Ralph TUI
Open-source desktop app for Mac, Windows and Linux. Agents run on your computer or on your own server.
Product websiteralph-tui.com Compare with others Missing or wrong data?
- Desktop app
- Mac, Windows and Linux
- Browser
- No
- Phone
- No
- Command line
- Yes
- Agents
- Claude Code, Codex, Cursor Agent, Gemini CLI, OpenCode and Pi
- Your subscription
- Works with your own plan
- Agents run
- On your computer and on your own server
- Laptop closed
- Keeps going
- Isolation
- Worktree
- Parallel agents
- Yes
- Scheduled runs
- Yes
- Coordinator agent
- No
- Built-in issues
- Yes
- Pull requests
- No
- Review screen
- No
- Team
- Not listed
- Self-hosting
- Yes
- Licence
- MIT
- Price from
- Free
- How it charges
- No paid plan and bring your own key
- Paid plans from
- Not listed
Receipts
- Checked
- · oldest answer 2026-11-08
- Findings
- 40 published · 30 cross-checked by a second agent · read them all
- Sources
- 20 public pages on 2 sites: github.com, ralph-tui.com
- Screenshot
- raw.githubusercontent.com, retrieved · first party repository · shows appearance, not capability
- Data
- /products/ralph-tui.json · .md
- Method
- Same rules for all 103 researched ADEs. How we check · Report wrong data
What it does
Where you use it
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
Source and scope for Where you use itAgents it runs
Release v0.12.0 names Claude Code, OpenCode, Factory Droid, Cursor CLI, Gemini CLI, Codex and Kiro CLI, plus a built-in Pi Coding Agent plugin; agent and model can be overridden. Amp appears only in later main-branch docs. The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
Source and scope for Agents it runsAccounts and subscriptions
Ralph TUI launches the installed agent CLI and by default strips *_API_KEY variables from its environment so the CLI uses its own sign-in, such as a Claude Max subscription. An API key can be passed through deliberately.
Source and scope for Accounts and subscriptionsWhere agents run
Run engine on the local machine or a user-controlled VPS/CI host and control it from another TUI.
Source and scope for Where agents runHow work is kept apart
Ralph TUI uses Git worktrees for parallel file separation and optional bubblewrap/macOS sandbox-exec process policies. It provides no built-in Docker/container or VM mode.
Source and scope for How work is kept apartParallel work
Optional parallel worker pool executes independent tasks simultaneously in worktrees.
Source and scope for Parallel workScheduled runs
An external program can start work through the documented remote WebSocket orchestrate:start API. The server owns execution independently of a human TUI action for each run.
Source and scope for Scheduled runsGit and pull requests
Ralph TUI integrates worktree merging, conflict handling and optional commits, then prints a manual gh pr create command for the user.
Source and scope for Git and pull requestsLicence and source
Ralph TUI application source uses MIT license.
Source and scope for Licence and sourcePrice
Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
Source and scope for PriceTyped filters
Every value links to its receipt below. Read plan and surface restrictions before using a filter to decide.
| Filter | Value |
|---|---|
| Desktop OS: macOS | Yes |
| Desktop OS: Windows | Yes |
| Desktop OS: Linux | Yes |
| Web client | No |
| Phone access: Native iOS | No |
| Phone access: Native Android | No |
| Phone access: Phone browser | No |
| CLI | Yes |
| Coding agents: Claude Code | Yes cross-checked |
| Coding agents: Codex | Yes cross-checked |
| Coding agents: Cursor Agent | Yes cross-checked |
| Coding agents: Gemini CLI | Yes cross-checked |
| Coding agents: OpenCode | Yes cross-checked |
| Coding agents: Pi | Yes cross-checked |
| Coding agents: Grok | Unknown |
| Coding agents: Aider | Unknown |
| Coding agents: Antigravity | Unknown |
| Use my subscription | Yes cross-checked |
| Where agents run: On your computer | Yes cross-checked |
| Where agents run: On your own server | Yes cross-checked |
| Where agents run: In the vendor's cloud | No |
| Workspace isolation: Git worktree | Yes |
| Workspace isolation: Container | No |
| Workspace isolation: VM or microVM | No |
| Parallel runs | Yes cross-checked |
| Coordinator | No |
| Native issue tracking | Yes cross-checked |
| Unattended triggers | Yes |
| Shared prompting | Unknown |
| Team roles | No |
| SSO | No |
| Pull request workflow | No |
| Review interface | No |
| Product license: Apache-2.0 | No. MIT cross-checked |
| Product license: MIT | Yes. MIT cross-checked |
| Product license: GPL-3.0 | No. MIT cross-checked |
| Product license: GPL-3.0-or-later | No. MIT cross-checked |
| Product license: AGPL-3.0 | No. MIT cross-checked |
| Product license: AGPL-3.0-or-later | No. MIT cross-checked |
| Product license: Elastic License 2.0 | No. MIT cross-checked |
| Product license: Custom terms | No. MIT cross-checked |
| Product license: Proprietary | No. MIT cross-checked |
| Self-hosting | Yes cross-checked |
| Telemetry opt-out | No |
| Free access | Yes cross-checked |
| Starting price: Free | Yes. Free cross-checked |
| Starting price: Up to $20/month | No. Free cross-checked |
| Starting price: $21–40/month | No. Free cross-checked |
| Starting price: $41–100/month | No. Free cross-checked |
| Starting price: Over $100/month | No. Free cross-checked |
| Starting price: Contact for pricing | No. Free cross-checked |
| Pricing model: Free, no paid plan | Yes |
| Pricing model: Free plan | No |
| Pricing model: Subscription | No |
| Pricing model: Per seat | No |
| Pricing model: Usage-based | No |
| Pricing model: Per machine or compute | No |
| Pricing model: Bring your own key | Yes |
| Pricing model: One-time purchase | No |
| Paid plans from: No paid plan | Unknown |
| Paid plans from: $10 or less/month | Unknown |
| Paid plans from: $11–20/month | Unknown |
| Paid plans from: $21–40/month | Unknown |
| Paid plans from: $41–100/month | Unknown |
| Paid plans from: Over $100/month | Unknown |
| Paid plans from: Contact sales | Unknown |
| Lifecycle status: Active | Unknown |
| Lifecycle status: Sunsetting | Unknown |
| Keeps going with laptop closed | Yes cross-checked |
| Open-source license | Yes. MIT cross-checked |
| Any desktop app | Yes |
| Paid plan up to $20/month | Unknown |
| Paid plan up to $40/month | Unknown |
| Any phone client | No |
Filter receipts
Desktop OS: macOS: Yes
Desktop OS: macOS
Typed value: Yes.
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Desktop os: The installation guide provides macOS/Linux and native Windows PowerShell prerequisites and global installation.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Ralph TUI: website/content/docs/getting-started/installation.mdx
Finding: The first-party installation page provides macOS/Linux and native Windows PowerShell Bun setup followed by global ralph-tui installation. The primary product is a terminal CLI/TUI, not a desktop GUI.
Desktop OS: Windows: Yes
Desktop OS: Windows
Typed value: Yes.
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Desktop os: The installation guide provides macOS/Linux and native Windows PowerShell prerequisites and global installation.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Ralph TUI: website/content/docs/getting-started/installation.mdx
Finding: The first-party installation page provides macOS/Linux and native Windows PowerShell Bun setup followed by global ralph-tui installation. The primary product is a terminal CLI/TUI, not a desktop GUI.
Desktop OS: Linux: Yes
Desktop OS: Linux
Typed value: Yes.
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Desktop os: The installation guide provides macOS/Linux and native Windows PowerShell prerequisites and global installation.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Ralph TUI: website/content/docs/getting-started/installation.mdx
Finding: The first-party installation page provides macOS/Linux and native Windows PowerShell Bun setup followed by global ralph-tui installation. The primary product is a terminal CLI/TUI, not a desktop GUI.
Web client: No
Web client
Typed value: No.
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Mobile apps: There is no native Ralph TUI iOS/Android application in the scoped client code.
- Browser: The product exposes terminal clients and a WebSocket control API, without a browser application.
- Mobile browser: The remote listener is a control API, without a phone-browser work UI.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
Phone access: Native iOS: No
Phone access: Native iOS
Typed value: No.
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Mobile apps: There is no native Ralph TUI iOS/Android application in the scoped client code.
- Browser: The product exposes terminal clients and a WebSocket control API, without a browser application.
- Mobile browser: The remote listener is a control API, without a phone-browser work UI.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
Phone access: Native Android: No
Phone access: Native Android
Typed value: No.
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Mobile apps: There is no native Ralph TUI iOS/Android application in the scoped client code.
- Browser: The product exposes terminal clients and a WebSocket control API, without a browser application.
- Mobile browser: The remote listener is a control API, without a phone-browser work UI.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
Phone access: Phone browser: No
Phone access: Phone browser
Typed value: No.
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Mobile apps: There is no native Ralph TUI iOS/Android application in the scoped client code.
- Browser: The product exposes terminal clients and a WebSocket control API, without a browser application.
- Mobile browser: The remote listener is a control API, without a phone-browser work UI.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
CLI: Yes
CLI
Typed value: Yes.
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Cli: ralph-tui CLI and interactive terminal UI.
- Task actions: Remote TUI supports viewing tasks and pause/resume/cancel/start.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Coding agents: Claude Code: Yes cross-checked
Coding agents: Claude Code
Typed value: Yes.
Release v0.12.0 names Claude Code, OpenCode, Factory Droid, Cursor CLI, Gemini CLI, Codex and Kiro CLI, plus a built-in Pi Coding Agent plugin; agent and model can be overridden. Amp appears only in later main-branch docs. The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
- Model providers: Model identifier passed through selected agent.
- Custom adapters: Configurable custom executable/wrapper.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
- Antigravity harness support remains unknown; installing skills into an editor is not running its coding agent.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Coding agents: Codex: Yes cross-checked
Coding agents: Codex
Typed value: Yes.
Release v0.12.0 names Claude Code, OpenCode, Factory Droid, Cursor CLI, Gemini CLI, Codex and Kiro CLI, plus a built-in Pi Coding Agent plugin; agent and model can be overridden. Amp appears only in later main-branch docs. The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
- Model providers: Model identifier passed through selected agent.
- Custom adapters: Configurable custom executable/wrapper.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
- Antigravity harness support remains unknown; installing skills into an editor is not running its coding agent.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Coding agents: Cursor Agent: Yes cross-checked
Coding agents: Cursor Agent
Typed value: Yes.
Release v0.12.0 names Claude Code, OpenCode, Factory Droid, Cursor CLI, Gemini CLI, Codex and Kiro CLI, plus a built-in Pi Coding Agent plugin; agent and model can be overridden. Amp appears only in later main-branch docs. The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
- Model providers: Model identifier passed through selected agent.
- Custom adapters: Configurable custom executable/wrapper.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
- Antigravity harness support remains unknown; installing skills into an editor is not running its coding agent.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Coding agents: Gemini CLI: Yes cross-checked
Coding agents: Gemini CLI
Typed value: Yes.
Release v0.12.0 names Claude Code, OpenCode, Factory Droid, Cursor CLI, Gemini CLI, Codex and Kiro CLI, plus a built-in Pi Coding Agent plugin; agent and model can be overridden. Amp appears only in later main-branch docs. The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
- Model providers: Model identifier passed through selected agent.
- Custom adapters: Configurable custom executable/wrapper.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
- Antigravity harness support remains unknown; installing skills into an editor is not running its coding agent.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Coding agents: OpenCode: Yes cross-checked
Coding agents: OpenCode
Typed value: Yes.
Release v0.12.0 names Claude Code, OpenCode, Factory Droid, Cursor CLI, Gemini CLI, Codex and Kiro CLI, plus a built-in Pi Coding Agent plugin; agent and model can be overridden. Amp appears only in later main-branch docs. The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
- Model providers: Model identifier passed through selected agent.
- Custom adapters: Configurable custom executable/wrapper.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
- Antigravity harness support remains unknown; installing skills into an editor is not running its coding agent.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Coding agents: Pi: Yes cross-checked
Coding agents: Pi
Typed value: Yes.
Release v0.12.0 names Claude Code, OpenCode, Factory Droid, Cursor CLI, Gemini CLI, Codex and Kiro CLI, plus a built-in Pi Coding Agent plugin; agent and model can be overridden. Amp appears only in later main-branch docs. The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
- Harnesses: Seven named coding CLIs in the v0.12.0 README plus a built-in Pi Coding Agent plugin in the same release. The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
- Antigravity harness support remains unknown; installing skills into an editor is not running its coding agent.
-
Ralph TUI Pi agent plugin (v0.12.0)
Finding: Release v0.12.0 bundles a built-in Pi Coding Agent plugin that launches the installed pi command as the task agent. A Grok plugin exists only on main after that release.
Use my subscription: Yes cross-checked
Use my subscription
Typed value: Yes.
Ralph TUI launches the installed agent CLI and by default strips *_API_KEY variables from its environment so the CLI uses its own sign-in, such as a Claude Max subscription. An API key can be passed through deliberately.
- Agent subscriptions: Documented default: API keys are excluded from agent subprocesses so the agent CLI uses its subscription login, with Claude Max as the named example.
- Api accounts: envPassthrough lets a chosen API key reach the agent for API-based auth.
- Per-provider plan terms were not checked. No installed-product test was run.
-
Ralph TUI docs: Environment Variable Exclusion (v0.12.0)
Finding: Ralph TUI blocks *_API_KEY variables from agent subprocesses by default so the agent CLI uses its own sign-in, such as a Claude Max subscription, instead of an API key; envPassthrough allows a key through when wanted.
Where agents run: On your computer: Yes cross-checked
Where agents run: On your computer
Typed value: Yes.
Run engine on the local machine or a user-controlled VPS/CI host and control it from another TUI.
- Local: Local terminal execution.
- Byo remote: Documented remote instances on VPS/CI/dev boxes.
- Self hosted: User operates listener and engine.
- Client disconnect: Remote engine/listener runs on server; control reconnects after interruptions.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Where agents run: On your own server: Yes cross-checked
Where agents run: On your own server
Typed value: Yes.
Run engine on the local machine or a user-controlled VPS/CI host and control it from another TUI.
- Local: Local terminal execution.
- Byo remote: Documented remote instances on VPS/CI/dev boxes.
- Self hosted: User operates listener and engine.
- Client disconnect: Remote engine/listener runs on server; control reconnects after interruptions.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Where agents run: In the vendor's cloud: No
Where agents run: In the vendor's cloud
Typed value: No.
Run engine on the local machine or a user-controlled VPS/CI host and control it from another TUI.
- Local: Local terminal execution.
- Byo remote: Documented remote instances on VPS/CI/dev boxes.
- Self hosted: User operates listener and engine.
- Client disconnect: Remote engine/listener runs on server; control reconnects after interruptions.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Workspace isolation: Git worktree: Yes
Workspace isolation: Git worktree
Typed value: Yes.
Ralph TUI uses Git worktrees for parallel file separation and optional bubblewrap/macOS sandbox-exec process policies. It provides no built-in Docker/container or VM mode.
- Worktree: One Git worktree per worker.
- Os sandbox: Bubblewrap Linux or sandbox-exec macOS; Windows WSL2 only.
- Filesystem: Writable and read-only allowlists.
- Network: --no-network blocks all network including model APIs.
- Security boundary: Not a VM boundary; no resource limits.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: checkout, branch, process, secrets, code_separation.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
Workspace isolation: Container: No
Workspace isolation: Container
Typed value: No.
Ralph TUI uses Git worktrees for parallel file separation and optional bubblewrap/macOS sandbox-exec process policies. It provides no built-in Docker/container or VM mode.
- Container: The complete mode union has native bubblewrap/sandbox-exec/off policies, without Docker or OCI container provisioning.
- Vm: The complete mode union and wrapper implementation provision no VM.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: checkout, branch, process, secrets, code_separation.
-
Ralph TUI: src/config/types.ts
Finding: The complete sandbox mode union is auto, bwrap, sandbox-exec and off. The implementation wraps native commands with Linux bubblewrap or macOS sandbox-exec. It does not provision Docker/OCI containers or virtual machines; Git worktrees are separate.
Workspace isolation: VM or microVM: No
Workspace isolation: VM or microVM
Typed value: No.
Ralph TUI uses Git worktrees for parallel file separation and optional bubblewrap/macOS sandbox-exec process policies. It provides no built-in Docker/container or VM mode.
- Container: The complete mode union has native bubblewrap/sandbox-exec/off policies, without Docker or OCI container provisioning.
- Vm: The complete mode union and wrapper implementation provision no VM.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: checkout, branch, process, secrets, code_separation.
-
Ralph TUI: src/config/types.ts
Finding: The complete sandbox mode union is auto, bwrap, sandbox-exec and off. The implementation wraps native commands with Linux bubblewrap or macOS sandbox-exec. It does not provision Docker/OCI containers or virtual machines; Git worktrees are separate.
Parallel runs: Yes cross-checked
Parallel runs
Typed value: Yes.
Optional parallel worker pool executes independent tasks simultaneously in worktrees.
- Concurrent tasks: Independent graph tasks run in parallel.
- Quotas: --parallel 4 requests four workers; provider quotas separate.
- Status visibility: TUI exposes worker/merge details and source scope.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
Coordinator: No
Coordinator
Typed value: No.
Ralph TUI has a deterministic parallel worker engine and provider-subagent tracing, without a product-supplied agent coordinator role that delegates and inspects workers.
- Roles: The product worker engine is program logic; the scoped source ships no supervising agent role with worker-delegation tools.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: messages, results_return, cross_harness, orchestration_owner, activation, authentication, retained_results, followup, invocation.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
Native issue tracking: Yes cross-checked
Native issue tracking
Typed value: Yes.
The built-in JSON tracker stores tasks as user stories in a local prd.json with priority, acceptance criteria and a completion state that Ralph TUI updates. Beads, Jira and Linear trackers are optional alternatives.
- Task session separation: Tasks live in prd.json independently of any agent session.
- Metadata: Each task has id, title, description, acceptance criteria and priority.
- Durable states: The passes flag persists in prd.json and is set true as tasks complete.
- External tracker requirement: No external tracker is required; the JSON tracker needs only a local file.
- Discussion, labels, assignment, search and views were not established for the JSON tracker. No installed-product test was run.
-
Ralph TUI docs: JSON Tracker (v0.12.0)
Finding: The built-in JSON tracker keeps tasks as user stories in a local prd.json with id, title, description, acceptance criteria, priority and a passes state; Ralph TUI marks passes true as tasks complete. External trackers (Beads, Jira, Linear) are optional alternatives.
Unattended triggers: Yes
Unattended triggers
Typed value: Yes.
An external program can start work through the documented remote WebSocket orchestrate:start API. The server owns execution independently of a human TUI action for each run.
- Events: The remote API accepts programmatic orchestrate:start requests and starts the owned parallel engine.
- Scheduler location: Requests execute on the user-operated remote listener host; no cron scheduler is established.
- Client independence: Execution is server-owned and event subscribers are separate clients.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: schedules, recurring_tasks, completion_conditions, overlaps, retries.
-
Ralph TUI: website/content/docs/cli/remote.mdx
Finding: The documented remote WebSocket API can programmatically start parallel execution with orchestrate:start, then pause/resume/stop or query it. The product-owned server executes on the remote host and streams events to subscribers. This permits an external program to start work without a human operating the TUI for each run; it does not establish a cron scheduler or distinct human collaboration identities.
Team roles: No
Team roles
Typed value: No.
Ralph TUI uses remote server/connection tokens without named human workspace roles or enterprise SSO.
- Organization roles: No human organization/workspace role-permission model is implemented in the scoped client/server source.
- Sso: The remote listener uses token authentication rather than enterprise SAML/OIDC SSO.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: project_roles, task_roles, membership, provisioning, unauthorized_access, revocation, plan_scope.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
SSO: No
SSO
Typed value: No.
Ralph TUI uses remote server/connection tokens without named human workspace roles or enterprise SSO.
- Organization roles: No human organization/workspace role-permission model is implemented in the scoped client/server source.
- Sso: The remote listener uses token authentication rather than enterprise SAML/OIDC SSO.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: project_roles, task_roles, membership, provisioning, unauthorized_access, revocation, plan_scope.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
Pull request workflow: No
Pull request workflow
Typed value: No.
Ralph TUI integrates worktree merging, conflict handling and optional commits, then prints a manual gh pr create command for the user.
- Pull requests: The run summary prints manual GitHub CLI guidance rather than a product PR creation/management action.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: branches, comments, ci_response, rebase, product_state.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
Review interface: No
Review interface
Typed value: No.
The TUI displays execution output and task state without proposed-code diff review. Configuration preview output is a separate administration feature.
- Change understanding: The inspected TUI components have no proposed-code diff review renderer; internal merge diffs and config preview are separate.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: pending_reviews, evidence_navigation, comments, request_changes, shipping_guidance, effort, missed_evidence, keyboard, narrow_screen, context_switching, roles, retained_context.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
Product license: Apache-2.0: No. MIT cross-checked
Product license: Apache-2.0
Typed value: No. MIT.
Ralph TUI application source uses MIT license.
- Client license: MIT.
- Source visibility: Public application source.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
Product license: MIT: Yes. MIT cross-checked
Product license: MIT
Typed value: Yes. MIT.
Ralph TUI application source uses MIT license.
- Client license: MIT.
- Source visibility: Public application source.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
Product license: GPL-3.0: No. MIT cross-checked
Product license: GPL-3.0
Typed value: No. MIT.
Ralph TUI application source uses MIT license.
- Client license: MIT.
- Source visibility: Public application source.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
Product license: GPL-3.0-or-later: No. MIT cross-checked
Product license: GPL-3.0-or-later
Typed value: No. MIT.
Ralph TUI application source uses MIT license.
- Client license: MIT.
- Source visibility: Public application source.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
Product license: AGPL-3.0: No. MIT cross-checked
Product license: AGPL-3.0
Typed value: No. MIT.
Ralph TUI application source uses MIT license.
- Client license: MIT.
- Source visibility: Public application source.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
Product license: AGPL-3.0-or-later: No. MIT cross-checked
Product license: AGPL-3.0-or-later
Typed value: No. MIT.
Ralph TUI application source uses MIT license.
- Client license: MIT.
- Source visibility: Public application source.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
Product license: Elastic License 2.0: No. MIT cross-checked
Product license: Elastic License 2.0
Typed value: No. MIT.
Ralph TUI application source uses MIT license.
- Client license: MIT.
- Source visibility: Public application source.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
Product license: Custom terms: No. MIT cross-checked
Product license: Custom terms
Typed value: No. MIT.
Ralph TUI application source uses MIT license.
- Client license: MIT.
- Source visibility: Public application source.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
Product license: Proprietary: No. MIT cross-checked
Product license: Proprietary
Typed value: No. MIT.
Ralph TUI application source uses MIT license.
- Client license: MIT.
- Source visibility: Public application source.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
Self-hosting: Yes cross-checked
Self-hosting
Typed value: Yes.
Core CLI/TUI engine and authenticated listener run on user-controlled machines.
- Self hosting: Install on VPS, CI host or development machine.
- Formats: TOML config, JSON PRD, Markdown progress and iteration logs.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Telemetry opt-out: No
Telemetry opt-out
Typed value: No.
Ralph TUI parses agent usage output locally and writes execution state/logs. The scoped configuration and runtime implement no ADE telemetry opt-out switch; provider telemetry remains separate.
- Telemetry disclosure: Usage counters are parsed from the coding-agent output and used by the local TUI.
- Telemetry controls: No product telemetry opt-out control is implemented in the inspected CLI, config or runtime.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: code_data, prompts, credentials, logs, artifacts, retention, deletion, training, residency, subprocessors, telemetry_defaults, consent_prompt, provider_separation.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
Free access: Yes cross-checked
Free access
Typed value: Yes.
Ralph TUI is free MIT software installed from npm; the website and README show no paid plan. Agent and model costs are paid to their providers.
- Ongoing free: The CLI/TUI is a free MIT package installed from npm.
- Self host cost: Runs on the user's own machine or server with no ADE fee shown; agent provider charges stay separate.
- Paid only: Not paid-only: the package installs free from npm.
- Agent free tiers and caps were not checked. No installed-product test was run.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: Ralph TUI installs as a public package from npm with bun install -g ralph-tui; no account or price is required.
Starting price: Free: Yes. Free cross-checked
Starting price: Free
Typed value: Yes. Free.
Ralph TUI is free MIT software installed from npm; the website and README show no paid plan. Agent and model costs are paid to their providers.
- Ongoing free: The CLI/TUI is a free MIT package installed from npm.
- Self host cost: Runs on the user's own machine or server with no ADE fee shown; agent provider charges stay separate.
- Paid only: Not paid-only: the package installs free from npm.
- Agent free tiers and caps were not checked. No installed-product test was run.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: Ralph TUI installs as a public package from npm with bun install -g ralph-tui; no account or price is required.
Starting price: Up to $20/month: No. Free cross-checked
Starting price: Up to $20/month
Typed value: No. Free.
Ralph TUI is free MIT software installed from npm; the website and README show no paid plan. Agent and model costs are paid to their providers.
- Ongoing free: The CLI/TUI is a free MIT package installed from npm.
- Self host cost: Runs on the user's own machine or server with no ADE fee shown; agent provider charges stay separate.
- Paid only: Not paid-only: the package installs free from npm.
- Agent free tiers and caps were not checked. No installed-product test was run.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: Ralph TUI installs as a public package from npm with bun install -g ralph-tui; no account or price is required.
Starting price: $21–40/month: No. Free cross-checked
Starting price: $21–40/month
Typed value: No. Free.
Ralph TUI is free MIT software installed from npm; the website and README show no paid plan. Agent and model costs are paid to their providers.
- Ongoing free: The CLI/TUI is a free MIT package installed from npm.
- Self host cost: Runs on the user's own machine or server with no ADE fee shown; agent provider charges stay separate.
- Paid only: Not paid-only: the package installs free from npm.
- Agent free tiers and caps were not checked. No installed-product test was run.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: Ralph TUI installs as a public package from npm with bun install -g ralph-tui; no account or price is required.
Starting price: $41–100/month: No. Free cross-checked
Starting price: $41–100/month
Typed value: No. Free.
Ralph TUI is free MIT software installed from npm; the website and README show no paid plan. Agent and model costs are paid to their providers.
- Ongoing free: The CLI/TUI is a free MIT package installed from npm.
- Self host cost: Runs on the user's own machine or server with no ADE fee shown; agent provider charges stay separate.
- Paid only: Not paid-only: the package installs free from npm.
- Agent free tiers and caps were not checked. No installed-product test was run.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: Ralph TUI installs as a public package from npm with bun install -g ralph-tui; no account or price is required.
Starting price: Over $100/month: No. Free cross-checked
Starting price: Over $100/month
Typed value: No. Free.
Ralph TUI is free MIT software installed from npm; the website and README show no paid plan. Agent and model costs are paid to their providers.
- Ongoing free: The CLI/TUI is a free MIT package installed from npm.
- Self host cost: Runs on the user's own machine or server with no ADE fee shown; agent provider charges stay separate.
- Paid only: Not paid-only: the package installs free from npm.
- Agent free tiers and caps were not checked. No installed-product test was run.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: Ralph TUI installs as a public package from npm with bun install -g ralph-tui; no account or price is required.
Starting price: Contact for pricing: No. Free cross-checked
Starting price: Contact for pricing
Typed value: No. Free.
Ralph TUI is free MIT software installed from npm; the website and README show no paid plan. Agent and model costs are paid to their providers.
- Ongoing free: The CLI/TUI is a free MIT package installed from npm.
- Self host cost: Runs on the user's own machine or server with no ADE fee shown; agent provider charges stay separate.
- Paid only: Not paid-only: the package installs free from npm.
- Agent free tiers and caps were not checked. No installed-product test was run.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: Ralph TUI installs as a public package from npm with bun install -g ralph-tui; no account or price is required.
Pricing model: Free, no paid plan: Yes
Pricing model: Free, no paid plan
Typed value: Yes.
Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
- Model cost: The Pi plugin requires the user Anthropic, OpenAI or Google provider key; those accounts pay for model use.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: currency, interval, seats, seat_minimum, workspace_fees, included_usage, credits, trial, overages, taxes, compute_cost.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: The MIT License lets anyone obtain, use, copy and modify the software free of charge.
Pricing model: Free plan: No
Pricing model: Free plan
Typed value: No.
Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
- Model cost: The Pi plugin requires the user Anthropic, OpenAI or Google provider key; those accounts pay for model use.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: currency, interval, seats, seat_minimum, workspace_fees, included_usage, credits, trial, overages, taxes, compute_cost.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: The MIT License lets anyone obtain, use, copy and modify the software free of charge.
Pricing model: Subscription: No
Pricing model: Subscription
Typed value: No.
Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
- Model cost: The Pi plugin requires the user Anthropic, OpenAI or Google provider key; those accounts pay for model use.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: currency, interval, seats, seat_minimum, workspace_fees, included_usage, credits, trial, overages, taxes, compute_cost.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: The MIT License lets anyone obtain, use, copy and modify the software free of charge.
Pricing model: Per seat: No
Pricing model: Per seat
Typed value: No.
Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
- Model cost: The Pi plugin requires the user Anthropic, OpenAI or Google provider key; those accounts pay for model use.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: currency, interval, seats, seat_minimum, workspace_fees, included_usage, credits, trial, overages, taxes, compute_cost.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: The MIT License lets anyone obtain, use, copy and modify the software free of charge.
Pricing model: Usage-based: No
Pricing model: Usage-based
Typed value: No.
Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
- Model cost: The Pi plugin requires the user Anthropic, OpenAI or Google provider key; those accounts pay for model use.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: currency, interval, seats, seat_minimum, workspace_fees, included_usage, credits, trial, overages, taxes, compute_cost.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: The MIT License lets anyone obtain, use, copy and modify the software free of charge.
Pricing model: Per machine or compute: No
Pricing model: Per machine or compute
Typed value: No.
Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
- Model cost: The Pi plugin requires the user Anthropic, OpenAI or Google provider key; those accounts pay for model use.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: currency, interval, seats, seat_minimum, workspace_fees, included_usage, credits, trial, overages, taxes, compute_cost.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: The MIT License lets anyone obtain, use, copy and modify the software free of charge.
Pricing model: Bring your own key: Yes
Pricing model: Bring your own key
Typed value: Yes.
Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
- Model cost: The Pi plugin requires the user Anthropic, OpenAI or Google provider key; those accounts pay for model use.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: currency, interval, seats, seat_minimum, workspace_fees, included_usage, credits, trial, overages, taxes, compute_cost.
-
Ralph TUI: website/content/docs/plugins/agents/pi.mdx
Finding: The first-party Pi plugin guide explicitly requires a user API key for at least one provider and names ANTHROPIC_API_KEY, OPENAI_API_KEY and GEMINI_API_KEY. Those provider accounts pay for model use; Ralph TUI remains free MIT software.
Pricing model: One-time purchase: No
Pricing model: One-time purchase
Typed value: No.
Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
- Model cost: The Pi plugin requires the user Anthropic, OpenAI or Google provider key; those accounts pay for model use.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: currency, interval, seats, seat_minimum, workspace_fees, included_usage, credits, trial, overages, taxes, compute_cost.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: The MIT License lets anyone obtain, use, copy and modify the software free of charge.
Keeps going with laptop closed: Yes cross-checked
Keeps going with laptop closed
Typed value: Yes.
Run engine on the local machine or a user-controlled VPS/CI host and control it from another TUI.
- Local: Local terminal execution.
- Byo remote: Documented remote instances on VPS/CI/dev boxes.
- Self hosted: User operates listener and engine.
- Client disconnect: Remote engine/listener runs on server; control reconnects after interruptions.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
Open-source license: Yes. MIT cross-checked
Open-source license
Typed value: Yes. MIT.
Ralph TUI application source uses MIT license.
- Client license: MIT.
- Source visibility: Public application source.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
Any desktop app: Yes
Any desktop app
Typed value: Yes.
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Desktop os: The installation guide provides macOS/Linux and native Windows PowerShell prerequisites and global installation.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Ralph TUI: website/content/docs/getting-started/installation.mdx
Finding: The first-party installation page provides macOS/Linux and native Windows PowerShell Bun setup followed by global ralph-tui installation. The primary product is a terminal CLI/TUI, not a desktop GUI.
Any phone client: No
Any phone client
Typed value: No.
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Mobile apps: There is no native Ralph TUI iOS/Android application in the scoped client code.
- Browser: The product exposes terminal clients and a WebSocket control API, without a browser application.
- Mobile browser: The remote listener is a control API, without a phone-browser work UI.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
All public findings
Findings may summarize a source. They are not presented as verbatim statements.
ralph-tui:ACC-01: Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
ralph-tui:ACC-01
Ralph TUI is a native terminal CLI/TUI with documented macOS, Linux and Windows installation and remote terminal control, without a product browser or native phone app.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: editor_extensions, notifications, cross_device_continuity.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
Ralph TUI: website/content/docs/getting-started/installation.mdx
Finding: The first-party installation page provides macOS/Linux and native Windows PowerShell Bun setup followed by global ralph-tui installation. The primary product is a terminal CLI/TUI, not a desktop GUI.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
ralph-tui:ACC-02: Release v0.12.0 names Claude Code, OpenCode, Factory Droid, Cursor CLI, Gemini CLI, Codex and Kiro CLI, plus a built-in Pi Coding Agent plugin; agent and model can be overridden. Amp appears only in later main-branch docs. The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration. cross-checked
ralph-tui:ACC-02
Release v0.12.0 names Claude Code, OpenCode, Factory Droid, Cursor CLI, Gemini CLI, Codex and Kiro CLI, plus a built-in Pi Coding Agent plugin; agent and model can be overridden. Amp appears only in later main-branch docs. The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
- Antigravity harness support remains unknown; installing skills into an editor is not running its coding agent.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
website/content/docs/configuration/options.mdx
Finding: Configurable agent/model/custom command, iteration caps, progress/log paths, optional autoCommit and subagent tracing levels.
-
Ralph TUI Pi agent plugin (v0.12.0)
Finding: Release v0.12.0 bundles a built-in Pi Coding Agent plugin that launches the installed pi command as the task agent. A Grok plugin exists only on main after that release.
-
Finding: At release v0.12.0 the README names Claude Code, OpenCode, Factory Droid, Cursor CLI, Gemini CLI, Codex and Kiro CLI as supported coding assistants.
-
Ralph TUI Antigravity support at v0.12.0
Finding: The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
-
Ralph TUI Antigravity support at v0.12.0
Finding: The v0.12.0 Antigravity reference is a skills-installer test fixture. The released agent list and plugins do not establish an Antigravity launch integration.
ralph-tui:ACC-03: Ralph TUI launches the installed agent CLI and by default strips *_API_KEY variables from its environment so the CLI uses its own sign-in, such as a Claude Max subscription. An API key can be passed through deliberately. cross-checked
ralph-tui:ACC-03
Ralph TUI launches the installed agent CLI and by default strips *_API_KEY variables from its environment so the CLI uses its own sign-in, such as a Claude Max subscription. An API key can be passed through deliberately.
- Per-provider plan terms were not checked. No installed-product test was run.
-
Ralph TUI docs: Environment Variable Exclusion (v0.12.0)
Finding: Ralph TUI blocks *_API_KEY variables from agent subprocesses by default so the agent CLI uses its own sign-in, such as a Claude Max subscription, instead of an API key; envPassthrough allows a key through when wanted.
ralph-tui:ACC-04: Install with Bun, run setup, create a PRD and start from prd.json or a Beads epic. cross-checked
ralph-tui:ACC-04
Install with Bun, run setup, create a PRD and start from prd.json or a Beads epic.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
ralph-tui:ACC-05: Product-owned CLI supports run, resume, status, logs, configuration, tracker conversion and remote management. cross-checked
ralph-tui:ACC-05
Product-owned CLI supports run, resume, status, logs, configuration, tracker conversion and remote management.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
ralph-tui:ACC-06: Agent/tracker plugins and installable PRD/task skills extend the orchestrator. cross-checked
ralph-tui:ACC-06
Agent/tracker plugins and installable PRD/task skills extend the orchestrator.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
website/content/docs/configuration/options.mdx
Finding: Configurable agent/model/custom command, iteration caps, progress/log paths, optional autoCommit and subagent tracing levels.
ralph-tui:ACC-07: Terminal interface uses OpenTUI/React and Bun; no Electron desktop client was established. cross-checked
ralph-tui:ACC-07
Terminal interface uses OpenTUI/React and Bun; no Electron desktop client was established.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
ralph-tui:PLAN-01: Intake supports local prd.json and external Beads, plus conversion into tracker formats. cross-checked
ralph-tui:PLAN-01
Intake supports local prd.json and external Beads, plus conversion into tracker formats.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
ralph-tui:PLAN-02: Interactive AI-assisted PRD creation and progress files retain plan/context between iterations. cross-checked
ralph-tui:PLAN-02
Interactive AI-assisted PRD creation and progress files retain plan/context between iterations.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
ralph-tui:PLAN-03: One graph spans selected epics and blocks dependent work on incomplete external dependencies. cross-checked
ralph-tui:PLAN-03
One graph spans selected epics and blocks dependent work on incomplete external dependencies.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
ralph-tui:PLAN-04: Product progress file retains cross-iteration context and custom skills/templates can be selected. cross-checked
ralph-tui:PLAN-04
Product progress file retains cross-iteration context and custom skills/templates can be selected.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
website/content/docs/configuration/options.mdx
Finding: Configurable agent/model/custom command, iteration caps, progress/log paths, optional autoCommit and subagent tracing levels.
ralph-tui:PLAN-06: The built-in JSON tracker stores tasks as user stories in a local prd.json with priority, acceptance criteria and a completion state that Ralph TUI updates. Beads, Jira and Linear trackers are optional alternatives. cross-checked
ralph-tui:PLAN-06
The built-in JSON tracker stores tasks as user stories in a local prd.json with priority, acceptance criteria and a completion state that Ralph TUI updates. Beads, Jira and Linear trackers are optional alternatives.
- Discussion, labels, assignment, search and views were not established for the JSON tracker. No installed-product test was run.
-
Ralph TUI docs: JSON Tracker (v0.12.0)
Finding: The built-in JSON tracker keeps tasks as user stories in a local prd.json with id, title, description, acceptance criteria, priority and a passes state; Ralph TUI marks passes true as tasks complete. External trackers (Beads, Jira, Linear) are optional alternatives.
ralph-tui:COORD-01: Optional parallel worker pool executes independent tasks simultaneously in worktrees. cross-checked
ralph-tui:COORD-01
Optional parallel worker pool executes independent tasks simultaneously in worktrees.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
ralph-tui:COORD-02: Scheduler selects highest-priority ready tasks and dispatches them from a dependency graph. cross-checked
ralph-tui:COORD-02
Scheduler selects highest-priority ready tasks and dispatches them from a dependency graph.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
ralph-tui:COORD-03: Ralph TUI has a deterministic parallel worker engine and provider-subagent tracing, without a product-supplied agent coordinator role that delegates and inspects workers.
ralph-tui:COORD-03
Ralph TUI has a deterministic parallel worker engine and provider-subagent tracing, without a product-supplied agent coordinator role that delegates and inspects workers.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: messages, results_return, cross_harness, orchestration_owner, activation, authentication, retained_results, followup, invocation.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
website/content/docs/configuration/options.mdx
Finding: Configurable agent/model/custom command, iteration caps, progress/log paths, optional autoCommit and subagent tracing levels.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
ralph-tui:COORD-04: An external program can start work through the documented remote WebSocket orchestrate:start API. The server owns execution independently of a human TUI action for each run.
ralph-tui:COORD-04
An external program can start work through the documented remote WebSocket orchestrate:start API. The server owns execution independently of a human TUI action for each run.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: schedules, recurring_tasks, completion_conditions, overlaps, retries.
-
Ralph TUI: website/content/docs/cli/remote.mdx
Finding: The documented remote WebSocket API can programmatically start parallel execution with orchestrate:start, then pause/resume/stop or query it. The product-owned server executes on the remote host and streams events to subscribers. This permits an external program to start work without a human operating the TUI for each run; it does not establish a cron scheduler or distinct human collaboration identities.
ralph-tui:COORD-05: Keyboard and remote actions pause, resume, interrupt or cancel execution, with persisted logs. cross-checked
ralph-tui:COORD-05
Keyboard and remote actions pause, resume, interrupt or cancel execution, with persisted logs.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
ralph-tui:COORD-06: Multi-epic sessions show source scopes and shared workers in a single scheduler UI; no fleet usability fixture was run. cross-checked
ralph-tui:COORD-06
Multi-epic sessions show source scopes and shared workers in a single scheduler UI; no fleet usability fixture was run.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
ralph-tui:COORD-07: Agent/model selection, custom templates and task-tracker plugins configure execution. cross-checked
ralph-tui:COORD-07
Agent/model selection, custom templates and task-tracker plugins configure execution.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
website/content/docs/configuration/options.mdx
Finding: Configurable agent/model/custom command, iteration caps, progress/log paths, optional autoCommit and subagent tracing levels.
ralph-tui:RUN-01: Run engine on the local machine or a user-controlled VPS/CI host and control it from another TUI. cross-checked
ralph-tui:RUN-01
Run engine on the local machine or a user-controlled VPS/CI host and control it from another TUI.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
ralph-tui:RUN-02: Ralph TUI uses Git worktrees for parallel file separation and optional bubblewrap/macOS sandbox-exec process policies. It provides no built-in Docker/container or VM mode.
ralph-tui:RUN-02
Ralph TUI uses Git worktrees for parallel file separation and optional bubblewrap/macOS sandbox-exec process policies. It provides no built-in Docker/container or VM mode.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: checkout, branch, process, secrets, code_separation.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
-
website/content/docs/configuration/sandbox.mdx
Finding: Opt-in Bubblewrap on Linux or sandbox-exec on macOS; Windows needs WSL2. Configurable writable/read-only paths and network disable. No VM-level boundary or CPU/memory/process limits; auto can fall back off.
-
Ralph TUI: src/config/types.ts
Finding: The complete sandbox mode union is auto, bwrap, sandbox-exec and off. The implementation wraps native commands with Linux bubblewrap or macOS sandbox-exec. It does not provision Docker/OCI containers or virtual machines; Git worktrees are separate.
ralph-tui:RUN-03: Sessions can be resumed and remote connections use bounded reconnect retries. Open bug reports mean recovery is not proven here. cross-checked
ralph-tui:RUN-03
Sessions can be resumed and remote connections use bounded reconnect retries. Open bug reports mean recovery is not proven here.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
ralph-tui:RUN-05: Iteration limit bounds loop count; preflight timeout bounds setup checks, not all runtime/tool effects. cross-checked
ralph-tui:RUN-05
Iteration limit bounds loop count; preflight timeout bounds setup checks, not all runtime/tool effects.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
website/content/docs/configuration/options.mdx
Finding: Configurable agent/model/custom command, iteration caps, progress/log paths, optional autoCommit and subagent tracing levels.
ralph-tui:RUN-06: Dashboard displays agent/model, tracker, branch, sandbox and remote status; log files retain iteration output. cross-checked
ralph-tui:RUN-06
Dashboard displays agent/model, tracker, branch, sandbox and remote status; log files retain iteration output.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
website/content/docs/configuration/options.mdx
Finding: Configurable agent/model/custom command, iteration caps, progress/log paths, optional autoCommit and subagent tracing levels.
ralph-tui:RUN-08: Worker pool and dependency grouping bound concurrency; OS sandbox explicitly does not enforce CPU/RAM/process limits. cross-checked
ralph-tui:RUN-08
Worker pool and dependency grouping bound concurrency; OS sandbox explicitly does not enforce CPU/RAM/process limits.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
-
website/content/docs/configuration/sandbox.mdx
Finding: Opt-in Bubblewrap on Linux or sandbox-exec on macOS; Windows needs WSL2. Configurable writable/read-only paths and network disable. No VM-level boundary or CPU/memory/process limits; auto can fall back off.
ralph-tui:TEAM-01: Remote terminal clients can share server state and execution controls through server/connection tokens. This does not establish distinct human accounts or a multi-person prompting workflow.
ralph-tui:TEAM-01
Remote terminal clients can share server state and execution controls through server/connection tokens. This does not establish distinct human accounts or a multi-person prompting workflow.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: discussion, comments, presence, ordering, conflicts, attribution, permissions, plan_restrictions, owner_online.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
-
Ralph TUI: website/content/docs/cli/remote.mdx
Finding: The documented remote WebSocket API can programmatically start parallel execution with orchestrate:start, then pause/resume/stop or query it. The product-owned server executes on the remote host and streams events to subscribers. This permits an external program to start work without a human operating the TUI for each run; it does not establish a cron scheduler or distinct human collaboration identities.
ralph-tui:TEAM-02: Ralph TUI uses remote server/connection tokens without named human workspace roles or enterprise SSO.
ralph-tui:TEAM-02
Ralph TUI uses remote server/connection tokens without named human workspace roles or enterprise SSO.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: project_roles, task_roles, membership, provisioning, unauthorized_access, revocation, plan_scope.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
ralph-tui:TEAM-03: Remote tokens and opt-in sandbox paths/network constrain access; no enterprise approval workflow established. cross-checked
ralph-tui:TEAM-03
Remote tokens and opt-in sandbox paths/network constrain access; no enterprise approval workflow established.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
website/content/docs/configuration/sandbox.mdx
Finding: Opt-in Bubblewrap on Linux or sandbox-exec on macOS; Windows needs WSL2. Configurable writable/read-only paths and network disable. No VM-level boundary or CPU/memory/process limits; auto can fall back off.
ralph-tui:TEAM-04: Remote actions are logged to a local audit.log; immutable retention is unestablished. cross-checked
ralph-tui:TEAM-04
Remote actions are logged to a local audit.log; immutable retention is unestablished.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
ralph-tui:TEAM-05: The TUI displays execution output and task state without proposed-code diff review. Configuration preview output is a separate administration feature.
ralph-tui:TEAM-05
The TUI displays execution output and task state without proposed-code diff review. Configuration preview output is a separate administration feature.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: pending_reviews, evidence_navigation, comments, request_changes, shipping_guidance, effort, missed_evidence, keyboard, narrow_screen, context_switching, roles, retained_context.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
ralph-tui:SHIP-01: PRD/progress and per-iteration logs persist on disk. cross-checked
ralph-tui:SHIP-01
PRD/progress and per-iteration logs persist on disk.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
website/content/docs/configuration/options.mdx
Finding: Configurable agent/model/custom command, iteration caps, progress/log paths, optional autoCommit and subagent tracing levels.
ralph-tui:SHIP-03: Ralph TUI integrates worktree merging, conflict handling and optional commits, then prints a manual gh pr create command for the user.
ralph-tui:SHIP-03
Ralph TUI integrates worktree merging, conflict handling and optional commits, then prints a manual gh pr create command for the user.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: branches, comments, ci_response, rebase, product_state.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
-
website/content/docs/configuration/options.mdx
Finding: Configurable agent/model/custom command, iteration caps, progress/log paths, optional autoCommit and subagent tracing levels.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
ralph-tui:SHIP-04: Orchestrator merge queue brings completed worker branches back sequentially. Exact revision human authority is unestablished. cross-checked
ralph-tui:SHIP-04
Orchestrator merge queue brings completed worker branches back sequentially. Exact revision human authority is unestablished.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
ralph-tui:SHIP-05: Backup tags are created before each merge for rollback safety. cross-checked
ralph-tui:SHIP-05
Backup tags are created before each merge for rollback safety.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
website/content/docs/parallel/overview.mdx
Finding: Optional parallel workers use task worktrees, a dependency graph, one session branch and serial merge queue, with conflict resolution and backup tags. Default is sequential; external incomplete dependencies block tasks.
ralph-tui:OWN-01: Ralph TUI application source uses MIT license. cross-checked
ralph-tui:OWN-01
Ralph TUI application source uses MIT license.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Application source uses MIT license.
ralph-tui:OWN-02: Core CLI/TUI engine and authenticated listener run on user-controlled machines. cross-checked
ralph-tui:OWN-02
Core CLI/TUI engine and authenticated listener run on user-controlled machines.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
ralph-tui:OWN-03: Ralph TUI parses agent usage output locally and writes execution state/logs. The scoped configuration and runtime implement no ADE telemetry opt-out switch; provider telemetry remains separate.
ralph-tui:OWN-03
Ralph TUI parses agent usage output locally and writes execution state/logs. The scoped configuration and runtime implement no ADE telemetry opt-out switch; provider telemetry remains separate.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: code_data, prompts, credentials, logs, artifacts, retention, deletion, training, residency, subprocessors, telemetry_defaults, consent_prompt, provider_separation.
-
Ralph TUI: src/commands/run.tsx
Finding: Bounded inspection of the product command tree, TUI components, parallel executor and remote listener found no browser or native phone client, proposed-code diff review renderer, human team-role/SSO model or product telemetry opt-out setting. ParallelExecutor is a deterministic dependency/worktree worker engine, not an agent coordinator role with delegation tools. The remote listener uses owner/server and connection tokens. Run guidance prints a manual gh pr create command rather than creating a PR itself.
ralph-tui:OWN-04: Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
ralph-tui:OWN-04
Ralph TUI is free MIT software without a paid plan. Its Pi plugin explicitly accepts user model-provider API keys, and provider/compute charges are separate.
- Documentation and pinned source inspection only; no hands-on scenario was run.
- Other atomic questions remain unestablished: currency, interval, seats, seat_minimum, workspace_fees, included_usage, credits, trial, overages, taxes, compute_cost.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: The MIT License lets anyone obtain, use, copy and modify the software free of charge.
-
Ralph TUI: website/content/docs/plugins/agents/pi.mdx
Finding: The first-party Pi plugin guide explicitly requires a user API key for at least one provider and names ANTHROPIC_API_KEY, OPENAI_API_KEY and GEMINI_API_KEY. Those provider accounts pay for model use; Ralph TUI remains free MIT software.
ralph-tui:OWN-06: Latest listed release v0.12.0 was published May 13, 2026; current default-branch documentation may include later changes. cross-checked
ralph-tui:OWN-06
Latest listed release v0.12.0 was published May 13, 2026; current default-branch documentation may include later changes.
- Only the named facts are documented. Unanswered atomic questions remain unknown; no execution test was performed.
-
Finding: Terminal orchestrator runs named coding CLIs against prd.json/Beads tasks. It offers PRD generation, persisted sessions, progress memory, subagent tracing, headless execution and authenticated remote control on VPS/CI hosts.
-
Finding: v0.12.0 was published 2026-05-13. Release notes describe the scoped changes; this is not a hands-on check.
ralph-tui:OWN-08: Ralph TUI is free MIT software installed from npm; the website and README show no paid plan. Agent and model costs are paid to their providers. cross-checked
ralph-tui:OWN-08
Ralph TUI is free MIT software installed from npm; the website and README show no paid plan. Agent and model costs are paid to their providers.
- Agent free tiers and caps were not checked. No installed-product test was run.
-
Finding: Says the tool is open source under the MIT License. The site shows no price or pricing page.
-
Finding: The MIT License lets anyone obtain, use, copy and modify the software free of charge.
-
Finding: Ralph TUI installs as a public package from npm with bun install -g ralph-tui; no account or price is required.