text (single line)

   Retrieved

   2026-10-09

Product license: GPL-3.0-or-later

Typed value: No. Proprietary.

Claude Code is proprietary: the public repository LICENSE.md states "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The sandbox runtime (Apache-2.0) and the Python Agent SDK (MIT, per GitHub) are open source components.

• Client license: LICENSE.md: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The GitHub API reports no license for the repository.
• Source visibility: A public GitHub repository exists for issues, changelog, plugins and examples; the CLI source was not inspected.
• Source revision: Latest release is v2.1.295 (2026-10-08); its tag resolves to commit 602df92 (release page). v2.1.294 is the prior release. The stored head commit 71cdddec623889d38af14b7a489670a03186f659 is not shown on the release page and was not re-verified on 2026-10-09; no source file was inspected.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No source-inspected evidence recorded.
• Head commit 71cdddec623889d38af14b7a489670a03186f659 was not re-verified on 2026-10-09; the repository main page did not expose it.

1. anthropics/claude-code repository

   Finding: Releases page marks v2.1.295 (08 Oct 19:48) as Latest, above stored v2.1.294. Issues tab shows 13,889 open (stored 14,591). Main page shows 149.8k stars; no license name or head commit visible.

   Locator

   Repository metadata, LICENSE.md, releases and issues

   Retrieved

   2026-10-09

2. anthropics/claude-code LICENSE.md (main branch)

   Finding: The whole file reads: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." No open-source license grant appears in the file.

   Locator

   Full file text (single line)

   Retrieved

   2026-10-09

Product license: AGPL-3.0

Typed value: No. Proprietary.

Claude Code is proprietary: the public repository LICENSE.md states "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The sandbox runtime (Apache-2.0) and the Python Agent SDK (MIT, per GitHub) are open source components.

• Client license: LICENSE.md: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The GitHub API reports no license for the repository.
• Source visibility: A public GitHub repository exists for issues, changelog, plugins and examples; the CLI source was not inspected.
• Source revision: Latest release is v2.1.295 (2026-10-08); its tag resolves to commit 602df92 (release page). v2.1.294 is the prior release. The stored head commit 71cdddec623889d38af14b7a489670a03186f659 is not shown on the release page and was not re-verified on 2026-10-09; no source file was inspected.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No source-inspected evidence recorded.
• Head commit 71cdddec623889d38af14b7a489670a03186f659 was not re-verified on 2026-10-09; the repository main page did not expose it.

1. anthropics/claude-code repository

   Finding: Releases page marks v2.1.295 (08 Oct 19:48) as Latest, above stored v2.1.294. Issues tab shows 13,889 open (stored 14,591). Main page shows 149.8k stars; no license name or head commit visible.

   Locator

   Repository metadata, LICENSE.md, releases and issues

   Retrieved

   2026-10-09

2. anthropics/claude-code LICENSE.md (main branch)

   Finding: The whole file reads: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." No open-source license grant appears in the file.

   Locator

   Full file text (single line)

   Retrieved

   2026-10-09

Product license: AGPL-3.0-or-later

Typed value: No. Proprietary.

Claude Code is proprietary: the public repository LICENSE.md states "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The sandbox runtime (Apache-2.0) and the Python Agent SDK (MIT, per GitHub) are open source components.

• Client license: LICENSE.md: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The GitHub API reports no license for the repository.
• Source visibility: A public GitHub repository exists for issues, changelog, plugins and examples; the CLI source was not inspected.
• Source revision: Latest release is v2.1.295 (2026-10-08); its tag resolves to commit 602df92 (release page). v2.1.294 is the prior release. The stored head commit 71cdddec623889d38af14b7a489670a03186f659 is not shown on the release page and was not re-verified on 2026-10-09; no source file was inspected.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No source-inspected evidence recorded.
• Head commit 71cdddec623889d38af14b7a489670a03186f659 was not re-verified on 2026-10-09; the repository main page did not expose it.

1. anthropics/claude-code repository

   Finding: Releases page marks v2.1.295 (08 Oct 19:48) as Latest, above stored v2.1.294. Issues tab shows 13,889 open (stored 14,591). Main page shows 149.8k stars; no license name or head commit visible.

   Locator

   Repository metadata, LICENSE.md, releases and issues

   Retrieved

   2026-10-09

2. anthropics/claude-code LICENSE.md (main branch)

   Finding: The whole file reads: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." No open-source license grant appears in the file.

   Locator

   Full file text (single line)

   Retrieved

   2026-10-09

Product license: Elastic License 2.0

Typed value: No. Proprietary.

Claude Code is proprietary: the public repository LICENSE.md states "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The sandbox runtime (Apache-2.0) and the Python Agent SDK (MIT, per GitHub) are open source components.

• Client license: LICENSE.md: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The GitHub API reports no license for the repository.
• Source visibility: A public GitHub repository exists for issues, changelog, plugins and examples; the CLI source was not inspected.
• Source revision: Latest release is v2.1.295 (2026-10-08); its tag resolves to commit 602df92 (release page). v2.1.294 is the prior release. The stored head commit 71cdddec623889d38af14b7a489670a03186f659 is not shown on the release page and was not re-verified on 2026-10-09; no source file was inspected.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No source-inspected evidence recorded.
• Head commit 71cdddec623889d38af14b7a489670a03186f659 was not re-verified on 2026-10-09; the repository main page did not expose it.

1. anthropics/claude-code repository

   Finding: Releases page marks v2.1.295 (08 Oct 19:48) as Latest, above stored v2.1.294. Issues tab shows 13,889 open (stored 14,591). Main page shows 149.8k stars; no license name or head commit visible.

   Locator

   Repository metadata, LICENSE.md, releases and issues

   Retrieved

   2026-10-09

2. anthropics/claude-code LICENSE.md (main branch)

   Finding: The whole file reads: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." No open-source license grant appears in the file.

   Locator

   Full file text (single line)

   Retrieved

   2026-10-09

Product license: Custom terms

Typed value: No. Proprietary.

Claude Code is proprietary: the public repository LICENSE.md states "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The sandbox runtime (Apache-2.0) and the Python Agent SDK (MIT, per GitHub) are open source components.

• Client license: LICENSE.md: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The GitHub API reports no license for the repository.
• Source visibility: A public GitHub repository exists for issues, changelog, plugins and examples; the CLI source was not inspected.
• Source revision: Latest release is v2.1.295 (2026-10-08); its tag resolves to commit 602df92 (release page). v2.1.294 is the prior release. The stored head commit 71cdddec623889d38af14b7a489670a03186f659 is not shown on the release page and was not re-verified on 2026-10-09; no source file was inspected.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No source-inspected evidence recorded.
• Head commit 71cdddec623889d38af14b7a489670a03186f659 was not re-verified on 2026-10-09; the repository main page did not expose it.

1. anthropics/claude-code repository

   Finding: Releases page marks v2.1.295 (08 Oct 19:48) as Latest, above stored v2.1.294. Issues tab shows 13,889 open (stored 14,591). Main page shows 149.8k stars; no license name or head commit visible.

   Locator

   Repository metadata, LICENSE.md, releases and issues

   Retrieved

   2026-10-09

2. anthropics/claude-code LICENSE.md (main branch)

   Finding: The whole file reads: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." No open-source license grant appears in the file.

   Locator

   Full file text (single line)

   Retrieved

   2026-10-09

Product license: Proprietary

Typed value: Yes. Proprietary.

Claude Code is proprietary: the public repository LICENSE.md states "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The sandbox runtime (Apache-2.0) and the Python Agent SDK (MIT, per GitHub) are open source components.

• Client license: LICENSE.md: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The GitHub API reports no license for the repository.
• Source visibility: A public GitHub repository exists for issues, changelog, plugins and examples; the CLI source was not inspected.
• Source revision: Latest release is v2.1.295 (2026-10-08); its tag resolves to commit 602df92 (release page). v2.1.294 is the prior release. The stored head commit 71cdddec623889d38af14b7a489670a03186f659 is not shown on the release page and was not re-verified on 2026-10-09; no source file was inspected.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No source-inspected evidence recorded.
• Head commit 71cdddec623889d38af14b7a489670a03186f659 was not re-verified on 2026-10-09; the repository main page did not expose it.

1. anthropics/claude-code repository

   Finding: Releases page marks v2.1.295 (08 Oct 19:48) as Latest, above stored v2.1.294. Issues tab shows 13,889 open (stored 14,591). Main page shows 149.8k stars; no license name or head commit visible.

   Locator

   Repository metadata, LICENSE.md, releases and issues

   Retrieved

   2026-10-09

2. anthropics/claude-code LICENSE.md (main branch)

   Finding: The whole file reads: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." No open-source license grant appears in the file.

   Locator

   Full file text (single line)

   Retrieved

   2026-10-09

Self-hosting

Typed value: Yes.

The installed Claude Code CLI can run its full agent loop, tools and context management on user-controlled infrastructure, including scripts and CI/CD. Model inference/authentication still uses Anthropic or a supported provider. This does not self-host the Claude web/desktop cloud control plane; cloud self-hosted runners are a separate worker feature.

• Self hosting: The CLI runs the full Claude Code agent loop and tools locally through claude -p, including user-managed scripts and CI/CD.

Checked

2026-11-13 · checked 2 days ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code CLI headless documentation checked 2026-11-13 for local-core self-hosting. Other atomic answers retain their individually dated source scope.

• The mapping covers the local CLI core only; hosted cloud/desktop service self-hosting is not established.

1. Run Claude Code programmatically

   Finding: The CLI exposes the same agent loop, tools and context management as Claude Code through claude -p for scripts and CI/CD. The installed local core runs with model-provider credentials; this is more than a remote execution worker for the hosted web product.

   Locator

   Agent SDK introduction; CLI scripts and CI/CD; Basic usage

   Retrieved

   2026-11-13

Telemetry opt-out

Typed value: Yes.

Prompts and outputs go to Anthropic (or the chosen provider) over TLS 1.2+; consumer plans train on data only when the setting is on, commercial plans do not by default; standard retention is 30 days for commercial and 30 days or 5 years for consumers. Usage metrics default on for the Claude API and off on Amazon Bedrock, Google Cloud's Agent Platform and Microsoft Foundry; error reports default on only for Pro, Max and Ultra sign-ins on v2.1.198+; both have documented opt-outs.

• Code data: Local files stay on the machine; code in prompts and tool results is sent to the model API; cloud sessions clone repositories into a VM.
• Prompts: Prompts and outputs are sent over TLS 1.2+ and retained per account type.
• Credentials: For Anthropic-hosted cloud sessions, GitHub credentials stay encrypted on Anthropic servers and never enter the VM. Customer-managed self-hosted runners use customer-managed Git credentials.
• Logs: Local CLI transcripts are plaintext for 30 days by default; transcripts of sessions started or most recently continued in Claude Desktop or Cowork are exempt from that cleanup by default (data-usage page).
• Retention: Consumer: 5 years if training allowed, else 30 days; commercial: 30 days, ZDR for qualified Enterprise.
• Deletion: Cloud sessions can be deleted permanently; local transcripts are swept by cleanupPeriodDays.
• Training: Consumer plans train when the setting is on; commercial plans do not by default.
• Telemetry disclosure: Metrics and error reports are described.
• Telemetry defaults: Metrics default on for the Claude API; error reports on for Pro/Max/Ultra sign-ins on v2.1.198+; both off by default on Bedrock, Foundry, Agent Platform.
• Telemetry controls: DISABLE_TELEMETRY, DISABLE_ERROR_REPORTING and CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC opt out.
• Consent prompt: First-run telemetry prompts were not observed. The documented session-survey follow-up asks before any transcript upload; that is a separate consent step, not a telemetry prompt.
• Provider separation: Third-party providers default telemetry off, except that hosts setting CLAUDE_CODE_PROVIDER_MANAGED_BY_HOST default metrics on.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Retention and training differ by plan

Surface

Claude Code CLI

Version

Current first-party documentation read 2026-11-07; changelog through v2.1.319

• Plan-specific; Remote Control and cloud flows differ.

1. Data usage

   Finding: Current data policy states opt-in consumer training and five-year versus thirty-day retention by choice; commercial no training by default and thirty-day standard retention. Local plaintext CLI transcripts default to thirty days, with Desktop/Cowork exemptions. Metrics and error reports have separate opt-outs. Error reports default on for direct Claude API Pro, Max and Ultra sign-ins on v2.1.198+, otherwise off. Provider-managed-host metrics are an explicit exception to third-party-provider default-off behavior.

   Locator

   Data training policy; Data retention; Telemetry services; Default behaviors by API provider

   Retrieved

   2026-11-07

Free access

Typed value: No.

Free excludes Claude Code; all paid plans include it. Evaluation requires an eligible subscription, Console/API billing or supported provider route; bundled platform credits do not pay Code usage.

• Ongoing free: Pricing page lists Claude Code as No for the Free plan.
• Trial: Trial terms are not established. The pricing comparison marks Trials No for Team and self-serve Enterprise and Yes only for sales-assisted Enterprise (pricing page).
• Paid only: All paid Claude plans include Code; Free does not.
• Usage caps: Usage limits apply to every plan.
• Paid evaluation: Pro is $20 monthly or $17 on annual billing.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Free plan excluded

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Console API credit promotions and trials were not checked.
• Console API credit promotions were not checked; trial length and terms for sales-assisted Enterprise are not stated.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Starting price: Free

Typed value: No. Up to $20/month.

Free excludes Claude Code; all paid plans include it. Evaluation requires an eligible subscription, Console/API billing or supported provider route; bundled platform credits do not pay Code usage.

• Ongoing free: Pricing page lists Claude Code as No for the Free plan.
• Trial: Trial terms are not established. The pricing comparison marks Trials No for Team and self-serve Enterprise and Yes only for sales-assisted Enterprise (pricing page).
• Paid only: All paid Claude plans include Code; Free does not.
• Usage caps: Usage limits apply to every plan.
• Paid evaluation: Pro is $20 monthly or $17 on annual billing.

Checked

2026-11-13 · checked 2 days ago

Plan

Free plan excluded

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Console API credit promotions and trials were not checked.
• Console API credit promotions were not checked; trial length and terms for sales-assisted Enterprise are not stated.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Starting price: Up to $20/month

Typed value: Yes. Up to $20/month.

Free excludes Claude Code; all paid plans include it. Evaluation requires an eligible subscription, Console/API billing or supported provider route; bundled platform credits do not pay Code usage.

• Ongoing free: Pricing page lists Claude Code as No for the Free plan.
• Trial: Trial terms are not established. The pricing comparison marks Trials No for Team and self-serve Enterprise and Yes only for sales-assisted Enterprise (pricing page).
• Paid only: All paid Claude plans include Code; Free does not.
• Usage caps: Usage limits apply to every plan.
• Paid evaluation: Pro is $20 monthly or $17 on annual billing.

Checked

2026-11-13 · checked 2 days ago

Plan

Free plan excluded

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Console API credit promotions and trials were not checked.
• Console API credit promotions were not checked; trial length and terms for sales-assisted Enterprise are not stated.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Starting price: $21–40/month

Typed value: No. Up to $20/month.

Free excludes Claude Code; all paid plans include it. Evaluation requires an eligible subscription, Console/API billing or supported provider route; bundled platform credits do not pay Code usage.

• Ongoing free: Pricing page lists Claude Code as No for the Free plan.
• Trial: Trial terms are not established. The pricing comparison marks Trials No for Team and self-serve Enterprise and Yes only for sales-assisted Enterprise (pricing page).
• Paid only: All paid Claude plans include Code; Free does not.
• Usage caps: Usage limits apply to every plan.
• Paid evaluation: Pro is $20 monthly or $17 on annual billing.

Checked

2026-11-13 · checked 2 days ago

Plan

Free plan excluded

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Console API credit promotions and trials were not checked.
• Console API credit promotions were not checked; trial length and terms for sales-assisted Enterprise are not stated.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Starting price: $41–100/month

Typed value: No. Up to $20/month.

Free excludes Claude Code; all paid plans include it. Evaluation requires an eligible subscription, Console/API billing or supported provider route; bundled platform credits do not pay Code usage.

• Ongoing free: Pricing page lists Claude Code as No for the Free plan.
• Trial: Trial terms are not established. The pricing comparison marks Trials No for Team and self-serve Enterprise and Yes only for sales-assisted Enterprise (pricing page).
• Paid only: All paid Claude plans include Code; Free does not.
• Usage caps: Usage limits apply to every plan.
• Paid evaluation: Pro is $20 monthly or $17 on annual billing.

Checked

2026-11-13 · checked 2 days ago

Plan

Free plan excluded

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Console API credit promotions and trials were not checked.
• Console API credit promotions were not checked; trial length and terms for sales-assisted Enterprise are not stated.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Starting price: Over $100/month

Typed value: No. Up to $20/month.

Free excludes Claude Code; all paid plans include it. Evaluation requires an eligible subscription, Console/API billing or supported provider route; bundled platform credits do not pay Code usage.

• Ongoing free: Pricing page lists Claude Code as No for the Free plan.
• Trial: Trial terms are not established. The pricing comparison marks Trials No for Team and self-serve Enterprise and Yes only for sales-assisted Enterprise (pricing page).
• Paid only: All paid Claude plans include Code; Free does not.
• Usage caps: Usage limits apply to every plan.
• Paid evaluation: Pro is $20 monthly or $17 on annual billing.

Checked

2026-11-13 · checked 2 days ago

Plan

Free plan excluded

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Console API credit promotions and trials were not checked.
• Console API credit promotions were not checked; trial length and terms for sales-assisted Enterprise are not stated.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Starting price: Contact for pricing

Typed value: No. Up to $20/month.

Free excludes Claude Code; all paid plans include it. Evaluation requires an eligible subscription, Console/API billing or supported provider route; bundled platform credits do not pay Code usage.

• Ongoing free: Pricing page lists Claude Code as No for the Free plan.
• Trial: Trial terms are not established. The pricing comparison marks Trials No for Team and self-serve Enterprise and Yes only for sales-assisted Enterprise (pricing page).
• Paid only: All paid Claude plans include Code; Free does not.
• Usage caps: Usage limits apply to every plan.
• Paid evaluation: Pro is $20 monthly or $17 on annual billing.

Checked

2026-11-13 · checked 2 days ago

Plan

Free plan excluded

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Console API credit promotions and trials were not checked.
• Console API credit promotions were not checked; trial length and terms for sales-assisted Enterprise are not stated.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Pricing model: Free, no paid plan

Typed value: No.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-13 · checked 2 days ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Pricing model: Free plan

Typed value: No.

Free excludes Claude Code; all paid plans include it. Evaluation requires an eligible subscription, Console/API billing or supported provider route; bundled platform credits do not pay Code usage.

• Ongoing free: Pricing page lists Claude Code as No for the Free plan.
• Trial: Trial terms are not established. The pricing comparison marks Trials No for Team and self-serve Enterprise and Yes only for sales-assisted Enterprise (pricing page).
• Paid only: All paid Claude plans include Code; Free does not.
• Usage caps: Usage limits apply to every plan.
• Paid evaluation: Pro is $20 monthly or $17 on annual billing.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-10

Plan

Free plan excluded

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Console API credit promotions and trials were not checked.
• Console API credit promotions were not checked; trial length and terms for sales-assisted Enterprise are not stated.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Pricing model: Subscription

Typed value: Yes.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-10

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Pricing model: Per seat

Typed value: Yes.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-10

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Pricing model: Usage-based

Typed value: Yes.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Credits: Max/Ultra monthly platform API credits cannot be used for Claude Code.
• Overages: Paid usage credits or Console API credits are separate billing routes.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-10

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Manage costs effectively

   Finding: /usage shows client-side session cost estimate at list price, plan usage bars and attribution to skills/subagents/plugins/MCP; usage credits past plan limits; --max-budget-usd and --max-turns in print mode; org spend limits for Team/Enterprise; average enterprise cost about $13 per developer per active day.

   Locator

   Track your costs; Plan usage breakdown; Usage credits; Manage costs for your organization

   Retrieved

   2026-11-07

2. Use Claude Code with your Pro or Max plan

   Finding: Pro, Max and Ultra include Code in a shared subscription quota; Max and Ultra platform API credits cannot pay Code usage. ANTHROPIC_API_KEY causes API billing instead of subscription. No workload usage test.

   Locator

   Connect; usage limits; Claude Code billing

   Retrieved

   2026-11-07

Pricing model: Per machine or compute

Typed value: No.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Compute cost: Cloud sessions have no separate compute charge under the documented subscription route.

Checked

2026-11-13 · checked 2 days ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-11-07

Pricing model: Bring your own key

Typed value: Yes.

Pro, Max, Ultra, Team and Enterprise include Claude Code. Console/API and third-party cloud-provider routes are separate. Free excludes Code; platform API credits bundled with Max/Ultra cannot pay Code usage.

• Api accounts: Anthropic Console API key or Console login, plus Bedrock, Google Cloud Agent Platform, Microsoft Foundry and Claude Platform on AWS credentials.
• Account switching: Separate CLAUDE_CONFIG_DIR directories keep settings, session history and claude.ai logins separate, so several accounts can stay signed in; /login and /logout switch accounts. Two Console sign-ins without API keys are not isolated by config directory because their credentials are stored outside it (authentication page).

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-10

Plan

Pro, Max, Ultra, Team, Enterprise, Console API, third-party providers

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Local-model statement is the vendor support position; unofficial base-URL workarounds were not tested.
• No trial or promotional pricing beyond the public pricing page was checked.
• Two Console sign-ins without API keys are not isolated by config directory (authentication page); the account_switching atom is partial for that case.

1. Authentication

   Finding: Account types: Claude Pro/Max, Teams/Enterprise, Claude Console, Bedrock/Vertex(Agent Platform)/Foundry, Claude apps gateway. Multiple accounts via separate CLAUDE_CONFIG_DIR directories. ANTHROPIC_API_KEY approval skips login.

   Locator

   Log in to Claude Code; Log in with multiple accounts

   Retrieved

   2026-11-07

Pricing model: One-time purchase

Typed value: No.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-13 · checked 2 days ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Paid plans from: No paid plan

Typed value: No. $11–20/month.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Paid plans from: $10 or less/month

Typed value: No. $11–20/month.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Paid plans from: $11–20/month

Typed value: Yes. $11–20/month.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Paid plans from: $21–40/month

Typed value: No. $11–20/month.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Paid plans from: $41–100/month

Typed value: No. $11–20/month.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Paid plans from: Over $100/month

Typed value: No. $11–20/month.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Paid plans from: Contact sales

Typed value: No. $11–20/month.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Keeps going with laptop closed

Typed value: Yes.

Sessions persist as local transcripts and resume; cloud sessions restore history after VM reclamation but lose background processes; Remote Control reconnects after sleep, and server mode gives up after roughly 10 minutes without network; background local sessions survive sleep but not shutdown.

• Client disconnect: Cloud sessions keep running. Remote Control reconnects after a brief network loss; in server mode Claude Code gives up after roughly 10 minutes and the claude remote-control process exits. Local sessions need the machine.
• Expiry: Cloud VMs are reclaimed after inactivity; conversation history is restored on reopen but running background work is not.
• Process reconnect: claude attach re-enters a running background session; Remote Control reconnects.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-10

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No crash or restart experiments were run; claims are from documentation.

1. Remote Control

   Finding: Drive a local CLI/Desktop/VS Code session from claude.ai/code or the mobile app; machine must stay on; Pro/Max/Team/Enterprise only (no API keys, Bedrock/Vertex/Foundry, gateways); server mode default capacity 32; transcript stored on Anthropic servers while connected.

   Locator

   Requirements; Start a Remote Control session; Limitations; Connection and security

   Retrieved

   2026-10-09

2. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-10-09

Open-source license

Typed value: No. Proprietary.

Claude Code is proprietary: the public repository LICENSE.md states "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The sandbox runtime (Apache-2.0) and the Python Agent SDK (MIT, per GitHub) are open source components.

• Client license: LICENSE.md: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." The GitHub API reports no license for the repository.
• Source visibility: A public GitHub repository exists for issues, changelog, plugins and examples; the CLI source was not inspected.
• Source revision: Latest release is v2.1.295 (2026-10-08); its tag resolves to commit 602df92 (release page). v2.1.294 is the prior release. The stored head commit 71cdddec623889d38af14b7a489670a03186f659 is not shown on the release page and was not re-verified on 2026-10-09; no source file was inspected.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No source-inspected evidence recorded.
• Head commit 71cdddec623889d38af14b7a489670a03186f659 was not re-verified on 2026-10-09; the repository main page did not expose it.

1. anthropics/claude-code repository

   Finding: Releases page marks v2.1.295 (08 Oct 19:48) as Latest, above stored v2.1.294. Issues tab shows 13,889 open (stored 14,591). Main page shows 149.8k stars; no license name or head commit visible.

   Locator

   Repository metadata, LICENSE.md, releases and issues

   Retrieved

   2026-10-09

2. anthropics/claude-code LICENSE.md (main branch)

   Finding: The whole file reads: "© Anthropic PBC. All rights reserved. Use is subject to Anthropic's Commercial Terms of Service." No open-source license grant appears in the file.

   Locator

   Full file text (single line)

   Retrieved

   2026-10-09

Any desktop app

Typed value: Yes.

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

• Desktop os: CLI supports macOS, Windows and Linux; Desktop supports macOS/Windows with Linux beta on Ubuntu/Debian. Direct CLI --desktop launch and managed 3P Desktop are separate auth paths.
• Cli: CLI: macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+, x64 or ARM64, 4 GB+ RAM; the built-in sandbox works on macOS, Linux and WSL2 only.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Advanced setup

   Finding: CLI runs on macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+; 4 GB+ RAM; x64 or ARM64; internet required; Bash, Zsh, PowerShell or CMD.

   Locator

   System requirements

   Retrieved

   2026-11-07

2. Desktop application

   Finding: Desktop Code tab has parallel sessions with Git isolation, diff review and comments, app preview, integrated terminal, permission modes, scheduled tasks and Dispatch from phone; needs a paid subscription; Desktop does not accept a Console API key directly.

   Locator

   Sessions; Choose a permission mode; Work in parallel; Dispatch; Coming from the CLI

   Retrieved

   2026-11-07

3. Claude Desktop on Linux (beta)

   Finding: Linux desktop app is beta on Ubuntu 22.04+/Debian 12+; Fedora/Arch should use the CLI; signs in with claude.ai subscription or SSO, not a Console API key.

   Locator

   Requirements; What is not in the Linux beta yet

   Retrieved

   2026-11-07

4. Feature availability

   Finding: Local CLI features work across supported providers. Desktop is an exception: gateway/3P deployments can run Code on Bedrock, Google Agent Platform, Foundry, self-hosted gateways and the Claude API with Console key. Subscription-only features have separate plan restrictions.

   Locator

   Availability by model provider; Features that require a Claude subscription

   Retrieved

   2026-11-07

Paid plan up to $20/month

Typed value: Yes. $11–20/month.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Paid plan up to $40/month

Typed value: Yes. $11–20/month.

Pro $20/month or $200/year; Max $100/month; Ultra $200/month. Team seats $25/$125 monthly or $20/$100 monthly-equivalent annual. Enterprise $20/seat/month annual with 20-seat minimum, plus API usage.

• Currency: US dollar public prices; taxes excluded.
• Interval: Pro monthly or annual; Max/Ultra monthly; Team monthly/annual; Enterprise annual.
• Seats: Team Standard/Premium seats are mixable, for 2-150 users. Enterprise minimum is 20 seats.
• Seat minimum: Team 2 users; Enterprise 20 seats.
• Included usage: Paid plans include Code with shared usage limits; task/token allowance varies by workload.
• Taxes: Applicable taxes excluded.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• No workload, billing or authenticated client test was performed; unestablished details remain unknown.

1. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

Any phone client

Typed value: Yes.

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

• Mobile apps: Claude iOS and Android apps (iPad uses the iOS app) have a Code tab for cloud sessions, projects, Remote Control and Dispatch; code never runs on the phone.
• Task actions: Mobile reaches cloud sessions, Remote Control local sessions, or Desktop Dispatch. Dispatch is on Pro/Max/Ultra; the host must stay awake for local execution.
• Notifications: Push notifications to the phone when Remote Control is active (task finished or decision needed), Dispatch notifications, terminal bell/desktop notification settings in the CLI, and a Notification hook.
• Cross device continuity: Cloud sessions persist across devices; Remote Control continues a local CLI/Desktop/VS Code session from phone or browser while the machine stays on; /teleport pulls a cloud session into the terminal; /desktop continues a CLI session in Desktop. A terminal session cannot be pushed to the cloud (handoff is one-way); a desktop-to-phone move was not tested.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Claude Code on mobile

   Finding: The Claude iOS/Android app is a client for cloud sessions, projects, Remote Control and Dispatch (no code runs on the phone). Push notifications when Remote Control is active; local-only commands such as /plugin and /resume do not work; Bypass permissions cannot be selected from the app.

   Locator

   Work from your phone; Get push notifications; Limitations

   Retrieved

   2026-11-07

claude-code:ACC-01

Claude Code has CLI, Desktop, IDE, browser and mobile clients. Execution is local or cloud by surface; mobile is a client only. Desktop Linux is beta; 3P Desktop has separately documented provider and Console-key routes.

Checked

2026-11-07 · checked 1 week ago

Plan

Pro, Max, Ultra, Team and Enterprise. Desktop, web and mobile need a claude.ai subscription, except the Desktop exception: gateway routing can be configured in the app or by an administrator; Enterprise deployments can route Desktop to Google Cloud Agent Platform or a gateway via managed settings; and Claude Desktop on 3P runs the Code tab on Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry or a self-hosted LLM gateway (feature-availability).

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Capabilities and authentication differ across CLI, Desktop, IDE, mobile and cloud.
• No installed-client or authenticated scenario was run.
• Desktop 3P supports the documented Console-key route; mobile/cloud still require eligible claude.ai accounts.

1. Claude Code overview

   Finding: Documents terminal (macOS/Linux/WSL/Windows), VS Code, desktop (macOS, Windows x64/ARM64, Ubuntu/Debian beta), web and JetBrains surfaces; most surfaces need a Claude subscription or Console account; CLI, VS Code and JetBrains also support third-party providers.

   Locator

   Get started tabs; Terminal install; Desktop app; Web

   Retrieved

   2026-11-07

2. Platforms and integrations

   Finding: Compares CLI, Desktop, VS Code, JetBrains, Web and Mobile; CLI is the most complete surface (scripting and Agent SDK are CLI-only); Dispatch, Remote Control, Channels, Slack, self-hosted environments and scheduled tasks listed with where Claude runs and plan requirements.

   Locator

   Where to run Claude Code; Work when you are away from your terminal

   Retrieved

   2026-11-07

3. Advanced setup

   Finding: CLI runs on macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+; 4 GB+ RAM; x64 or ARM64; internet required; Bash, Zsh, PowerShell or CMD.

   Locator

   System requirements

   Retrieved

   2026-11-07

4. Claude Code on mobile

   Finding: The Claude iOS/Android app is a client for cloud sessions, projects, Remote Control and Dispatch (no code runs on the phone). Push notifications when Remote Control is active; local-only commands such as /plugin and /resume do not work; Bypass permissions cannot be selected from the app.

   Locator

   Work from your phone; Get push notifications; Limitations

   Retrieved

   2026-11-07

5. Remote Control

   Finding: Drive a local CLI/Desktop/VS Code session from claude.ai/code or the mobile app; machine must stay on; Pro/Max/Team/Enterprise only (no API keys, Bedrock/Vertex/Foundry, gateways); server mode default capacity 32; transcript stored on Anthropic servers while connected.

   Locator

   Requirements; Start a Remote Control session; Limitations; Connection and security

   Retrieved

   2026-11-07

6. Desktop application

   Finding: Desktop Code tab has parallel sessions with Git isolation, diff review and comments, app preview, integrated terminal, permission modes, scheduled tasks and Dispatch from phone; needs a paid subscription; Desktop does not accept a Console API key directly.

   Locator

   Sessions; Choose a permission mode; Work in parallel; Dispatch; Coming from the CLI

   Retrieved

   2026-11-07

7. Claude Desktop on Linux (beta)

   Finding: Linux desktop app is beta on Ubuntu 22.04+/Debian 12+; Fedora/Arch should use the CLI; signs in with claude.ai subscription or SSO, not a Console API key.

   Locator

   Requirements; What is not in the Linux beta yet

   Retrieved

   2026-11-07

8. Use Claude Code in VS Code

   Finding: VS Code 1.94.0+; extension version equals bundled Claude Code version; inline diffs, plan review, permission mode indicator; also installable in Cursor; shares ~/.claude settings with the CLI.

   Locator

   Requirements; Permission modes; Git integration

   Retrieved

   2026-11-07

9. JetBrains IDEs

   Finding: Plugin runs the separately installed claude CLI in the IDE terminal and adds diff viewing, selection and diagnostic sharing; it does not bundle the CLI.

   Locator

   Supported IDEs; Installation; Features

   Retrieved

   2026-11-07

10. Use Claude Code in the cloud

    Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

    Locator

    Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

    Retrieved

    2026-11-07

11. Configure the sandboxed Bash tool

    Finding: OS-enforced sandbox for shell commands on macOS, Linux and WSL2 (off by default; native Windows runs unsandboxed); built on open source sandbox-runtime; reads allowed broadly by default, writes limited to working directory, network via domain-allowlist proxy; file tools, MCP servers and hooks run outside it.

    Locator

    What the sandbox restricts; What runs outside the sandbox

    Retrieved

    2026-11-07

12. Automate actions with hooks

    Finding: Hooks fire on events such as PreToolUse, PostToolUse, PermissionRequest, Notification, Stop, SubagentStop, SessionStart/End, TaskCreated/Completed, WorktreeCreate and can be commands, prompts, agents or HTTP calls; can block or auto-approve actions.

    Locator

    What you can automate; Hook types; Hooks and permission modes

    Retrieved

    2026-11-07

13. Feature availability

    Finding: Local CLI features work across supported providers. Desktop is an exception: gateway/3P deployments can run Code on Bedrock, Google Agent Platform, Foundry, self-hosted gateways and the Claude API with Console key. Subscription-only features have separate plan restrictions.

    Locator

    Availability by model provider; Features that require a Claude subscription

    Retrieved

    2026-11-07

14. Commands

    Finding: Lists /usage, /export, /import (Codex, Gemini CLI, Cursor), /rewind, /branch, /fork, /goal, /loop, /schedule, /teleport, /desktop, /code-review, /security-review, /install-github-app, /web-setup, /tasks, /sandbox, /diff.

    Locator

    All commands

    Retrieved

    2026-11-07

15. All settings

    Finding: Settings include agentPushNotifEnabled, inputNeededNotifEnabled, preferredNotifChannel, remoteControlAtStartup, cleanupPeriodDays, autoMemoryEnabled and sandbox keys.

    Locator

    Remote, desktop, and notifications

    Retrieved

    2026-11-07

16. Continue local sessions from any device with Remote Control

    Finding: Remote Control connects claude.ai/code or the mobile apps to a local Claude session. Users can send messages from the terminal, browser and phone; the local host must remain running. Eligible Claude subscriptions are required.

    Locator

    Introduction; Connect from another device; Requirements

    Retrieved

    2026-11-13

claude-code:ACC-02

Claude Code embeds its own harness and runs Claude models only. Providers: Anthropic (subscription or Console), Amazon Bedrock, Google Cloud Agent Platform, Microsoft Foundry, Claude Platform on AWS, and gateways that expose an Anthropic-format API. Anthropic does not support routing to non-Claude models and the docs state that workers in every multi-agent mode are Claude sessions.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Scope is the Claude Code harness; unofficial third-party routing to other model families is not covered by the documentation and was not tested.

1. Authentication

   Finding: Account types: Claude Pro/Max, Teams/Enterprise, Claude Console, Bedrock/Vertex(Agent Platform)/Foundry, Claude apps gateway. Multiple accounts via separate CLAUDE_CONFIG_DIR directories. ANTHROPIC_API_KEY approval skips login.

   Locator

   Log in to Claude Code; Log in with multiple accounts

   Retrieved

   2026-11-07

2. Run Claude Code programmatically

   Finding: claude -p runs non-interactively with exit codes, --output-format json/stream-json, --bare, --allowedTools; Agent SDK for Python/TypeScript.

   Locator

   Basic usage; Bare mode; Structured output

   Retrieved

   2026-11-07

3. Agent SDK overview

   Finding: Python and TypeScript libraries run the Claude Code agent loop with built-in tools, hooks, subagents, MCP, permissions and sessions; non-Python/TS languages can drive the CLI as a subprocess.

   Locator

   Compare the Agent SDK to other Claude tools; Capabilities

   Retrieved

   2026-11-07

4. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

5. Connect Claude Code to tools via MCP

   Finding: MCP client supports http, sse, stdio and websocket servers with OAuth 2.0, local/project/user scopes and claude.ai connectors; claude mcp serve exposes Claude Code as a stdio MCP server; channels push events.

   Locator

   Installing MCP servers; Authenticate with remote MCP servers; Use Claude Code as an MCP server; Scopes

   Retrieved

   2026-11-07

6. Feature availability

   Finding: Local CLI features work across supported providers. Desktop is an exception: gateway/3P deployments can run Code on Bedrock, Google Agent Platform, Foundry, self-hosted gateways and the Claude API with Console key. Subscription-only features have separate plan restrictions.

   Locator

   Availability by model provider; Features that require a Claude subscription

   Retrieved

   2026-11-07

7. Other LLM gateways

   Finding: Any gateway exposing a supported Anthropic API format works; Anthropic does not support routing Claude Code to non-Claude models through any gateway.

   Locator

   What a gateway provides; Roll out a gateway

   Retrieved

   2026-11-07

claude-code:ACC-03

Pro, Max, Ultra, Team and Enterprise include Claude Code. Console/API and third-party cloud-provider routes are separate. Free excludes Code; platform API credits bundled with Max/Ultra cannot pay Code usage.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Pro, Max, Ultra, Team, Enterprise, Console API, third-party providers

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Local-model statement is the vendor support position; unofficial base-URL workarounds were not tested.
• No trial or promotional pricing beyond the public pricing page was checked.
• Two Console sign-ins without API keys are not isolated by config directory (authentication page); the account_switching atom is partial for that case.

1. Claude Code overview

   Finding: Documents terminal (macOS/Linux/WSL/Windows), VS Code, desktop (macOS, Windows x64/ARM64, Ubuntu/Debian beta), web and JetBrains surfaces; most surfaces need a Claude subscription or Console account; CLI, VS Code and JetBrains also support third-party providers.

   Locator

   Get started tabs; Terminal install; Desktop app; Web

   Retrieved

   2026-11-07

2. Authentication

   Finding: Account types: Claude Pro/Max, Teams/Enterprise, Claude Console, Bedrock/Vertex(Agent Platform)/Foundry, Claude apps gateway. Multiple accounts via separate CLAUDE_CONFIG_DIR directories. ANTHROPIC_API_KEY approval skips login.

   Locator

   Log in to Claude Code; Log in with multiple accounts

   Retrieved

   2026-11-07

3. Claude Code on mobile

   Finding: The Claude iOS/Android app is a client for cloud sessions, projects, Remote Control and Dispatch (no code runs on the phone). Push notifications when Remote Control is active; local-only commands such as /plugin and /resume do not work; Bypass permissions cannot be selected from the app.

   Locator

   Work from your phone; Get push notifications; Limitations

   Retrieved

   2026-11-07

4. Remote Control

   Finding: Drive a local CLI/Desktop/VS Code session from claude.ai/code or the mobile app; machine must stay on; Pro/Max/Team/Enterprise only (no API keys, Bedrock/Vertex/Foundry, gateways); server mode default capacity 32; transcript stored on Anthropic servers while connected.

   Locator

   Requirements; Start a Remote Control session; Limitations; Connection and security

   Retrieved

   2026-11-07

5. Push events into a running session with channels

   Finding: Research preview. MCP-based channels push Telegram/Discord/iMessage or webhook events into an open session; requires claude.ai login or Console key; not on Bedrock/Agent Platform/Foundry; Team/Enterprise must enable.

   Locator

   Overview; Enterprise controls

   Retrieved

   2026-11-07

6. Manage costs effectively

   Finding: /usage shows client-side session cost estimate at list price, plan usage bars and attribution to skills/subagents/plugins/MCP; usage credits past plan limits; --max-budget-usd and --max-turns in print mode; org spend limits for Team/Enterprise; average enterprise cost about $13 per developer per active day.

   Locator

   Track your costs; Plan usage breakdown; Usage credits; Manage costs for your organization

   Retrieved

   2026-11-07

7. Feature availability

   Finding: Local CLI features work across supported providers. Desktop is an exception: gateway/3P deployments can run Code on Bedrock, Google Agent Platform, Foundry, self-hosted gateways and the Claude API with Console key. Subscription-only features have separate plan restrictions.

   Locator

   Availability by model provider; Features that require a Claude subscription

   Retrieved

   2026-11-07

8. Other LLM gateways

   Finding: Any gateway exposing a supported Anthropic API format works; Anthropic does not support routing Claude Code to non-Claude models through any gateway.

   Locator

   What a gateway provides; Roll out a gateway

   Retrieved

   2026-11-07

9. Claude pricing

   Finding: Current list prices: Pro $20 monthly or $200/year displayed as $17/month; Max $100 monthly; Ultra $200 monthly. Team 2-150 users: Standard $25 monthly/$20 annual per month, Premium $125/$100. Enterprise $20/seat/month annual, 20-seat minimum, plus API usage. All paid plans include Code; taxes excluded. Max/Ultra platform API credits cannot pay Code usage.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-11-07

10. Use Claude Code with your Pro or Max plan

    Finding: Pro, Max and Ultra include Code in a shared subscription quota; Max and Ultra platform API credits cannot pay Code usage. ANTHROPIC_API_KEY causes API billing instead of subscription. No workload usage test.

    Locator

    Connect; usage limits; Claude Code billing

    Retrieved

    2026-11-07

claude-code:ACC-04

CLI onboarding is one install command plus browser login, and works in any directory. Worktree isolation needs git; cloud sessions need GitHub (or a local bundle). Self-hosted runners exist for Team/Enterprise. Remote Control is off by default on Team and Enterprise plans until an Owner enables it.

Checked

2026-11-07 · checked 1 week ago

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Current first-party documentation read 2026-11-07; changelog through v2.1.319

• Onboarding was read from documentation, not timed hands-on.
• Self-hosted environments are public beta and exclude ZDR and HIPAA organizations.

1. Claude Code overview

   Finding: Documents terminal (macOS/Linux/WSL/Windows), VS Code, desktop (macOS, Windows x64/ARM64, Ubuntu/Debian beta), web and JetBrains surfaces; most surfaces need a Claude subscription or Console account; CLI, VS Code and JetBrains also support third-party providers.

   Locator

   Get started tabs; Terminal install; Desktop app; Web

   Retrieved

   2026-10-09

2. Advanced setup

   Finding: CLI runs on macOS 13+, Windows 10 1809+/Server 2019+, Ubuntu 20.04+, Debian 10+, Alpine 3.19+; 4 GB+ RAM; x64 or ARM64; internet required; Bash, Zsh, PowerShell or CMD.

   Locator

   System requirements

   Retrieved

   2026-10-09

3. Authentication

   Finding: Account types: Claude Pro/Max, Teams/Enterprise, Claude Console, Bedrock/Vertex(Agent Platform)/Foundry, Claude apps gateway. Multiple accounts via separate CLAUDE_CONFIG_DIR directories. ANTHROPIC_API_KEY approval skips login.

   Locator

   Log in to Claude Code; Log in with multiple accounts

   Retrieved

   2026-10-09

4. Remote Control

   Finding: Drive a local CLI/Desktop/VS Code session from claude.ai/code or the mobile app; machine must stay on; Pro/Max/Team/Enterprise only (no API keys, Bedrock/Vertex/Foundry, gateways); server mode default capacity 32; transcript stored on Anthropic servers while connected.

   Locator

   Requirements; Start a Remote Control session; Limitations; Connection and security

   Retrieved

   2026-10-09

5. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-10-09

6. Get started with Claude Code in the cloud

   Finding: Browser onboarding connects GitHub and sets up a Default cloud environment.

   Locator

   Connect GitHub; first session

   Retrieved

   2026-10-09

7. Self-hosted environments

   Finding: Public beta on Team and Enterprise (off by default); runners on customer infrastructure execute cloud sessions; unavailable with ZDR or HIPAA; model inference still via Anthropic API unless Bedrock/Agent Platform configured; session content goes to Anthropic.

   Locator

   How self-hosted environments work; Availability and limitations

   Retrieved

   2026-10-09

8. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-10-09

9. Run parallel sessions with worktrees

   Finding: claude --worktree creates .claude/worktrees/<name> on a new branch; requires a git repository (hooks can replace git logic); subagents can use isolation: worktree.

   Locator

   Start Claude in a worktree; Isolate subagents; Non-git version control

   Retrieved

   2026-10-09

10. Claude Code with GitHub Enterprise Server

    Finding: Self-hosted GitHub Enterprise Server instances supported for Team and Enterprise plans.

    Locator

    Overview

    Retrieved

    2026-10-09

11. What is new

    Finding: Weekly digests list cross-session messaging and self-hosted environments (Week 32, public beta), auto mode default (from August 14 for Pro/Max/Team), Desktop resume of CLI sessions (Week 35), Fable 5.1 (Week 36); no future-dated commitments found.

    Locator

    Weeks 32 to 37 (2026)

    Retrieved

    2026-10-09

12. Configure cloud environments

    Finding: Anthropic-hosted sessions use fresh Ubuntu 24.04 x86_64 VMs with approximate ceilings of 4 vCPUs, 16 GB RAM and 30 GB disk. Setup under roughly five minutes is cached as a filesystem snapshot, expiring after roughly seven days. Foreground Bash defaults to two minutes, up to ten; timed-out commands move to background except sleep-prefixed commands. Unattended background commands have a separate time limit. Idle VMs pause and can be reclaimed. Network secrets are available on Pro, Max, Ultra, Team and Enterprise, with role/configuration requirements.

    Locator

    Installed tools; Time limits; Setup scripts; Environment caching; Network access

    Retrieved

    2026-11-07

claude-code:ACC-05

Programmable access is strong on the CLI: claude -p with JSON output, Python/TypeScript Agent SDK, claude mcp serve and HTTP/webhook triggers for routines. The routine /fire endpoint is experimental and several surfaces are research preview.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• API/webhook coverage is limited to routines and the SDK; no hands-on integration test.

1. Authentication

   Finding: Account types: Claude Pro/Max, Teams/Enterprise, Claude Console, Bedrock/Vertex(Agent Platform)/Foundry, Claude apps gateway. Multiple accounts via separate CLAUDE_CONFIG_DIR directories. ANTHROPIC_API_KEY approval skips login.

   Locator

   Log in to Claude Code; Log in with multiple accounts

   Retrieved

   2026-11-07

2. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-11-07

3. Automate work with routines

   Finding: Cloud routines are now named cloud scheduled tasks; shared web/Desktop/CLI account records, schedule/API/GitHub triggers, research preview fire endpoint, and per-account execution. /schedule creates them and /routines remains an alias.

   Locator

   Triggers; Add an API trigger; GitHub event; Usage and limits

   Retrieved

   2026-11-07

4. Run Claude Code programmatically

   Finding: claude -p runs non-interactively with exit codes, --output-format json/stream-json, --bare, --allowedTools; Agent SDK for Python/TypeScript.

   Locator

   Basic usage; Bare mode; Structured output

   Retrieved

   2026-11-07

5. Agent SDK overview

   Finding: Python and TypeScript libraries run the Claude Code agent loop with built-in tools, hooks, subagents, MCP, permissions and sessions; non-Python/TS languages can drive the CLI as a subprocess.

   Locator

   Compare the Agent SDK to other Claude tools; Capabilities

   Retrieved

   2026-11-07

6. claude-agent-sdk-python repository

   Finding: Public repository for the Python Agent SDK; GitHub reports an MIT license (retrieved via the GitHub repository API on 2026-10-08).

   Locator

   GitHub repository metadata (license field)

   Retrieved

   2026-11-07

7. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-11-07

8. Automate actions with hooks

   Finding: Hooks fire on events such as PreToolUse, PostToolUse, PermissionRequest, Notification, Stop, SubagentStop, SessionStart/End, TaskCreated/Completed, WorktreeCreate and can be commands, prompts, agents or HTTP calls; can block or auto-approve actions.

   Locator

   What you can automate; Hook types; Hooks and permission modes

   Retrieved

   2026-11-07

9. Push events into a running session with channels

   Finding: Research preview. MCP-based channels push Telegram/Discord/iMessage or webhook events into an open session; requires claude.ai login or Console key; not on Bedrock/Agent Platform/Foundry; Team/Enterprise must enable.

   Locator

   Overview; Enterprise controls

   Retrieved

   2026-11-07

10. Connect Claude Code to tools via MCP

    Finding: MCP client supports http, sse, stdio and websocket servers with OAuth 2.0, local/project/user scopes and claude.ai connectors; claude mcp serve exposes Claude Code as a stdio MCP server; channels push events.

    Locator

    Installing MCP servers; Authenticate with remote MCP servers; Use Claude Code as an MCP server; Scopes

    Retrieved

    2026-11-07

11. CLI reference

    Finding: Documents claude agents, claude remote-control, claude mcp, claude gateway; flags --worktree, --max-turns and --max-budget-usd (print mode only), --fallback-model, --output-format, --permission-mode, --bare, --channels.

    Locator

    CLI commands; CLI flags

    Retrieved

    2026-11-07

claude-code:ACC-06

Claude Code is an MCP client (http, sse, stdio, websocket) with OAuth and scoped config, supports plugins/marketplaces, skills, hooks, custom subagents, reads AGENTS.md, and can import Codex, Gemini CLI and Cursor configuration.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Integration ownership (which connectors Anthropic maintains) was not established.
• Direct AGENTS.md reading requires Claude Code v2.1.277 or later, and in some sessions Claude cannot read AGENTS.md; the docs suggest importing it from CLAUDE.md there.
• v2.1.295 negotiates MCP protocol revision 2026-07-28 by default; MCP_PROTOCOL_NEGOTIATION=legacy opts out (changelog, 2026-10-08). The stored answers for this claim do not address protocol versions.

1. Agent SDK overview

   Finding: Python and TypeScript libraries run the Claude Code agent loop with built-in tools, hooks, subagents, MCP, permissions and sessions; non-Python/TS languages can drive the CLI as a subprocess.

   Locator

   Compare the Agent SDK to other Claude tools; Capabilities

   Retrieved

   2026-10-09

2. Automate actions with hooks

   Finding: Hooks fire on events such as PreToolUse, PostToolUse, PermissionRequest, Notification, Stop, SubagentStop, SessionStart/End, TaskCreated/Completed, WorktreeCreate and can be commands, prompts, agents or HTTP calls; can block or auto-approve actions.

   Locator

   What you can automate; Hook types; Hooks and permission modes

   Retrieved

   2026-10-09

3. How Claude remembers your project

   Finding: CLAUDE.md at managed/user/project/local scope with imports and .claude/rules; AGENTS.md read when no CLAUDE.md (v2.1.277+); auto memory stored per project under ~/.claude/projects/<project>/memory/ as editable markdown; /memory browses and toggles.

   Locator

   CLAUDE.md files; AGENTS.md; Auto memory; View and edit with /memory

   Retrieved

   2026-10-09

4. Extend Claude with skills

   Finding: Skills follow the Agent Skills open standard; custom commands merged into skills; bundled skills include /loop, /batch, /code-review.

   Locator

   Bundled skills

   Retrieved

   2026-10-09

5. Plugins overview

   Finding: Plugins package skills, agents, hooks and MCP servers, installed from marketplaces via /plugin.

   Locator

   Plugin contents; marketplaces

   Retrieved

   2026-10-09

6. Connect Claude Code to tools via MCP

   Finding: MCP client supports http, sse, stdio and websocket servers with OAuth 2.0, local/project/user scopes and claude.ai connectors; claude mcp serve exposes Claude Code as a stdio MCP server; channels push events.

   Locator

   Installing MCP servers; Authenticate with remote MCP servers; Use Claude Code as an MCP server; Scopes

   Retrieved

   2026-10-09

7. Set up Claude Code for your organization

   Finding: Managed settings delivered via claude.ai admin console, MDM or file; SSO, SCIM and seat assignment at the Claude account level; Owners govern cloud environments and the default environment; org effort limits enforced server-side.

   Locator

   API providers; managed settings; usage monitoring; cloud sessions admin

   Retrieved

   2026-10-09

8. Commands

   Finding: Lists /usage, /export, /import (Codex, Gemini CLI, Cursor), /rewind, /branch, /fork, /goal, /loop, /schedule, /teleport, /desktop, /code-review, /security-review, /install-github-app, /web-setup, /tasks, /sandbox, /diff.

   Locator

   All commands

   Retrieved

   2026-10-09

9. CLI reference

   Finding: Documents claude agents, claude remote-control, claude mcp, claude gateway; flags --worktree, --max-turns and --max-budget-usd (print mode only), --fallback-model, --output-format, --permission-mode, --bare, --channels.

   Locator

   CLI commands; CLI flags

   Retrieved

   2026-10-09

claude-code:ACC-07

The mobile app is a client for cloud or Remote Control sessions, and IDE extensions wrap or bundle the CLI engine. Desktop framework and inspected-source evidence were not established.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Desktop UI framework and CLI implementation language were not established from first-party sources; the public repository holds issues, plugins and the changelog under an all-rights-reserved license, and no source inspection was done.

1. Platforms and integrations

   Finding: Compares CLI, Desktop, VS Code, JetBrains, Web and Mobile; CLI is the most complete surface (scripting and Agent SDK are CLI-only); Dispatch, Remote Control, Channels, Slack, self-hosted environments and scheduled tasks listed with where Claude runs and plan requirements.

   Locator

   Where to run Claude Code; Work when you are away from your terminal

   Retrieved

   2026-10-09

2. Claude Code on mobile

   Finding: The Claude iOS/Android app is a client for cloud sessions, projects, Remote Control and Dispatch (no code runs on the phone). Push notifications when Remote Control is active; local-only commands such as /plugin and /resume do not work; Bypass permissions cannot be selected from the app.

   Locator

   Work from your phone; Get push notifications; Limitations

   Retrieved

   2026-10-09

3. Desktop application

   Finding: Desktop Code tab has parallel sessions with Git isolation, diff review and comments, app preview, integrated terminal, permission modes, scheduled tasks and Dispatch from phone; needs a paid subscription; Desktop does not accept a Console API key directly.

   Locator

   Sessions; Choose a permission mode; Work in parallel; Dispatch; Coming from the CLI

   Retrieved

   2026-10-09

4. Use Claude Code in VS Code

   Finding: VS Code 1.94.0+; extension version equals bundled Claude Code version; inline diffs, plan review, permission mode indicator; also installable in Cursor; shares ~/.claude settings with the CLI.

   Locator

   Requirements; Permission modes; Git integration

   Retrieved

   2026-10-09

5. JetBrains IDEs

   Finding: Plugin runs the separately installed claude CLI in the IDE terminal and adds diff viewing, selection and diagnostic sharing; it does not bundle the CLI.

   Locator

   Supported IDEs; Installation; Features

   Retrieved

   2026-10-09

claude-code:PLAN-01

No native issue tracker was found in the documentation, changelog or repository checked; work arrives by prompt, Slack/Claude Tag mentions, GitHub @claude mentions and GitHub-event routines, and external trackers such as Linear or Notion are reached through MCP connectors. Authoritative-tracker, ID and sync semantics depend on the connector and were not established.

Checked

2026-11-07 · checked 1 week ago

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Current first-party documentation read 2026-11-07; changelog through v2.1.319

• A native issue/backlog tracker was not established; session Task tools are a separate documented checklist; the pages checked do not describe tracker identity or sync.

1. Claude Code overview

   Finding: Documents terminal (macOS/Linux/WSL/Windows), VS Code, desktop (macOS, Windows x64/ARM64, Ubuntu/Debian beta), web and JetBrains surfaces; most surfaces need a Claude subscription or Console account; CLI, VS Code and JetBrains also support third-party providers.

   Locator

   Get started tabs; Terminal install; Desktop app; Web

   Retrieved

   2026-10-09

2. Platforms and integrations

   Finding: Compares CLI, Desktop, VS Code, JetBrains, Web and Mobile; CLI is the most complete surface (scripting and Agent SDK are CLI-only); Dispatch, Remote Control, Channels, Slack, self-hosted environments and scheduled tasks listed with where Claude runs and plan requirements.

   Locator

   Where to run Claude Code; Work when you are away from your terminal

   Retrieved

   2026-10-09

3. Automate work with routines

   Finding: Research preview. Cloud routines with schedule, API (/fire endpoint under experimental beta header, claude.ai users only) and GitHub event triggers; Pro/Max/Team/Enterprise; run on Anthropic cloud or self-hosted environment; belong to individual account; hourly/webhook caps apply.

   Locator

   Triggers; Add an API trigger; GitHub event; Usage and limits

   Retrieved

   2026-10-09

4. Run Claude Code programmatically

   Finding: claude -p runs non-interactively with exit codes, --output-format json/stream-json, --bare, --allowedTools; Agent SDK for Python/TypeScript.

   Locator

   Basic usage; Bare mode; Structured output

   Retrieved

   2026-10-09

5. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-10-09

6. Connect Claude Code to tools via MCP

   Finding: MCP client supports http, sse, stdio and websocket servers with OAuth 2.0, local/project/user scopes and claude.ai connectors; claude mcp serve exposes Claude Code as a stdio MCP server; channels push events.

   Locator

   Installing MCP servers; Authenticate with remote MCP servers; Use Claude Code as an MCP server; Scopes

   Retrieved

   2026-10-09

7. Claude Code GitHub Actions

   Finding: claude-code-action runs Claude Code in GitHub workflows from @claude mentions or any GitHub event; turns issues into PRs.

   Locator

   Setup; mentions

   Retrieved

   2026-10-09

8. Claude Code in Slack

   Finding: The earlier Slack bot is available on Pro, Max and Ultra only in workspaces not connected to an organization Claude Tag. Team/Enterprise use Claude Tag. The retirement notice effective October 5 applies in an organization-connected workspace using the earlier bot.

   Locator

   Setup; plans

   Retrieved

   2026-11-07

9. Claude Code tools reference

   Finding: Product Task tools create, read, update and delete tasks. They are model-dependent or opt-in since v2.1.268, and provided across models in cloud/background sessions.

   Locator

   Task tool availability; TaskCreate, TaskGet, TaskList, TaskUpdate

   Retrieved

   2026-11-07

10. Claude Code interactive mode

    Finding: Task checklists are distinct from running shells/subagents, retain state through compaction and resume, and can share a named on-disk task directory across sessions.

    Locator

    Task list

    Retrieved

    2026-11-07

claude-code:PLAN-02

Plan mode proposes a plan before edits and plans can be saved into the repository; /goal tracks a completion condition. Durable plan objects with revisions and acceptance-criteria editors were not found.

Checked

2026-10-09 · checked 1 month ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Plan files are written to ~/.claude/plans or a project-relative plansDirectory; plan revision history, amendment tracking and links from plans to implementation were not established.

1. Agent teams

   Finding: Experimental, disabled by default (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1). Lead plus independent teammates with own context, direct messaging, shared task list with pending/in-progress/completed states and task dependencies, file-locked claiming; limitations include no in-process teammate resume, one team per session, status lag.

   Locator

   Enablement; shared task list; Assign and claim tasks; Limitations

   Retrieved

   2026-10-09

2. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-10-09

3. Run prompts on a schedule

   Finding: /loop and cron tools are session-scoped (1 minute minimum, recurring tasks expire after 7 days, only fire while Claude Code is running and idle, no catch-up for missed fires); cloud routines minimum 1 hour; Desktop tasks need machine on.

   Locator

   Compare scheduling options; Seven-day expiry; Limitations

   Retrieved

   2026-10-09

4. Keep Claude working toward a goal

   Finding: /goal sets a completion condition that a model checks after each turn; goal clears when met, judged impossible, or on an unfixable error.

   Locator

   Compare ways to keep a session running; How evaluation works

   Retrieved

   2026-10-09

5. Choose a permission mode

   Finding: Modes default/Manual, acceptEdits, plan, auto (classifier), dontAsk, bypassPermissions; deny rules apply in every mode; auto is the built-in starting mode for interactive terminal and VS Code sessions from v2.1.283.

   Locator

   Available modes; Actions no mode auto-approves

   Retrieved

   2026-10-09

6. Manage sessions

   Finding: Sessions saved continuously as local JSONL transcripts; --continue, --resume, --from-pr, /branch; running the same session in two terminals interleaves one transcript; /export writes plain text.

   Locator

   Resume a session; Export and locate session data

   Retrieved

   2026-10-09

7. Claude Code permission modes

   Finding: Ctrl+G opens the proposed plan in the default text editor before approval; approval exits plan mode.

   Locator

   Review and approve a plan

   Retrieved

   2026-11-07

8. Claude Code settings reference

   Finding: plansDirectory sets where plan mode stores plan files, resolved relative to the project root; unset defaults to ~/.claude/plans, and paths outside the project root fall back to ~/.claude/plans.

   Locator

   plansDirectory

   Retrieved

   2026-11-07

claude-code:PLAN-03

Ordinary TaskUpdate changes task dependencies. Task tools are available by default on selected older model families, can be enabled on other models, and are provided on every model in background and cloud sessions. Experimental agent teams add a shared task list where unresolved dependencies block claiming. Hierarchy and epics were not established.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Agent teams are experimental and off by default; no epic/story hierarchy, task-to-epic movement or identity retention was found.

1. Agent teams

   Finding: Experimental, disabled by default (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1). Lead plus independent teammates with own context, direct messaging, shared task list with pending/in-progress/completed states and task dependencies, file-locked claiming; limitations include no in-process teammate resume, one team per session, status lag.

   Locator

   Enablement; shared task list; Assign and claim tasks; Limitations

   Retrieved

   2026-11-07

2. Automate actions with hooks

   Finding: Hooks fire on events such as PreToolUse, PostToolUse, PermissionRequest, Notification, Stop, SubagentStop, SessionStart/End, TaskCreated/Completed, WorktreeCreate and can be commands, prompts, agents or HTTP calls; can block or auto-approve actions.

   Locator

   What you can automate; Hook types; Hooks and permission modes

   Retrieved

   2026-11-07

3. Tools reference

   Finding: Lists built-in tools including TaskCreate/TaskUpdate/TaskList, TaskStop, WebFetch, WebSearch, Workflow; task tools are provided by default only on some model families.

   Locator

   Tool list; Task tool availability; WebFetch and WebSearch

   Retrieved

   2026-11-07

claude-code:PLAN-04

Persistent instructions and memory are documented: CLAUDE.md at managed/user/project/local scope, AGENTS.md reading, skills, imports and rules, and editable per-project auto memory stored locally.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Scope verified from documentation only; no hands-on takeover test.
• Direct AGENTS.md reading requires Claude Code v2.1.277 or later, and in some sessions Claude cannot read AGENTS.md; the docs suggest importing it from CLAUDE.md there.

1. How Claude remembers your project

   Finding: CLAUDE.md at managed/user/project/local scope with imports and .claude/rules; AGENTS.md read when no CLAUDE.md (v2.1.277+); auto memory stored per project under ~/.claude/projects/<project>/memory/ as editable markdown; /memory browses and toggles.

   Locator

   CLAUDE.md files; AGENTS.md; Auto memory; View and edit with /memory

   Retrieved

   2026-10-09

2. Extend Claude with skills

   Finding: Skills follow the Agent Skills open standard; custom commands merged into skills; bundled skills include /loop, /batch, /code-review.

   Locator

   Bundled skills

   Retrieved

   2026-10-09

3. Plugins overview

   Finding: Plugins package skills, agents, hooks and MCP servers, installed from marketplaces via /plugin.

   Locator

   Plugin contents; marketplaces

   Retrieved

   2026-10-09

claude-code:PLAN-05

Code understanding relies on agentic search (grep, glob, file reads) plus optional LSP code-intelligence plugins and user-supplied MCP indexes; images can be attached; web search and fetch tools exist. A built-in semantic index and citation or conflict-detection features were not found.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No retrieval-quality or citation test was run.

1. Platforms and integrations

   Finding: Compares CLI, Desktop, VS Code, JetBrains, Web and Mobile; CLI is the most complete surface (scripting and Agent SDK are CLI-only); Dispatch, Remote Control, Channels, Slack, self-hosted environments and scheduled tasks listed with where Claude runs and plan requirements.

   Locator

   Where to run Claude Code; Work when you are away from your terminal

   Retrieved

   2026-10-09

2. Claude Code on mobile

   Finding: The Claude iOS/Android app is a client for cloud sessions, projects, Remote Control and Dispatch (no code runs on the phone). Push notifications when Remote Control is active; local-only commands such as /plugin and /resume do not work; Bypass permissions cannot be selected from the app.

   Locator

   Work from your phone; Get push notifications; Limitations

   Retrieved

   2026-10-09

3. Desktop application

   Finding: Desktop Code tab has parallel sessions with Git isolation, diff review and comments, app preview, integrated terminal, permission modes, scheduled tasks and Dispatch from phone; needs a paid subscription; Desktop does not accept a Console API key directly.

   Locator

   Sessions; Choose a permission mode; Work in parallel; Dispatch; Coming from the CLI

   Retrieved

   2026-10-09

4. Connect Claude Code to tools via MCP

   Finding: MCP client supports http, sse, stdio and websocket servers with OAuth 2.0, local/project/user scopes and claude.ai connectors; claude mcp serve exposes Claude Code as a stdio MCP server; channels push events.

   Locator

   Installing MCP servers; Authenticate with remote MCP servers; Use Claude Code as an MCP server; Scopes

   Retrieved

   2026-10-09

5. CLI reference

   Finding: Documents claude agents, claude remote-control, claude mcp, claude gateway; flags --worktree, --max-turns and --max-budget-usd (print mode only), --fallback-model, --output-format, --permission-mode, --bare, --channels.

   Locator

   CLI commands; CLI flags

   Retrieved

   2026-10-09

6. Set up Claude Code in a monorepo or large codebase

   Finding: Finding symbols can cost many file reads and grep calls; LSP code intelligence plugins add definition/reference lookup; an organization code search or RAG index can be exposed as an MCP tool.

   Locator

   Code intelligence plugins; Expose your own index over MCP

   Retrieved

   2026-10-09

7. Common workflows

   Finding: Claude can create commits and pull requests on request; --from-pr finds the session linked to a PR; images can be pasted or referenced by path; Claude reads files fresh on each tool call.

   Locator

   Create pull requests; Work with images

   Retrieved

   2026-10-09

claude-code:PLAN-06

Claude Code has persistent native task checklists with optional cross-session lists and experimental team assignment/dependencies. A full issue backlog is not established.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Task tools documented 2026-11-07; model-dependent or opt-in since v2.1.268; team assignment experimental

• Task tools are absent by default on newer interactive models unless enabled; this is a checklist rather than a full issue tracker. Team assignment requires experimental teams.

1. Claude Code tools reference

   Finding: Product Task tools create, read, update and delete tasks. They are model-dependent or opt-in since v2.1.268, and provided across models in cloud/background sessions.

   Locator

   Task tool availability; TaskCreate, TaskGet, TaskList, TaskUpdate

   Retrieved

   2026-11-07

2. Claude Code interactive mode

   Finding: Task checklists are distinct from running shells/subagents, retain state through compaction and resume, and can share a named on-disk task directory across sessions.

   Locator

   Task list

   Retrieved

   2026-11-07

3. Claude Code agent teams

   Finding: Experimental teams use pending, in-progress and completed tasks, dependencies, explicit assignment and file-locked self-claims. Task directories remain after sessions end.

   Locator

   Assign and claim tasks; Architecture; Permissions

   Retrieved

   2026-11-07

claude-code:PLAN-07

TaskCreate, TaskGet, TaskList and TaskUpdate manage session checklists. They are available by default on selected older model families, can be enabled on other models, and are provided on every model in background and cloud sessions. Experimental agent teams add shared claiming and assignment. A native shared issue tracker was not established.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Agent teams are experimental and disabled by default. Local foreground sessions provide task tools by default only on selected model families; other models can opt in. Background and cloud sessions provide them on every model. These are session task lists, not a native issue tracker.

1. Agent teams

   Finding: Experimental, disabled by default (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1). Lead plus independent teammates with own context, direct messaging, shared task list with pending/in-progress/completed states and task dependencies, file-locked claiming; limitations include no in-process teammate resume, one team per session, status lag.

   Locator

   Enablement; shared task list; Assign and claim tasks; Limitations

   Retrieved

   2026-11-07

2. Interactive mode

   Finding: Task list checklist; automatic continue after a claude.ai usage limit resets in interactive subscription sessions; background Bash commands.

   Locator

   Task list; Wait for a usage limit to reset

   Retrieved

   2026-11-07

3. Tools reference

   Finding: Lists built-in tools including TaskCreate/TaskUpdate/TaskList, TaskStop, WebFetch, WebSearch, Workflow; task tools are provided by default only on some model families.

   Locator

   Tool list; Task tool availability; WebFetch and WebSearch

   Retrieved

   2026-11-07

4. Claude Code tools reference

   Finding: Product Task tools create, read, update and delete tasks. They are model-dependent or opt-in since v2.1.268, and provided across models in cloud/background sessions.

   Locator

   Task tool availability; TaskCreate, TaskGet, TaskList, TaskUpdate

   Retrieved

   2026-11-07

5. Claude Code agent teams

   Finding: Experimental teams use pending, in-progress and completed tasks, dependencies, explicit assignment and file-locked self-claims. Task directories remain after sessions end.

   Locator

   Assign and claim tasks; Architecture; Permissions

   Retrieved

   2026-11-07

claude-code:PLAN-08

Living product specifications were not found as a native feature; instructions files and plans in the repository are the nearest documented mechanisms.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• These are general instruction files, not a specification object with approvals or drift detection.

1. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-10-09

2. How Claude remembers your project

   Finding: CLAUDE.md at managed/user/project/local scope with imports and .claude/rules; AGENTS.md read when no CLAUDE.md (v2.1.277+); auto memory stored per project under ~/.claude/projects/<project>/memory/ as editable markdown; /memory browses and toggles.

   Locator

   CLAUDE.md files; AGENTS.md; Auto memory; View and edit with /memory

   Retrieved

   2026-10-09

claude-code:COORD-01

Parallel work is supported locally (multiple terminals, worktrees, subagents, agent view background sessions, Remote Control server mode with default capacity 32) and in the cloud; every parallel session draws on the same subscription limits.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Agent view is research preview; no concurrency stress test was run, so observed_concurrency is unknown.

1. Remote Control

   Finding: Drive a local CLI/Desktop/VS Code session from claude.ai/code or the mobile app; machine must stay on; Pro/Max/Team/Enterprise only (no API keys, Bedrock/Vertex/Foundry, gateways); server mode default capacity 32; transcript stored on Anthropic servers while connected.

   Locator

   Requirements; Start a Remote Control session; Limitations; Connection and security

   Retrieved

   2026-10-09

2. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-10-09

3. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-10-09

4. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-10-09

5. Create custom subagents

   Finding: Subagents have their own context, tools, permissions, model, effort, maxTurns and optional worktree isolation; background subagents run concurrently; results return to the spawning conversation.

   Locator

   Built-in subagents; Supported frontmatter fields; Choose a model; Run subagents in foreground or background

   Retrieved

   2026-10-09

claude-code:COORD-02

Agent teams (experimental) document lead assignment and file-locked task claiming; outside teams, assignment is the user dispatching sessions.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Agent teams are experimental and disabled by default; one team per session.

1. Agent teams

   Finding: Experimental, disabled by default (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1). Lead plus independent teammates with own context, direct messaging, shared task list with pending/in-progress/completed states and task dependencies, file-locked claiming; limitations include no in-process teammate resume, one team per session, status lag.

   Locator

   Enablement; shared task list; Assign and claim tasks; Limitations

   Retrieved

   2026-11-07

2. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-11-07

3. Self-hosted environments

   Finding: Public beta on Team and Enterprise (off by default); runners on customer infrastructure execute cloud sessions; unavailable with ZDR or HIPAA; model inference still via Anthropic API unless Bedrock/Agent Platform configured; session content goes to Anthropic.

   Locator

   How self-hosted environments work; Availability and limitations

   Retrieved

   2026-11-07

4. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

5. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-11-07

6. Claude Code agent teams

   Finding: Experimental teams use pending, in-progress and completed tasks, dependencies, explicit assignment and file-locked self-claims. Task directories remain after sessions end.

   Locator

   Assign and claim tasks; Architecture; Permissions

   Retrieved

   2026-11-07

claude-code:COORD-03

Delegation is inspectable through subagents (own model, effort, tools), agent teams, cross-session messaging and dynamic workflows. All workers are Claude sessions; agent teams are experimental and off by default.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Agent teams are experimental; in-process teammates are not restored on resume.

1. Agent teams

   Finding: Experimental, disabled by default (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1). Lead plus independent teammates with own context, direct messaging, shared task list with pending/in-progress/completed states and task dependencies, file-locked claiming; limitations include no in-process teammate resume, one team per session, status lag.

   Locator

   Enablement; shared task list; Assign and claim tasks; Limitations

   Retrieved

   2026-11-07

2. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

3. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-11-07

4. Create custom subagents

   Finding: Subagents have their own context, tools, permissions, model, effort, maxTurns and optional worktree isolation; background subagents run concurrently; results return to the spawning conversation.

   Locator

   Built-in subagents; Supported frontmatter fields; Choose a model; Run subagents in foreground or background

   Retrieved

   2026-11-07

5. Message your other Claude Code sessions

   Finding: Claude can list and message other sessions on the same or another machine or in the cloud; v2.1.224+ on macOS/Linux, v2.1.234+ on Windows; messages are text only.

   Locator

   When to use; Availability

   Retrieved

   2026-11-07

6. Dynamic workflows

   Finding: A JavaScript script Claude writes orchestrates many subagents in the background; on all paid plans, API, Bedrock, Agent Platform and Foundry; /workflows lists runs.

   Locator

   When to use a workflow

   Retrieved

   2026-11-07

claude-code:COORD-04

Cloud scheduled tasks, formerly routines, support schedule/API/GitHub triggers independently of the laptop. Desktop local schedules and session /loop have separate uptime and expiry rules.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Scheduled tasks: Pro, Max, Ultra, Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Routines are research preview with experimental API; no scheduled run was executed.

1. Automate work with routines

   Finding: Cloud routines are now named cloud scheduled tasks; shared web/Desktop/CLI account records, schedule/API/GitHub triggers, research preview fire endpoint, and per-account execution. /schedule creates them and /routines remains an alias.

   Locator

   Triggers; Add an API trigger; GitHub event; Usage and limits

   Retrieved

   2026-11-07

2. Run prompts on a schedule

   Finding: /loop and cron tools are session-scoped (1 minute minimum, recurring tasks expire after 7 days, only fire while Claude Code is running and idle, no catch-up for missed fires); cloud routines minimum 1 hour; Desktop tasks need machine on.

   Locator

   Compare scheduling options; Seven-day expiry; Limitations

   Retrieved

   2026-11-07

3. Schedule recurring tasks in Claude Code Desktop

   Finding: Local Desktop scheduled tasks run on the user machine and fire only while the app is open and the computer is awake; remote routines run in the cloud.

   Locator

   Overview; Compare scheduling options

   Retrieved

   2026-11-07

4. Keep Claude working toward a goal

   Finding: /goal sets a completion condition that a model checks after each turn; goal clears when met, judged impossible, or on an unfixable error.

   Locator

   Compare ways to keep a session running; How evaluation works

   Retrieved

   2026-11-07

5. Push events into a running session with channels

   Finding: Research preview. MCP-based channels push Telegram/Discord/iMessage or webhook events into an open session; requires claude.ai login or Console key; not on Bedrock/Agent Platform/Foundry; Team/Enterprise must enable.

   Locator

   Overview; Enterprise controls

   Retrieved

   2026-11-07

claude-code:COORD-05

In the CLI a human can interrupt, stop, queue steering messages and resume sessions; background sessions show Needs input state; push notifications and hooks surface pending decisions. Some commands differ on mobile.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Pause semantics beyond interrupt/stop, escalation policy and error visibility were not tested.

1. Claude Code on mobile

   Finding: The Claude iOS/Android app is a client for cloud sessions, projects, Remote Control and Dispatch (no code runs on the phone). Push notifications when Remote Control is active; local-only commands such as /plugin and /resume do not work; Bypass permissions cannot be selected from the app.

   Locator

   Work from your phone; Get push notifications; Limitations

   Retrieved

   2026-10-09

2. Remote Control

   Finding: Drive a local CLI/Desktop/VS Code session from claude.ai/code or the mobile app; machine must stay on; Pro/Max/Team/Enterprise only (no API keys, Bedrock/Vertex/Foundry, gateways); server mode default capacity 32; transcript stored on Anthropic servers while connected.

   Locator

   Requirements; Start a Remote Control session; Limitations; Connection and security

   Retrieved

   2026-10-09

3. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-10-09

4. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-10-09

5. Automate actions with hooks

   Finding: Hooks fire on events such as PreToolUse, PostToolUse, PermissionRequest, Notification, Stop, SubagentStop, SessionStart/End, TaskCreated/Completed, WorktreeCreate and can be commands, prompts, agents or HTTP calls; can block or auto-approve actions.

   Locator

   What you can automate; Hook types; Hooks and permission modes

   Retrieved

   2026-10-09

6. Manage sessions

   Finding: Sessions saved continuously as local JSONL transcripts; --continue, --resume, --from-pr, /branch; running the same session in two terminals interleaves one transcript; /export writes plain text.

   Locator

   Resume a session; Export and locate session data

   Retrieved

   2026-10-09

7. Data usage

   Finding: Consumer plans: training on opt-in setting, retention 5 years if allowed else 30 days; commercial plans no training by default, 30-day standard retention, ZDR for qualified Enterprise accounts; local transcripts kept 30 days by default; telemetry/error reporting defaults per provider with env-var opt-outs.

   Locator

   Data training policy; Data retention; Telemetry services; Default behaviors by API provider

   Retrieved

   2026-10-09

8. All settings

   Finding: Settings include agentPushNotifEnabled, inputNeededNotifEnabled, preferredNotifChannel, remoteControlAtStartup, cleanupPeriodDays, autoMemoryEnabled and sandbox keys.

   Locator

   Remote, desktop, and notifications

   Retrieved

   2026-10-09

9. Tools reference

   Finding: Lists built-in tools including TaskCreate/TaskUpdate/TaskList, TaskStop, WebFetch, WebSearch, Workflow; task tools are provided by default only on some model families.

   Locator

   Tool list; Task tool availability; WebFetch and WebSearch

   Retrieved

   2026-10-09

claude-code:COORD-06

Agent view (research preview) is a terminal fleet screen that groups background sessions under Needs input, Ready for review, Working and Completed, with filters, peek and attach; Desktop has a session sidebar. Dependency graphs and parent-child epic views were not found.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Availability

beta

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Agent view is research preview and shows only local background sessions; no fixture with many agents was run.

1. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-11-07

claude-code:COORD-07

Quick tasks are one prompt away; larger efforts use subagents, agent teams, projects (cloud beta) or dynamic workflows, configured in natural language or markdown. Agent teams are experimental.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Orchestration surfaces are spread across several preview features; no end-to-end scenario run.

1. Claude Code overview

   Finding: Documents terminal (macOS/Linux/WSL/Windows), VS Code, desktop (macOS, Windows x64/ARM64, Ubuntu/Debian beta), web and JetBrains surfaces; most surfaces need a Claude subscription or Console account; CLI, VS Code and JetBrains also support third-party providers.

   Locator

   Get started tabs; Terminal install; Desktop app; Web

   Retrieved

   2026-11-07

2. Agent teams

   Finding: Experimental, disabled by default (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1). Lead plus independent teammates with own context, direct messaging, shared task list with pending/in-progress/completed states and task dependencies, file-locked claiming; limitations include no in-process teammate resume, one team per session, status lag.

   Locator

   Enablement; shared task list; Assign and claim tasks; Limitations

   Retrieved

   2026-11-07

3. Run Claude Code programmatically

   Finding: claude -p runs non-interactively with exit codes, --output-format json/stream-json, --bare, --allowedTools; Agent SDK for Python/TypeScript.

   Locator

   Basic usage; Bare mode; Structured output

   Retrieved

   2026-11-07

4. Run agents in parallel

   Finding: Subagents, agent view (research preview), agent teams (experimental), dynamic workflows, and Projects. In every approach the workers are Claude sessions; other tools are exposed via MCP. Worktrees isolate parallel sessions.

   Locator

   Five ways to work on several tasks; Choose an approach; Check on running work

   Retrieved

   2026-11-07

5. Create custom subagents

   Finding: Subagents have their own context, tools, permissions, model, effort, maxTurns and optional worktree isolation; background subagents run concurrently; results return to the spawning conversation.

   Locator

   Built-in subagents; Supported frontmatter fields; Choose a model; Run subagents in foreground or background

   Retrieved

   2026-11-07

6. Dynamic workflows

   Finding: A JavaScript script Claude writes orchestrates many subagents in the background; on all paid plans, API, Bedrock, Agent Platform and Foundry; /workflows lists runs.

   Locator

   When to use a workflow

   Retrieved

   2026-11-07

7. Projects

   Finding: Projects public beta gradually rolls out to Pro/Max/Ultra; not Team/Enterprise. Parallel cloud threads, optionally local through Remote Control. Projects are single-user and not shareable.

   Locator

   Overview; Availability note

   Retrieved

   2026-11-07

8. Extend Claude with skills

   Finding: Skills follow the Agent Skills open standard; custom commands merged into skills; bundled skills include /loop, /batch, /code-review.

   Locator

   Bundled skills

   Retrieved

   2026-11-07

claude-code:RUN-01

Claude Code runs locally by default; cloud sessions run on Anthropic VMs (4 vCPU, 16 GB, 30 GB) with no separate compute charge; cloud sessions are available on Pro, Max, Ultra and Team plans, and for Enterprise users with premium seats or Chat + Claude Code seats, optionally on customer-run self-hosted runners (Team/Enterprise beta); Remote Control lets other devices drive a local session.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Cloud: Pro, Max, Ultra, Team, Enterprise; self-hosted: Team, Enterprise (public beta)

Surface

Claude Code CLI; Claude Code web/mobile

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Region and exact idle-expiry durations were not found.

1. Claude Code overview

   Finding: Documents terminal (macOS/Linux/WSL/Windows), VS Code, desktop (macOS, Windows x64/ARM64, Ubuntu/Debian beta), web and JetBrains surfaces; most surfaces need a Claude subscription or Console account; CLI, VS Code and JetBrains also support third-party providers.

   Locator

   Get started tabs; Terminal install; Desktop app; Web

   Retrieved

   2026-11-07

2. Platforms and integrations

   Finding: Compares CLI, Desktop, VS Code, JetBrains, Web and Mobile; CLI is the most complete surface (scripting and Agent SDK are CLI-only); Dispatch, Remote Control, Channels, Slack, self-hosted environments and scheduled tasks listed with where Claude runs and plan requirements.

   Locator

   Where to run Claude Code; Work when you are away from your terminal

   Retrieved

   2026-11-07

3. Claude Code on mobile

   Finding: The Claude iOS/Android app is a client for cloud sessions, projects, Remote Control and Dispatch (no code runs on the phone). Push notifications when Remote Control is active; local-only commands such as /plugin and /resume do not work; Bypass permissions cannot be selected from the app.

   Locator

   Work from your phone; Get push notifications; Limitations

   Retrieved

   2026-11-07

4. Remote Control

   Finding: Drive a local CLI/Desktop/VS Code session from claude.ai/code or the mobile app; machine must stay on; Pro/Max/Team/Enterprise only (no API keys, Bedrock/Vertex/Foundry, gateways); server mode default capacity 32; transcript stored on Anthropic servers while connected.

   Locator

   Requirements; Start a Remote Control session; Limitations; Connection and security

   Retrieved

   2026-11-07

5. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-11-07

6. Configure cloud environments

   Finding: Anthropic-hosted sessions use fresh Ubuntu 24.04 x86_64 VMs with approximate ceilings of 4 vCPUs, 16 GB RAM and 30 GB disk. Setup under roughly five minutes is cached as a filesystem snapshot, expiring after roughly seven days. Foreground Bash defaults to two minutes, up to ten; timed-out commands move to background except sleep-prefixed commands. Unattended background commands have a separate time limit. Idle VMs pause and can be reclaimed. Network secrets are available on Pro, Max, Ultra, Team and Enterprise, with role/configuration requirements.

   Locator

   Installed tools; Time limits; Setup scripts; Environment caching; Network access

   Retrieved

   2026-11-07

7. Self-hosted environments

   Finding: Public beta on Team and Enterprise (off by default); runners on customer infrastructure execute cloud sessions; unavailable with ZDR or HIPAA; model inference still via Anthropic API unless Bedrock/Agent Platform configured; session content goes to Anthropic.

   Locator

   How self-hosted environments work; Availability and limitations

   Retrieved

   2026-11-07

8. Projects

   Finding: Projects public beta gradually rolls out to Pro/Max/Ultra; not Team/Enterprise. Parallel cloud threads, optionally local through Remote Control. Projects are single-user and not shareable.

   Locator

   Overview; Availability note

   Retrieved

   2026-11-07

9. All settings

   Finding: Settings include agentPushNotifEnabled, inputNeededNotifEnabled, preferredNotifChannel, remoteControlAtStartup, cleanupPeriodDays, autoMemoryEnabled and sandbox keys.

   Locator

   Remote, desktop, and notifications

   Retrieved

   2026-11-07

claude-code:RUN-02

Isolation options: git worktrees per session, an OS-level sandbox for shell commands (macOS/Linux/WSL2, off by default), dev containers or the sandbox runtime for the whole process, and a fresh VM per cloud session. The CLI by default runs with the user permissions and file tools outside the sandbox.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Current first-party documentation read 2026-11-07; changelog through v2.1.319

• No escape or boundary test was run.

1. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-10-09

2. Create custom subagents

   Finding: Subagents have their own context, tools, permissions, model, effort, maxTurns and optional worktree isolation; background subagents run concurrently; results return to the spawning conversation.

   Locator

   Built-in subagents; Supported frontmatter fields; Choose a model; Run subagents in foreground or background

   Retrieved

   2026-10-09

3. Run parallel sessions with worktrees

   Finding: claude --worktree creates .claude/worktrees/<name> on a new branch; requires a git repository (hooks can replace git logic); subagents can use isolation: worktree.

   Locator

   Start Claude in a worktree; Isolate subagents; Non-git version control

   Retrieved

   2026-10-09

4. Configure the sandboxed Bash tool

   Finding: OS-enforced sandbox for shell commands on macOS, Linux and WSL2 (off by default; native Windows runs unsandboxed); built on open source sandbox-runtime; reads allowed broadly by default, writes limited to working directory, network via domain-allowlist proxy; file tools, MCP servers and hooks run outside it.

   Locator

   What the sandbox restricts; What runs outside the sandbox

   Retrieved

   2026-10-09

5. Choose a sandbox environment

   Finding: Compares built-in sandbox, sandbox runtime, dev containers, containers and VMs for isolating Claude Code.

   Locator

   Sandbox runtime; dev containers; VMs

   Retrieved

   2026-10-09

6. Configure cloud environments

   Finding: Anthropic-hosted sessions use fresh Ubuntu 24.04 x86_64 VMs with approximate ceilings of 4 vCPUs, 16 GB RAM and 30 GB disk. Setup under roughly five minutes is cached as a filesystem snapshot, expiring after roughly seven days. Foreground Bash defaults to two minutes, up to ten; timed-out commands move to background except sleep-prefixed commands. Unattended background commands have a separate time limit. Idle VMs pause and can be reclaimed. Network secrets are available on Pro, Max, Ultra, Team and Enterprise, with role/configuration requirements.

   Locator

   Installed tools; Time limits; Setup scripts; Environment caching; Network access

   Retrieved

   2026-11-07

7. Self-hosted environments

   Finding: Public beta on Team and Enterprise (off by default); runners on customer infrastructure execute cloud sessions; unavailable with ZDR or HIPAA; model inference still via Anthropic API unless Bedrock/Agent Platform configured; session content goes to Anthropic.

   Locator

   How self-hosted environments work; Availability and limitations

   Retrieved

   2026-11-07

8. Run parallel sessions with worktrees

   Finding: The --worktree or -w flag starts a session in a separate Git working directory, by default .claude/worktrees/<name>, on a new worktree-<name> branch. Subagents can use isolation: worktree. Worktrees share repository history and Git metadata; this documents file separation, not a security sandbox. Source check only; no session was run.

   Locator

   Start Claude in a worktree; Isolate subagents with worktrees; What worktrees share with the main checkout

   Retrieved

   2026-11-07

9. Choose a sandbox environment

   Finding: Compares built-in sandbox, sandbox runtime, dev containers, containers and VMs for isolating Claude Code.

   Locator

   Sandbox runtime; dev containers; VMs

   Retrieved

   2026-11-07

claude-code:RUN-03

Sessions persist as local transcripts and resume; cloud sessions restore history after VM reclamation but lose background processes; Remote Control reconnects after sleep, and server mode gives up after roughly 10 minutes without network; background local sessions survive sleep but not shutdown.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI; Claude Code web/mobile

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No crash or restart experiments were run; claims are from documentation.

1. Agent teams

   Finding: Experimental, disabled by default (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1). Lead plus independent teammates with own context, direct messaging, shared task list with pending/in-progress/completed states and task dependencies, file-locked claiming; limitations include no in-process teammate resume, one team per session, status lag.

   Locator

   Enablement; shared task list; Assign and claim tasks; Limitations

   Retrieved

   2026-10-09

2. Remote Control

   Finding: Drive a local CLI/Desktop/VS Code session from claude.ai/code or the mobile app; machine must stay on; Pro/Max/Team/Enterprise only (no API keys, Bedrock/Vertex/Foundry, gateways); server mode default capacity 32; transcript stored on Anthropic servers while connected.

   Locator

   Requirements; Start a Remote Control session; Limitations; Connection and security

   Retrieved

   2026-10-09

3. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-10-09

4. Run prompts on a schedule

   Finding: /loop and cron tools are session-scoped (1 minute minimum, recurring tasks expire after 7 days, only fire while Claude Code is running and idle, no catch-up for missed fires); cloud routines minimum 1 hour; Desktop tasks need machine on.

   Locator

   Compare scheduling options; Seven-day expiry; Limitations

   Retrieved

   2026-10-09

5. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-10-09

6. Create custom subagents

   Finding: Subagents have their own context, tools, permissions, model, effort, maxTurns and optional worktree isolation; background subagents run concurrently; results return to the spawning conversation.

   Locator

   Built-in subagents; Supported frontmatter fields; Choose a model; Run subagents in foreground or background

   Retrieved

   2026-10-09

7. How Claude remembers your project

   Finding: CLAUDE.md at managed/user/project/local scope with imports and .claude/rules; AGENTS.md read when no CLAUDE.md (v2.1.277+); auto memory stored per project under ~/.claude/projects/<project>/memory/ as editable markdown; /memory browses and toggles.

   Locator

   CLAUDE.md files; AGENTS.md; Auto memory; View and edit with /memory

   Retrieved

   2026-10-09

8. Manage sessions

   Finding: Sessions saved continuously as local JSONL transcripts; --continue, --resume, --from-pr, /branch; running the same session in two terminals interleaves one transcript; /export writes plain text.

   Locator

   Resume a session; Export and locate session data

   Retrieved

   2026-10-09

9. Checkpointing

   Finding: Automatic snapshots before each prompt (100 most recent kept), /rewind restores code and/or conversation; file snapshots swept by retention (~30 days by default).

   Locator

   How checkpoints work

   Retrieved

   2026-10-09

10. Data usage

    Finding: Consumer plans: training on opt-in setting, retention 5 years if allowed else 30 days; commercial plans no training by default, 30-day standard retention, ZDR for qualified Enterprise accounts; local transcripts kept 30 days by default; telemetry/error reporting defaults per provider with env-var opt-outs.

    Locator

    Data training policy; Data retention; Telemetry services; Default behaviors by API provider

    Retrieved

    2026-10-09

11. Claude Code changelog

    Finding: Newest entry is v2.1.295 (2026-10-08), then v2.1.294 (2026-10-08); v2.1.287 to v2.1.295 are dated Oct 1 to 8. v2.1.295 negotiates MCP protocol 2026-07-28 by default; MCP_PROTOCOL_NEGOTIATION=legacy opts out.

    Locator

    Version 2.1.294 (October 8, 2026) and recent entries

    Retrieved

    2026-10-09

    Source date

    2026-10-08

claude-code:RUN-04

Cloud environments offer setup scripts, a filesystem-snapshot cache, network access levels and pre-installed toolchains; local runs inherit the host. Process state, ports and automatic repair are not covered.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code web/mobile

Version

Current first-party documentation read 2026-11-07; changelog through v2.1.319

• Cloud-only findings; the local CLI inherits whatever the host provides.

1. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-10-09

2. Run parallel sessions with worktrees

   Finding: claude --worktree creates .claude/worktrees/<name> on a new branch; requires a git repository (hooks can replace git logic); subagents can use isolation: worktree.

   Locator

   Start Claude in a worktree; Isolate subagents; Non-git version control

   Retrieved

   2026-10-09

3. Automate actions with hooks

   Finding: Hooks fire on events such as PreToolUse, PostToolUse, PermissionRequest, Notification, Stop, SubagentStop, SessionStart/End, TaskCreated/Completed, WorktreeCreate and can be commands, prompts, agents or HTTP calls; can block or auto-approve actions.

   Locator

   What you can automate; Hook types; Hooks and permission modes

   Retrieved

   2026-10-09

4. Configure cloud environments

   Finding: Anthropic-hosted sessions use fresh Ubuntu 24.04 x86_64 VMs with approximate ceilings of 4 vCPUs, 16 GB RAM and 30 GB disk. Setup under roughly five minutes is cached as a filesystem snapshot, expiring after roughly seven days. Foreground Bash defaults to two minutes, up to ten; timed-out commands move to background except sleep-prefixed commands. Unattended background commands have a separate time limit. Idle VMs pause and can be reclaimed. Network secrets are available on Pro, Max, Ultra, Team and Enterprise, with role/configuration requirements.

   Locator

   Installed tools; Time limits; Setup scripts; Environment caching; Network access

   Retrieved

   2026-11-07

claude-code:RUN-05

Limits include print-mode --max-turns and --max-budget-usd (client-side estimate), subagent maxTurns, tool timeouts, plan usage limits with automatic continue, organization spend limits, 10 automatic retries and --fallback-model.

Checked

2026-11-07 · checked 1 week ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• --max-budget-usd and --max-turns apply to print mode only; subagent maxTurns caps individual subagents in any session.

1. Configure cloud environments

   Finding: Anthropic-hosted sessions use fresh Ubuntu 24.04 x86_64 VMs with approximate ceilings of 4 vCPUs, 16 GB RAM and 30 GB disk. Setup under roughly five minutes is cached as a filesystem snapshot, expiring after roughly seven days. Foreground Bash defaults to two minutes, up to ten; timed-out commands move to background except sleep-prefixed commands. Unattended background commands have a separate time limit. Idle VMs pause and can be reclaimed. Network secrets are available on Pro, Max, Ultra, Team and Enterprise, with role/configuration requirements.

   Locator

   Installed tools; Time limits; Setup scripts; Environment caching; Network access

   Retrieved

   2026-11-07

2. Create custom subagents

   Finding: Subagents have their own context, tools, permissions, model, effort, maxTurns and optional worktree isolation; background subagents run concurrently; results return to the spawning conversation.

   Locator

   Built-in subagents; Supported frontmatter fields; Choose a model; Run subagents in foreground or background

   Retrieved

   2026-11-07

3. Connect Claude Code to tools via MCP

   Finding: MCP client supports http, sse, stdio and websocket servers with OAuth 2.0, local/project/user scopes and claude.ai connectors; claude mcp serve exposes Claude Code as a stdio MCP server; channels push events.

   Locator

   Installing MCP servers; Authenticate with remote MCP servers; Use Claude Code as an MCP server; Scopes

   Retrieved

   2026-11-07

4. Manage costs effectively

   Finding: /usage shows client-side session cost estimate at list price, plan usage bars and attribution to skills/subagents/plugins/MCP; usage credits past plan limits; --max-budget-usd and --max-turns in print mode; org spend limits for Team/Enterprise; average enterprise cost about $13 per developer per active day.

   Locator

   Track your costs; Plan usage breakdown; Usage credits; Manage costs for your organization

   Retrieved

   2026-11-07

5. CLI reference

   Finding: Documents claude agents, claude remote-control, claude mcp, claude gateway; flags --worktree, --max-turns and --max-budget-usd (print mode only), --fallback-model, --output-format, --permission-mode, --bare, --channels.

   Locator

   CLI commands; CLI flags

   Retrieved

   2026-11-07

6. Interactive mode

   Finding: Task list checklist; automatic continue after a claude.ai usage limit resets in interactive subscription sessions; background Bash commands.

   Locator

   Task list; Wait for a usage limit to reset

   Retrieved

   2026-11-07

7. Error reference

   Finding: Claude Code retries transient failures up to 10 times with exponential backoff; partial-response failures are not always retried.

   Locator

   Automatic retries

   Retrieved

   2026-11-07

8. Use Claude Code with your Pro or Max plan

   Finding: Pro, Max and Ultra include Code in a shared subscription quota; Max and Ultra platform API credits cannot pay Code usage. ANTHROPIC_API_KEY causes API billing instead of subscription. No workload usage test.

   Locator

   Connect; usage limits; Claude Code billing

   Retrieved

   2026-11-07

claude-code:RUN-06

Operational statistics are available via /usage, /insights, OpenTelemetry metrics, logs and traces (beta), and the Team/Enterprise analytics dashboard; failure reasons are documented in the error reference.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No accuracy or latency test was run.

1. Manage sessions

   Finding: Sessions saved continuously as local JSONL transcripts; --continue, --resume, --from-pr, /branch; running the same session in two terminals interleaves one transcript; /export writes plain text.

   Locator

   Resume a session; Export and locate session data

   Retrieved

   2026-10-09

2. Manage costs effectively

   Finding: /usage shows client-side session cost estimate at list price, plan usage bars and attribution to skills/subagents/plugins/MCP; usage credits past plan limits; --max-budget-usd and --max-turns in print mode; org spend limits for Team/Enterprise; average enterprise cost about $13 per developer per active day.

   Locator

   Track your costs; Plan usage breakdown; Usage credits; Manage costs for your organization

   Retrieved

   2026-10-09

3. Monitoring

   Finding: OpenTelemetry export of metrics (sessions, cost, tokens, lines of code, commits, PRs, active time), events and optional traces.

   Locator

   Quick start; Metrics; Events; Traces

   Retrieved

   2026-10-09

4. Track team usage with analytics

   Finding: Team/Enterprise dashboard at claude.ai/analytics/claude-code with usage and contribution metrics, leaderboard and CSV export; Console dashboard for API customers.

   Locator

   Access analytics for Team and Enterprise

   Retrieved

   2026-10-09

5. Commands

   Finding: Lists /usage, /export, /import (Codex, Gemini CLI, Cursor), /rewind, /branch, /fork, /goal, /loop, /schedule, /teleport, /desktop, /code-review, /security-review, /install-github-app, /web-setup, /tasks, /sandbox, /diff.

   Locator

   All commands

   Retrieved

   2026-10-09

6. Error reference

   Finding: Claude Code retries transient failures up to 10 times with exponential backoff; partial-response failures are not always retried.

   Locator

   Automatic retries

   Retrieved

   2026-10-09

claude-code:RUN-07

Cost views: /usage session cost (client-side estimate at list price unless an administrator's managed modelPricing table sets contracted rates), plan usage bars, OTel cost and token metrics, Console and Team/Enterprise dashboards; no separate compute charge for cloud VMs. Estimates can differ from the bill.

Checked

2026-11-07 · checked 1 week ago

Cross-check

cross-checked 2026-11-07

Plan

Pro, Max, Team, Enterprise, Console API

Surface

Claude Code CLI; Claude Code web/mobile

Version

First-party public docs reread 2026-11-07; changelog through v2.1.319

• Costs reconcile against provider billing only outside Claude Code.

1. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-11-07

2. Manage costs effectively

   Finding: /usage shows client-side session cost estimate at list price, plan usage bars and attribution to skills/subagents/plugins/MCP; usage credits past plan limits; --max-budget-usd and --max-turns in print mode; org spend limits for Team/Enterprise; average enterprise cost about $13 per developer per active day.

   Locator

   Track your costs; Plan usage breakdown; Usage credits; Manage costs for your organization

   Retrieved

   2026-11-07

3. Monitoring

   Finding: OpenTelemetry export of metrics (sessions, cost, tokens, lines of code, commits, PRs, active time), events and optional traces.

   Locator

   Quick start; Metrics; Events; Traces

   Retrieved

   2026-11-07

4. Track team usage with analytics

   Finding: Team/Enterprise dashboard at claude.ai/analytics/claude-code with usage and contribution metrics, leaderboard and CSV export; Console dashboard for API customers.

   Locator

   Access analytics for Team and Enterprise

   Retrieved

   2026-11-07

5. Use Claude Code with your Pro or Max plan

   Finding: Pro, Max and Ultra include Code in a shared subscription quota; Max and Ultra platform API credits cannot pay Code usage. ANTHROPIC_API_KEY causes API billing instead of subscription. No workload usage test.

   Locator

   Connect; usage limits; Claude Code billing

   Retrieved

   2026-11-07

claude-code:RUN-08

Compute-aware scheduling is limited: Remote Control and self-hosted runners cap concurrency, Bash commands have timeouts and can be cancelled, and subscription limits are shared. Workflow fan-outs run up to 16 agents by default and fewer on CPU-limited hosts, configurable from 1 to 256 (v2.1.269 or later). RAM-aware queues, check deduplication and fairness were not found.

Checked

2026-11-07 · checked 1 week ago

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Current first-party documentation read 2026-11-07; changelog through v2.1.319

• Workflow concurrency adapts to available CPUs; RAM-aware scheduling was not established.

1. Remote Control

   Finding: Drive a local CLI/Desktop/VS Code session from claude.ai/code or the mobile app; machine must stay on; Pro/Max/Team/Enterprise only (no API keys, Bedrock/Vertex/Foundry, gateways); server mode default capacity 32; transcript stored on Anthropic servers while connected.

   Locator

   Requirements; Start a Remote Control session; Limitations; Connection and security

   Retrieved

   2026-10-09

2. Self-hosted environments

   Finding: Public beta on Team and Enterprise (off by default); runners on customer infrastructure execute cloud sessions; unavailable with ZDR or HIPAA; model inference still via Anthropic API unless Bedrock/Agent Platform configured; session content goes to Anthropic.

   Locator

   How self-hosted environments work; Availability and limitations

   Retrieved

   2026-10-09

3. Manage multiple agents with agent view

   Finding: Research preview. claude agents dispatches and monitors background local sessions; sessions move into git worktrees before editing; background sessions use subscription quota like interactive ones, stop if the machine shuts down; claude agents --json output.

   Locator

   Quick start; How file edits are isolated; Limitations; Version history

   Retrieved

   2026-10-09

4. Dynamic workflows

   Finding: A JavaScript script Claude writes orchestrates many subagents in the background; on all paid plans, API, Bedrock, Agent Platform and Foundry; /workflows lists runs.

   Locator

   When to use a workflow

   Retrieved

   2026-10-09

5. Interactive mode

   Finding: Task list checklist; automatic continue after a claude.ai usage limit resets in interactive subscription sessions; background Bash commands.

   Locator

   Task list; Wait for a usage limit to reset

   Retrieved

   2026-10-09

6. Use Claude Code with your Pro or Max plan

   Finding: Pro/Max usage limits are shared across Claude and Claude Code; options at limit include usage credits, Console API credits or waiting; monthly API credits included with Max cannot be used in Claude Code; ANTHROPIC_API_KEY overrides subscription auth and causes API charges.

   Locator

   Connect; usage limits; Claude Code billing

   Retrieved

   2026-10-09

7. Configure cloud environments

   Finding: Anthropic-hosted sessions use fresh Ubuntu 24.04 x86_64 VMs with approximate ceilings of 4 vCPUs, 16 GB RAM and 30 GB disk. Setup under roughly five minutes is cached as a filesystem snapshot, expiring after roughly seven days. Foreground Bash defaults to two minutes, up to ten; timed-out commands move to background except sleep-prefixed commands. Unattended background commands have a separate time limit. Idle VMs pause and can be reclaimed. Network secrets are available on Pro, Max, Ultra, Team and Enterprise, with role/configuration requirements.

   Locator

   Installed tools; Time limits; Setup scripts; Environment caching; Network access

   Retrieved

   2026-11-07

8. Orchestrate subagents at scale with dynamic workflows

   Finding: Workflows run up to 16 concurrent agents by default and fewer on hosts or containers with fewer CPUs; CLAUDE_CODE_WORKFLOW_MAX_CONCURRENT_AGENTS accepts 1 to 256 and needs v2.1.269 or later.

   Locator

   How a workflow runs > Behavior and limits

   Retrieved

   2026-11-07

claude-code:TEAM-01

Live multi-human collaboration is limited. Cloud sessions can be shared by link (Team visibility or public for Pro/Max) but recipients see the latest state without real-time updates; multiple terminals on one session interleave a single transcript.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Team/Enterprise: Private or Team visibility; Pro/Max: Private or Public

Surface

Claude Code CLI; Claude Code web/mobile

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• No presence, co-prompting or comment threads were found for the CLI; no multi-user test was run.

1. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-10-09

2. Manage sessions

   Finding: Sessions saved continuously as local JSONL transcripts; --continue, --resume, --from-pr, /branch; running the same session in two terminals interleaves one transcript; /export writes plain text.

   Locator

   Resume a session; Export and locate session data

   Retrieved

   2026-10-09

3. Share session output as artifacts

   Finding: Artifacts publish live interactive pages from a session to a private claude.ai URL that can be shared with the organization or a public link; Pro, Max, Team and Enterprise with /login.

   Locator

   Overview; Availability

   Retrieved

   2026-10-09

claude-code:TEAM-02

Roles and access are managed at the Claude account level: Owners and admins configure cloud environments and settings; Team adds SSO and central billing; Enterprise adds role-based access, SCIM, audit logs and spend limits.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Team, Enterprise

Surface

Claude Code CLI; Claude Code desktop; Claude Code web/mobile; Claude Code IDE extensions

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Project-level and task-level roles were not found; role definitions were not read in detail.

1. Authentication

   Finding: Account types: Claude Pro/Max, Teams/Enterprise, Claude Console, Bedrock/Vertex(Agent Platform)/Foundry, Claude apps gateway. Multiple accounts via separate CLAUDE_CONFIG_DIR directories. ANTHROPIC_API_KEY approval skips login.

   Locator

   Log in to Claude Code; Log in with multiple accounts

   Retrieved

   2026-10-09

2. Use Claude Code in the cloud

   Finding: Cloud sessions on Pro, Max, Team and Enterprise (premium seats); Anthropic-managed VM by default; clone/PR creation requires GitHub (GHES for Team/Enterprise); non-GitHub repos via local bundle <100 MB; idle sessions pause and VMs are reclaimed, conversation history restored on reopen; shared rate limits, no separate compute charge; CLI --cloud and --teleport (one-way cloud to terminal); auto-fix of PR CI failures and review comments needs the Claude GitHub App.

   Locator

   Plans; GitHub authentication options; Environment expired; Limitations; Share sessions; Auto-fix pull requests

   Retrieved

   2026-10-09

3. Set up Claude Code for your organization

   Finding: Managed settings delivered via claude.ai admin console, MDM or file; SSO, SCIM and seat assignment at the Claude account level; Owners govern cloud environments and the default environment; org effort limits enforced server-side.

   Locator

   API providers; managed settings; usage monitoring; cloud sessions admin

   Retrieved

   2026-10-09

4. Claude pricing

   Finding: Free $0 (Claude Code: No); Pro $17/month annual or $20 monthly (Claude Code: Yes); Max from $100/month (5x or 20x Pro usage); Team standard seat $20 annual/$25 monthly, premium seat $100 annual/$125 monthly (Claude Code included); Enterprise US$20/seat/month plus usage at API rates; usage limits apply; prices exclude tax; model training opt-out on consumer plans, none by default on Team/Enterprise.

   Locator

   Individual and Team and Enterprise plan cards; feature comparison

   Retrieved

   2026-10-09

claude-code:TEAM-03

Approvals are per-user permission prompts governed by permission modes plus allow/ask/deny rules, hooks and organization managed settings. There is no separate approver identity; mobile and Remote Control sessions can answer prompts.

Checked

2026-10-09 · checked 1 month ago

Cross-check

cross-checked 2026-11-07

Plan

Not specified in this finding

Surface

Claude Code CLI

Version

Claude Code v2.1.295 (released 2026-10-08); official docs and pricing page read 2026-10-09

• Permission behaviour was read from documentation and not exercised.

1. Claude Code on mobile

   Finding: The Claude iOS/Android app is a client for cloud sessions, projects, Remote Control and Dispatch (no code runs on the phone). Push notifications when Remote Control is active; local-only commands such as /plugin and /resume do not work; Bypass permissions cannot be selected from the app.

   Locator

   Work from your phone; Get push notifications; Limitations

   Retrieved

   2026-10-09

2. Configure the sandboxed Bash tool

   Finding: OS-enforced sandbox for shell commands on macOS, Linux and WSL2 (off by default; native Windows runs unsandboxed); built on open source sandbox-runtime; reads allowed broadly by default, writes limited to working directory, network via domain-allowlist proxy; file tools, MCP servers and hooks run outside it.

   Locator

   What the sandbox restricts; What runs outside the sandbox

   Retrieved

   2026-10-09

3. Choose a permission mode

   Finding: Modes default/Manual, acceptEdits, plan, auto (classifier), dontAsk, bypassPermissions; deny rules apply in every mode; auto is the built-in starting mode for interactive terminal and VS Code sessions from v2.1.283.

   Locator

   Available modes; Actions no mode auto-approves

   Retrieved

   2026-10-09

4. Automate actions with hooks

   Finding: Hooks fire on events such as PreToolUse, PostToolUse, PermissionRequest, Notification, Stop, SubagentStop, SessionStart/End, TaskCreated/Completed, WorktreeCreate and can be commands, prompts, agents or HTTP calls; can block or auto-approve actions.

   Locator

   What you can automate; Hook types; Hooks and permission modes

   Retrieved

   2026-10-09

5. Connect Claude Code to tools via MCP

   Finding: MCP client supports http, sse, stdio and websocket servers with OAuth 2.0, local/project/user scopes and claude.ai connectors; claude mcp serve exposes Claude Code as a stdio MCP server; channels push events.

   Locator

   Installing MCP servers; Authenticate with remote MCP servers; Use Claude Code as an MCP server; Scopes

   Retrieved

   2026-10-09

6. Set up Claude Code for your organization

   Finding: Managed settings delivered via claude.ai admin console, MDM or file; SSO, SCIM and seat assignment at the Claude account level; Owners govern cloud environments and the default environment; org effort limits enforced server-side.

   Locator

   API providers; managed settings; usage monitoring; cloud sessions admin

   Retrieved

   2026-10-09

7. Configure permissions

   Finding: Allow, ask and deny rules for tools and Bash command patterns (for example git commit and git push patterns) layered on permission modes; deny rules apply in every mode; managed settings take precedence over local settings.

   Locator

   Manage permissions; permission rule syntax; protected paths

   Retrieved

   2026-10-09