November 2026 status archive, page 25 of 26
Shutdown notices, recent changes and all monthly archives
These events link to today's eligible findings, not historical before-and-after claims.
Cited events
20 ADEs
OpenHandsresearch correctionfreshness recheck39 findings · 14 sources
Findings today
- Canvas provides desktop installation paths, browser UI and documented phone/tablet browser access.
- Canvas manages backend-scoped MCP, skills and plugins, with health and credential checks.
- Canvas documents skills and a persistent workspace/user memory setting that can be enabled or disabled.
- Canvas has reusable profiles, goals and backend-specific configuration.
- Optional containers and worktrees separate execution/files, with documented shared-HOME and shared-workspace limits.
- Goals and experimental critic refinement have explicit iteration limits.
- Canvas displays automation phase/status and exports reported costs with null for unavailable cost.
- Canvas automation history exposes reported per-run LLM cost with explicit missing values.
- Canvas exposes combined commit/uncommitted diffs and experimental critic feedback.
- Canvas self-hosts and provides portable conversation and automation exports.
- Backend settings/secrets and browser-generated exports have explicit boundaries.
- Canvas integrates OpenHands and ACP agents with custom stdio ACP commands.
- Local ACP providers can reuse native logins, while clean containers/cloud require credentials and have provider limits.
- Canvas launches linked child conversations and returns their results to the parent.
- Automation Server owns schedules/events and Canvas exposes lifecycle/run history.
- Canvas supports goal interruption and preserves failed-message retry controls.
- Canvas shows linked children and grouped conversation metadata.
- Canvas reloads backend conversation history and keeps archived conversations without deleting them.
- Backend concurrency caps can skip automation dispatch; resource-aware fairness remains unknown.
- Cloud organization automation conversations support read-only team visibility.
- Canvas exports timestamped conversation/tool history and retains archived messages.
- Canvas renders file artifacts, images and unified code-change views.
- Experimental critic evaluates OpenHands-agent results and can request bounded refinement.
- Canvas Git controls prompt the agent to commit, pull, push and work with PRs.
- Local stack requires Node 24 and uv; Docker and hardened VM routes are separate.
- Canvas uses the Agent Server REST API; SDK ownership is separate.
- Canvas is a React/TypeScript web frontend.
- Issue-to-PR automations read tracker issues; Slack, GitHub and webhook events trigger agent runs.
- The Canvas Task List renders the latest TaskTrackerObservation from the current conversation, with position-derived…
- Parallel conversations can use separate Docker runtimes.
- Canvas switches between local, self-hosted remote and optional cloud backends.
- The user chooses workspace mounts and runtime configuration.
- Organizations document Member/Admin/Owner permissions; paid Teams/Enterprise separately advertise SSO/SAML.
- Self-hosted public mode requires the backend API key.
- The Canvas source distribution is MIT.
- The current public self-hosted offer is free Canvas, Teams at $500/month for ten users plus $50/month additional…
- Teams and Enterprise plans carry paid identity governance; support ranges from community support for Agent Canvas to…
- The vendor advertises free local and self-managed Canvas use.
- Pricing page publishes monthly Teams rate and session caps, with Enterprise by quote.
Sources
Open the OpenHands page
OpenSpecfreshness recheck36 findings · 10 sources
Findings today
- OpenSpec ships a terminal CLI on macOS, Windows and Linux plus assistant workflow skills, without a standalone…
- OpenSpec integrates workflow skills into Claude Code, Codex, Cursor, Gemini CLI, OpenCode, Pi, Grok Build and…
- OpenSpec adds skills and commands to the user's existing coding assistant and runs inside it, so the assistant's own…
- Initialize in a project directory or create a separate planning store.
- Product-owned CLI exposes machine-readable change, spec, readiness and instruction operations.
- OpenSpec delivers configurable workflow skills into supported coding assistants.
- Ideas become named change folders through an agent workflow or CLI.
- A change retains its proposal, requirements, design and implementation checklist as files.
- Schema dependencies order planning artifacts before implementation.
- Project config and shared stores provide persistent instructions and upstream specifications.
- Planning can reference upstream specifications; freshness depends on the user Git checkout.
- Named change records and task checklists can be created and archived independently of a chat session.
- Assistant workflows operate the same named change artifacts users inspect.
- OpenSpec maintains canonical requirements through change deltas and archive/sync workflows.
- OpenSpec tracks plans and task checklists but does not own or run concurrent coding-agent sessions.
- OpenSpec supplies workflow instructions for an existing assistant without a product-owned coordinator or…
- OpenSpec workflows are invoked by a user or an external assistant; the product has no schedule or event runner.
- Apply pauses when tasks are unclear or errors block implementation.
- Profiles, schemas and natural-language workflows customize planning and implementation.
- OpenSpec runs its CLI and planning workflow on the user-controlled machine with an installed assistant.
- OpenSpec edits artifacts in existing project directories; it provisions no Git worktree, container or VM isolation.
- CLI reports planning and checklist status, not verified agent runtime statistics.
- OpenSpec shares planning artifacts through ordinary Git repositories; it has no shared agent prompting session.
- OpenSpec has no workspace identity service, named team roles or enterprise SSO.
- Specification and archived change files remain available through the user repository.
- OpenSpec prints task and artifact progress and supplies verification instructions; it has no code-diff review interface.
- Change folders retain documents and implementation checklists.
- Structural validation and report-only verification are documented; archive offers bypasses.
- OpenSpec leaves code commits and pull requests to the user and coding assistant; its native workflow does not create…
- Bulk archive reviews several changes and syncs deltas before archiving.
- The OpenSpec application repository is MIT licensed.
- The CLI and durable Markdown planning files run on user-controlled machines.
- OpenSpec documents telemetry identifiers and opt-out controls.
- OpenSpec is an MIT open-source npm package with no paid plan or price published; it runs inside your existing AI…
- OpenSpec's current release is v2.1.1 (2026-11-19); multi-repo store workflows remain beta.
- OpenSpec is freely installable MIT software; coding assistant costs are separate.
Sources
Open the OpenSpec page
Orcafreshness recheckresearch correction29 findings · 66 sources
Findings today
- Desktop on macOS, Windows and Linux, plus beta iOS/Android companion.
- Orca launches and tracks existing CLI agents in terminals rather than embedding a harness: many CLI launchers are…
- Orca reuses the installed coding-agent native subscription logins, including Claude Code and Codex.
- Orca exposes a broad orca CLI (worktrees, terminals, files, browser, orchestration, automations, Linear, accounts,…
- Orca keeps lightweight task state: orchestration tasks carry a spec and status, worktrees carry a free-text comment…
- Workspaces can be nested under a parent workspace and orchestration tasks have dependencies and statuses including…
- Agents can read, create and update Linear issues through orca linear and set worktree comments and status, report…
- Orchestration dispatches a task to a specific worker terminal and treats the active dispatch as completion…
- Orca documents its own orchestration skill/CLI and supervised worker delegation, but the required Experimental panel…
- orca automations run a prompt on hourly, daily, weekday, weekly, cron or RRULE schedules against a repo or worktree,…
- Fleet views documented: sidebar worktree cards with N-children chips, an experimental kanban dashboard (Needs You,…
- Creating a worktree and launching an agent is a create-and-launch sequence with no mandatory workflow selection (the…
- Local desktop, user-owned SSH hosts and persistent Remote Orca Servers; experimental Cloud VM recipes use the user…
- No enforced time, turn, token or spend limit was found.
- Orca shows provider usage windows, resets and credit balances, and Stats with estimated cost from a local price…
- Orca is built around one developer machine or server; several paired clients can see the same server-owned workspaces.
- The self-managed Orca runtime uses revocable paired-device grants without named organization/project user roles or…
- A second client paired to a Remote Orca Server reaches the same server-owned workspaces, tabs and agent sessions.
- A done queue exists in the experimental dashboard and Agents feed with response previews, PR status filters and…
- Vendor policy describes local code and provider credentials, pseudonymous opt-out product telemetry and a vendor…
- The Orca application is advertised as free and open source with no price list; the Enterprise page offers rollout…
- No priced scenario was built.
- Latest GitHub release is v1.4.253, published November 7, 2026, following v1.4.252 the same day and v1.4.251 on…
- The desktop app is advertised as free and open source, so BYO-subscription and self-hosted scenarios carry no Orca…
- The site advertises the desktop app as free; no desktop seat, usage-unit or overage rate is established by the cited…
- A background daemon keeps local agent processes running across app quit, update relaunch and Orca app crashes.
- Orca documents native Run/Task/Dispatch records independent of external trackers, but the required Experimental…
- Native orchestration tasks retain a spec and link to execution attempts and decision gates.
- Approval handling is largely delegated to the agent CLIs: by default Orca pre-fills each supported CLI's…
Sources
Open the Orca page
Paseofreshness recheckresearch correction16 findings · 75 sources
Findings today
- Paseo offers desktop, native phone, browser and CLI clients for its daemon; the shared browser app includes compact…
- Current stable and homepage document native Claude Code, Codex, OpenCode, Pi, OMP, Antigravity and Muse plus a…
- Paseo reuses each provider CLI's own login for Claude plan, ChatGPT plan and API-key use, and adds no charge for…
- Paseo provides agent timelines (paseo logs), daemon and desktop logs, a Usage view of subscription quota windows and…
- In self-hosted Paseo Hub, distinct permitted Slack users can send instructions to one daemon agent by enabling Same…
- The Paseo app, daemon, CLI and SDK are free; hosted Hub is Free ($0, 50 agent runs/month, 1 seat) or Pro ($15 per…
- Fixed hosted Hub examples: one Free seat $0 for 50 runs/month; two Pro seats $30/month; five $75/month.
- Current stable is v0.15.3 on November 6, with separate Beta channel and gradual desktop rollout.
- The Paseo app, daemon, CLI and SDK are free and open source with no usage cap documented; hosted Hub offers a…
- Hub pricing lists the unit (seat), included runs and the seat definition; Paseo documents no model or compute charge…
- Paseo v0.15.3 root code is Apache-2.0 with third-party licence exceptions; its desktop manifest also declares…
- Paseo documents a plugin system and passes through the provider CLI's own MCP servers and skills; it does not…
- Paseo models parentage between agents (subagents, detach) and workspaces; epics, blockers and readiness states are…
- Paseo isolates file changes with optional git worktrees per workspace; same-workspace agents share files.
- Paseo persists agent records and timelines and resumes closed agents under the same ID.
- Stable v0.15.3 states no telemetry, tracking or required Paseo account for the local app.
Sources
- Paseo homepage
- Paseo: CHANGELOG.md
- Paseo: public-docs/orchestration-workflows.md
- Paseo: Why Paseo?
- Paseo: Supported providers
- Paseo: Workspaces
- Paseo: Git worktrees
- Paseo: Security
- Paseo: MCP reference
- Paseo: Hub security
- Paseo: Daemons in Hub
- Paseo: GitHub access
- Paseo: Docker
- Paseo: Run parallel tasks
- Paseo: Agent lifecycle
- Paseo: Troubleshooting
- Paseo: Plugins
- Paseo: Orchestration skills
- public-docs/why.md, line 23
- public-docs/why.md, lines 35 and 43
- env examples lines 56, 129 and 134; Multiple profiles line 118
- providers.openai.stt and tts apiKey examples, lines 94 and 98
- GitHub repository metadata
- Paseo homepage
- Paseo Hub plans and pricing
- Paseo GitHub releases
- Paseo: CHANGELOG.md
- Paseo: public-docs/orchestration.md
- Paseo: public-docs/orchestration-workflows.md
- Paseo: Getting started
- Paseo: Why Paseo?
- Paseo: Supported providers
- Paseo: Security
- Paseo: MCP reference
- Paseo: CLI reference
- Paseo: TypeScript SDK
- Paseo: Hub overview
- Paseo: Hosted Hub
- Paseo: Hub security
- Paseo: GitHub access
- Paseo: Hub activity
- Paseo: Hub public API
- Paseo: Self-hosting Hub
- Paseo: Claude Code provider
- Paseo: Codex provider
- Paseo: Agent lifecycle
- Paseo: Related projects
- Paseo: Updates
- Paseo: Troubleshooting
- Paseo: Self-hosting the web UI
- Paseo: Plugins
- Paseo: Custom providers
- public-docs/why.md, line 23
- public-docs/why.md, lines 35 and 43
- Native support (6 entries); ACP catalog (36 entries)
- GitHub API repos/getpaseo/paseo open_issues_count
- env examples lines 56, 129 and 134; Multiple profiles line 118
- providers.openai.stt and tts apiKey examples, lines 94 and 98
- Paseo homepage FAQ and integrations list
- Paseo Hub live page FAQ
- Paseo v0.11.1 Claude Code terminal access
- Paseo v0.15.3
- Paseo licence at v0.15.3
- Paseo desktop manifest at v0.15.3
- Paseo security docs at v0.15.3
- Paseo self-hosting Hub at v0.15.3
- Paseo v0.15.3 release
- Paseo v0.15.3: public-docs/supported-providers.md
- Paseo v0.15.3: public-docs/plugins/reference.md
- Paseo v0.15.3: public-docs/hub/workflows.md
- Paseo v0.15.3: public-docs/hub/security.md
- Paseo v0.15.3: public-docs/docker.md
- Paseo v0.15.3: docs/agent-lifecycle.md
- Paseo v0.15.3: docs/permissions.md
- Paseo Hub - GitHub, Slack, and Discord triggers
Open the Paseo page
Podiumfreshness recheckresearch correction11 findings · 72 sources
Findings today
- Docs state repositories and native harness credentials stay on machines you control and server state lives in a…
- Docs describe a first-run flow (install, podium, register a Git repository, confirm a detected logged-in harness)…
- Podium gives agents its own issue tools over CLI and MCP, and agent profiles import and export Claude Code agent files.
- v0.3.0 routes Podium issue instructions and relay environment through terminal and engine session launch.
- Not established.
- The spec CLI and service (living specs under pspec with stable SP IDs) are in v0.3.0.
- The server owns a scheduler for cron and one-time automations that spawn agent sessions, skipping overlaps and not…
- Issue actions, approval steps, comments and mail are recorded server-side.
- Artifact commands store immutable server-side copies of issue-linked files, including images, video, HTML and…
- The shipping service records revision-specific approval, configured validation and delivery receipts.
- Agent-session management requests go through an operator approval broker with approve and deny; the privacy docs say…
Sources
Open the Podium page
Proliferatefreshness recheckresearch correction38 findings · 15 sources
Findings today
- Repository documents Mac desktop and self-hosted browser app; native mobile release not established.
- The inspected Proliferate runtime supports the fixed native harness set Claude Code, Codex, Cursor, OpenCode and Grok.
- Detected native login, own API/provider config and managed gateway are distinct authentication routes.
- Source build and full self-host deployment recipes remain public despite missing website docs.
- Scoped runtime HTTP/MCP APIs exist; target general public agent API is excluded.
- Shared MCP, skills and custom tools extend native agent sessions.
- Desktop uses Tauri/Rust with compiled Vite web app on server.
- Prompted tasks execute in durable workspaces; source records do not establish tracker sync.
- Saved workflow definition carries prompts, inputs and materialized context docs.
- Workflow is a linear chain; child-agent hierarchy is bounded.
- Agents share configured skills/tools and retain durable conversation state.
- Proliferate persists agent sessions, local/worktree workspace lifecycle and workflow runs, without an accessible…
- Multiple task/child agents run concurrently.
- Parent agents dispatch and configure bounded child agents.
- Product-owned delegation keeps child identity, transcripts, messaging and parent result delivery.
- The current workflow implementation starts through a human click and a client courier; unattended schedule and…
- Close/open/interrupt and queued messaging control children; workflows can wait on a human.
- Current Agents pane exposes durable hierarchy and live state.
- Saved linear workflows and per-node agent picks customize work.
- Proliferate offers a local runtime and a self-hosted control plane.
- The current local workspace implementation provisions local clones and Git worktrees; subagents share their workspace.
- Durable runtime state supports archive/restore and history; restart interrupts workflow.
- Workspace setup scripts prepare execution; E2B cloud template status conflicts across repository docs.
- Child roster and workflow shape are bounded.
- Runtime statuses distinguish running, attention, human wait and completion.
- Organization owner/admin/member roles are documented in the current source contract.
- Workspace MCP enforces per-call role/relationship authority; workflow human-in-loop nodes park a run until approved.
- Frozen invocation and durable child results retain workflow history.
- Workspace diff surfaces and child transcript navigation support review.
- Workspace files, diffs, Cowork artifacts and workflow context docs are durable; artifacts are not session-linked.
- Workspace Git operations and a hosting PR read model are documented.
- Workspace Git API can revert changes; turn-start checkpoints are default-off and their revert consumers never landed.
- Main application repository is AGPL-3.0.
- Full control plane/browser application and local desktop runtime self-host from source.
- Desktop telemetry has a documented complete disable flag; self-host mode differs from hosted.
- User provider API configuration and native coding-agent accounts are supported.
- Public website is a holding page; source repository remains available.
- Application source can be built/self-hosted without a software license payment; hosted free offer unverified.
Sources
Open the Proliferate page
QM from YCresearch correctionfreshness recheck41 findings · 15 sources
Findings today
- QM offers web/PWA, Slack and a published Apple Silicon Mac desktop client.
- QM provides administrator-registered HTTP MCP connectors and own-deployment customization.
- Swarms dispatch durable reserved workers and deduplicate retried spawn/send requests by requestId.
- Web swarm strip groups workers under parents and exposes control/status.
- Scoped computers retain installed tools and support provider-specific snapshots.
- Operator error records and optional Sentry reporting expose logs and failure causes.
- QM has posture-based command rules and exact-argument single-use MCP approvals.
- Published Files retain artifacts after sandbox retirement.
- QM supports own-cloud deployment and durable Files/share outputs separate from sandbox lifetime.
- QM telemetry is deployment-configured with separate analytics and backend/browser error controls.
- QM is distributed as an MIT-licensed npm package and source for operator-run deployment; README links a separate…
- QM memory uses scoped revisioned records and grant-based skills.
- QM can search notes, documents, email, databases and web; scoped memory retrieval does not prove architecture…
- Headless QM crons, watches and webhooks run while users are away.
- Swarm workers can pause, resume or stop with retained session results.
- Feature-flagged swarm workers expose configuration and retained session identities.
- QM runs its headless core and agent resources on operator infrastructure.
- QM has per-scope sandbox computers.
- Durable core state, provider snapshots and file publication have separate recovery boundaries.
- Swarm goals/time budgets and resource reconciliation have bounded failure behavior.
- Swarm resource pools, durable cleanup and notification pools bound provisioning.
- QM supports people collaborating with an agent in shared Slack/web rooms.
- Organization security posture and project membership controls are documented.
- Durable shared session snapshots preserve selected visible history, separate from raw tool/audit history.
- QM advertises running tests and monitoring repository CI through the agent.
- QM advertises agent work in repositories including PR creation and CI monitoring.
- The current QM core runs the Pi coding-agent loop.
- QM supports both operator-provided model keys and personal Claude or ChatGPT/Codex subscription sign-in for human turns.
- QM supplies an agent-guided deployment directory and requires Postgres.
- Core HTTP APIs and scoped swarm routes provide programmatic operations.
- The web package uses Lit with Vite.
- QM accepts human prompts in personal and shared channels.
- The current QM API and web UI expose conversations/runs, schedules and approvals without a native independent issue…
- Swarms document independently provisioned worker sessions.
- QM swarm APIs retain linked worker sessions and addressable follow-ups.
- QM estimates model spend per request from token usage and list prices and reports it by person, scope, cron and…
- QM displays conversation artifacts, files and approvals without an integrated proposed-code diff review interface in…
- Slack and web share identity and configuration; full colleague takeover is unverified.
- QM repository is MIT.
- QM is MIT open source and self-hosted in your own cloud with your own model keys; YC publishes no paid plan, and the…
- QM is MIT-licensed software an organization deploys in its own cloud with its own model keys.
Sources
Open the QM from YC page
Replicasfreshness recheckresearch correction48 findings · 24 sources
Findings today
- Replicas provides a web/phone-browser client, desktop apps for three OS families, CLI and native iOS access to cloud…
- Replicas integrates the six-agent picker Claude Code, Codex, Cursor, Muse Code, OpenCode and Pi across app and…
- Own coding-agent subscription or API credentials are required in addition to Replicas account.
- Connect agent and source control, configure environment, send first task.
- CLI, versioned REST API and MCP control workspaces; sanctioned unattended paths are API and Automations.
- Environment skills/plugins/MCP tools extend selected agents.
- Desktop is a native window around hosted deployment; iOS includes native controls.
- Prompt, integration or API starts a workspace; external tracker context can initiate work.
- Claude/Codex plans open beside the chat in Canvas or the sidebar, with download and selection replies sent to the agent.
- Mothership/ships retain parent-child work and board state; no general issue dependency enforcement established.
- Shared learnings, environment instructions and durable orchestrator notes preserve context.
- Language-server navigation, images and MCP knowledge are available.
- Replicas Motherships own a persistent task-ticket board distinct from the derived ship/workspace and pull-request list.
- Concurrent chats/workspaces and separate-VM ships run parallel work.
- Orchestrator assigns ships and exchanges followups/results.
- Shipped Mothership orchestration delegates separate VMs and exposes child transcripts and messages.
- API/Automations run saved tasks without an open client.
- Steer or queue followups, stop sessions and answer questions from multiple clients.
- Ships board exposes fleet state and attention, with recent organization activity.
- Single chats or Motherships support configurable workflows.
- Managed cloud VMs run independently of clients; Enterprise can run the platform in own infrastructure.
- Separate Linux VM per workspace is documented; chats within one VM share its working files.
- Sleep/wake and archive/restore retain state subject to deletion and history settings.
- Warm/start hooks, pools, services and secrets configure development environments.
- Documented shell timeouts, provider retries and usage limits constrain runs.
- Visible VM/chat/background statuses and usage dashboards expose progress.
- Metered VM compute and usage drilldowns are separate from own model billing.
- CPU/memory drive automatic VM resizing; explicit broader fair scheduling not established.
- Shared teammates can run the same Mothership and workspaces.
- Organization roles and resource Read/Write apply; Enterprise advertises SCIM.
- Organization merge/identity policies and MCP permissions constrain actions.
- Persistent chat and append-only configuration audit retain different histories.
- PR/MR tour, ordered diffs and agent chat support contextual review.
- Shared workspace/Mothership retains conversation and runtime for teammates.
- Canvas retains plans/media and workspace changes for review subject to retention.
- Orchestrator follows CI/review; merge respects repository rules.
- Branches, PR/MR creation, review, checks, conflicts and merge operate inside product.
- Eligible reviewers approve and merge subject to host rules and org policy.
- Workspace/ships retain PR links and delivery state.
- Review queue and Ships board consolidate many requests.
- The main Replicas platform is proprietary closed-source software, offered as a paid managed service or Enterprise…
- Enterprise platform self-hosting and API history access are documented.
- Provider processing, masked analytics, history settings and deletion are disclosed.
- Developer starts$50/month with matching compute allowance; Team$200+; model costs separate.
- Public signup and current weekly releases are documented.
- Vendor claims SOC2TypeI with TypeII in progress; report scope/period not independently verified.
- Evaluation is$100 finite credit without card, not ongoing free hosted usage.
- Public compute rates and allowance are explicit; Enterprise terms require sales.
Sources
Open the Replicas page
Rookeryfreshness recheckresearch correction46 findings · 13 sources
Findings today
- Rookery supplies native desktop, CLI and Slack clients for its resident daemon.
- Claude and Codex are the only product-owned session/worker backends.
- Own provider API keys or native Claude/Codex login enable agents.
- Bundled desktop runtime or Node 22 headless setup are documented.
- CLI and scoped external MCP drive daemon workers.
- Trusted capability packs supply instructions/skills/MCP to scoped agents.
- Electron/React desktop wraps daemon-backed mission control.
- Prompt, spawn dialog, GitHub issue or Linear ticket can initiate worker.
- Plan cards and session memory retain planning conversation; no native spec approval lifecycle established.
- Home sessions retain child worker relationships; general issue dependency enforcement unknown.
- Local master memory and trusted scope-aware capability packs share context.
- Trusted MCPs and repository context inform agent tasks.
- Rookery persists master/worker agent sessions and external ticket references, without an independent native…
- Global fleet runs parallel worktree workers without a concurrency cap.
- Master or human dispatches workers and routes followups.
- Shipped daemon orchestrator delegates inspectable Claude/Codex worktree workers.
- Persistent daemon rules trigger master turns or worker spawns without client.
- Messages queue at turn boundaries; interrupt/stop and ranked attention support intervention.
- Mission control presents master sessions and global live worker fleet.
- Single session, worker fleet or automation supports natural-language custom orchestration.
- The operator-owned resident daemon runs on a local machine or deliberately configured headless server, independently…
- Rookery workers run native agent processes in separate Git worktrees on the daemon host; they have no…
- Daemon restart reconstructs resumable workers only if session handle and worktree survive.
- Bundled runtime and scoped secret/environment setup support local execution.
- Worker USD budgets and Claude per-send turn caps stop at observed turn results.
- Worker states, transcripts and budget/loop notices expose progress and failure.
- Local usage dashboard and worker cost estimates support spend visibility.
- No fleet concurrency cap; per-worker budgets provide limited control.
- Distinct permitted Slack users can send instructions into the same persistent Rookery master thread.
- Rookery has capability-token control, Slack user allowlists and external MCP read-only/full-control scope, without…
- Headless defaults autoapprove tools; trusted capability digest and MCP scope are separate gates.
- Local session/worker events and checkpoint records persist decisions and activity.
- Per-worker diffs, checkpoints, transcripts and attention queue guide review.
- Forks preserve conversation into same or different provider; colleague access itself unknown.
- Workspace files, diffs, transcripts and Git checkpoints remain locally.
- Reviewer spawning is configurable worker-settled automation, not an enforced exact-revision ship gate.
- Product manages worktrees/branches/diffs; instructed worker uses shell for commit/push/PR.
- Worker shell can run delivery commands when instructed; no separate native merge/deploy authority established.
- Per-turn checkpoints snapshot tracked and untracked work for local restore.
- Attention queue ranks many worker results for review.
- Main application is Apache-2.0; proprietary provider SDK stays separately licensed.
- Full local desktop/daemon source is self-hostable; history lives in local SQLite.
- Task context goes to the selected provider while Rookery history/settings remain local.
- Rookery is Apache-2.0 open source with no paid plan published; you bring your own Anthropic/Codex API keys or CLI…
- Public desktop installers and source are available; v0.6.0 released October 19.
- Apache application is freely installable; model usage and own compute cost separately.
Sources
Open the Rookery page
Solentafreshness recheckresearch correction48 findings · 27 sources
Findings today
Sources
Open the Solenta page
super.engineeringfreshness recheckresearch correction40 findings · 24 sources
Superconductorfreshness recheckresearch correction47 findings · 24 sources
Paradevfreshness recheckresearch correction39 findings · 125 sources
Findings today
- Paradev, formerly Superset, has macOS desktop, experimental Linux, CLI and iPhone/iPad clients.
- Superset does not sell model access: all plans use the user's own agent subscriptions or API keys, and it does not…
- Onboarding detects git, gh and installed agents; projects come from a local folder or Git URL and must be git…
- Current CLI and OAuth HTTP MCP tools cover tasks, workspaces, agents, terminals, pages, automations and hosts.
- Current Paradev plugins supply MCP tools and skills with agent-specific support and a vendor credential proxy.
- Native tasks and connected Linear/GitHub issue lists support workspace intake.
- Current automations combine schedules and provider-event triggers, up to 25 per automation, on selected hosts or cloud.
- Local and user-hosted remote execution plus managed Team/Ultra Cloud workspaces.
- Cloud GB-hour usage limits are checked at start/wake and during execution; a running workspace stops about ten…
- Usage estimates token costs from local session logs across supported agents, with workspace/model/session drilldown…
- Granted organization members can open the same remote host workspaces and send input to their shared agent terminals.
- Remote hosts have explicit member grants with owner promotion and removal; Enterprise lists SAML SSO and SCIM.
- Built-in diff and Pull Requests views stage, commit, push, create PRs and merge (squash, merge commit or rebase and…
- PRs can be merged (squash, merge commit, rebase and merge) from Superset on the real PR through the user's GitHub…
- Current desktop-v2.0.5 root repository code is MIT-licensed.
- Per-seat Pro $20 monthly/$192 yearly; Team $60/$648 yearly; Ultra $200 monthly.
- Fixed local/Pro examples: one Free user $0; two Pro users $40/month or $384/year; five $100/month or $960/year.
- Superset became Paradev on October 26, 2026; the latest stable GitHub release is desktop-v2.0.8 (Nov 10, 2026).
- Free covers one local desktop/CLI user without a card.
- Seat prices, pooled Cloud GB-hours and $0.06/GB-hour extras are public.
- Paradev states it completed a SOC 2 Type II audit (Aug 27, 2026 changelog entry) and lists the SOC 2 Type II report…
- Paradev documents 22 built-in terminal agents, custom CLI agents and mixed-harness orchestration.
- Paradev documents parallel agents in separate local worktrees, with sidebar activity and workspace list/board views.
- Paradev bundles the superset:orchestrate skill: a Claude Code or Codex coordinator launches mixed-agent workers via…
- Paradev documents split/unified and binary diffs, versioned Pages with workspace lists and retrieval, browser…
- The MIT local core can import a repository, run agents and review diffs signed out and offline on the user machine,…
- Bundled Superset skills are installed into agents, and a session can be handed to another agent with its last 36,000…
- Design mode sends screenshots and DOM context to agents and the Slack agent searches the web with citations; code…
- Native tasks support create, edit, priority, assignee, labels (with label hierarchy), semantic search,…
- Agents can list, read, create, update and soft-delete tasks over MCP, CLI and SDK.
- Local workspaces are git worktrees on separate branches; the FAQ states they are not sandboxes and share processes,…
- Terminals run in a background daemon that survives app restarts and updates; unexpected agent-terminal deaths…
- Setup and teardown commands run per workspace on create and delete (install dependencies there).
- Agents can pause for tool approval in the chat pane.
- Sessions, run history and task activity persist; export, retention and integrity are not documented.
- Pull Requests and diff views offer split/unified/focus diffs, check status, review threads and iPhone diff review.
- PR-based workspaces show pull request state in the sidebar, and an Open Preview button appears when the branch has a…
- The Pull Requests view lists PRs with check counts, and attention-based board columns (Needs attention, Needs…
- Paradev discloses desktop analytics.
Sources
Open the Paradev page
Synarafreshness recheckresearch correction48 findings · 22 sources
Findings today
- Synara provides macOS, Windows and Linux desktop clients and a self-hosted browser workspace.
- Synara's fixed runtime adapters cover Claude Code, Codex, OpenCode, Cursor, Grok, Pi, OMP, Droid, Antigravity and Devin.
- Existing provider accounts and subscriptions are reused.
- Local and headless setup have explicit prerequisites.
- Paired external MCP and server startup/status commands are documented.
- MCP clients and app-owned tools have explicit provider boundaries.
- The desktop application is packaged with Electron.
- To-dos and coordinator requests are separate intake routes.
- Durable task briefs survive without dispatch.
- Hubs retain coordinator-worker associations and queues.
- Hub instructions, memory and Library persist across workers.
- Editor intelligence and external MCP supply scoped context.
- Native to-dos remain separate from agent chats.
- Agents maintain the same native to-do IDs.
- Hub workers and standalone tasks can run concurrently.
- Hub queues and idempotent operations bound dispatch.
- The product-owned gateway coordinates visible task workers.
- Schedules and heartbeat conditions drive app-owned runs.
- Tasks expose intervention and pending approvals.
- Inbox and Hubs group attention and worker progress.
- Quick chats and coordinator workflows can evolve with provider handoffs.
- Core server and agents can run on owned infrastructure.
- Synara separates task code with managed worktrees and launches provider processes on the execution host.
- Server transcripts and provider recovery retain explicit limits.
- Headless remote-host installation is documented; project dependency, service and collision handling are not.
- Failure policy and opt-in limit recovery constrain repeated work.
- Usage snapshots and work logs expose recorded state.
- Recorded provider usage is distinguished from missing metrics.
- Hub parallelism limits and a persistent queue bound concurrent workers.
- Multiple authorized browser clients can send instructions into the same server thread through the shared…
- Synara authenticates owner/client sessions and guards owner-only account operations.
- Task and integration privileges are bounded without worker self-approval.
- Transcripts and versioned Library retain history.
- Editor diffs and PR repair keep review near the task.
- Provider handoff preserves task identity with stated context limits; human colleague takeover was not established.
- Library, diffs and previews retain deliverables.
- PR checks and separate review tasks are documented without a universal spec gate.
- GitHub PR and local Git operations are integrated.
- Remote PR actions use GitHub authority.
- Checkpoint and PR records retain recovery context.
- PR aggregation and attention groups help batch review.
- Main application code is public under MIT.
- The core ADE can be run from source or a headless release.
- Local state, connectors and diagnostics have separate boundaries.
- Local use of Synara is free ($0, no card or account) and MIT open source.
- Stable release and source support channels are public.
- MIT source can be used without an ADE license fee.
- Source license is free while provider billing remains external.
Sources
Open the Synara page
T3 Codefreshness recheckresearch correction37 findings · 76 sources
Findings today
- v0.1.7 documents Mac/Windows/Linux desktop, iOS/Android mobile, web and CLI.
- v0.1.7 documents installed Codex, Claude Code, Cursor, Grok Build, Kiro (beta, Early Access), OpenCode, Antigravity,…
- T3 Code reuses each provider's own subscription or login, supports multiple accounts per provider and per-instance…
- The t3 CLI and remote HTTP MCP expose project/thread operations in v0.1.7.
- Partly established for T3 Code v0.1.7 at thread level: threads can exist without a project, and settling keeps…
- Provider usage-limit stops (for example Claude subscription limits, and Codex windows when Codex reports them) are…
- v0.1.7 Usage shows provider/token estimates and model trend, cache-hit-rate and cost-per-million drilldown.
- Terms state no T3 Tools subscription fee currently, unless a feature states otherwise; they do not mention seats or…
- The current terms state no T3 Tools subscription fee and do not mention seats or user counts; provider usage and…
- Latest stable is v0.1.7, displayed November 4; v0.1.8 nightly builds are separate.
- T3 Code is currently offered without a T3 Tools subscription fee, and the terms allow paid features with prices…
- The pricing URL returned HTTP 404 and the homepage was reachable in the November 7 second pass.
- T3 Code relies on the underlying harness for MCP servers, plugins and skills; it reads skills from harness config…
- Plan mode and proposed-plan artifacts exist in current stable.
- Stable v0.1.7 has T3-owned MCP delegation across compatible providers, with role/model selection, scoped…
- Stable v0.1.7 runs server-owned recurring/one-time schedules and webhook tasks with signature, filtering,…
- Users can queue follow-ups or steer new messages into a running turn, interrupt the turn with the thread.stop…
- The sidebar groups and searches saved threads; Agents views and cards expose delegated children and retained results…
- Threads can start without a project and choose model or permission defaults.
- T3 Code uses project checkouts or per-thread Git worktrees and launches provider/ACP processes on the environment host.
- Threads persist on the environment server.
- Project actions and automatic submodule initialization for new worktrees are documented; desktop-managed SSH relies…
- The Usage page reports token use, cache savings, model breakdowns and estimated cost over 24 hours to 90 days, with…
- Optional beta guestlinks share recent/live thread history and accept guest messages/files, gated by the owner…
- T3 Code controls devices with environment pairing credentials, session tokens and scoped grants; administrators can…
- Thread permission modes expose command, edit and child approvals in the conversation, with provider-specific…
- Current stable PR review includes mobile diffs/draft comments and optional beta model-written guides with ordered…
- Partly established for T3 Code v0.1.7: pull requests can be reviewed, reviewers requested and merged from T3;…
- Threads can use per-thread worktrees and branches, commit, push and open pull requests with generated titles and…
- Linked pull requests, PR status badges and GitHub revert-PR creation are documented; rollback via checkpoints and…
- Optional beta PR review guides provide ordered stops, grouped mechanical changes, retained review context and agent…
- Self-hosted environment servers can use LAN, SSH or optional T3 Connect.
- The live privacy policy describes anonymous usage data sent by default; the pinned telemetry guide describes a…
- Agents run on the user's own machine or a machine they keep running (SSH, T3 Connect, LAN/tailnet); clients on other…
- Threads retain conversations and tool activity, but supported rewind can remove later conversation history.
- Portable provider handoff preserves a bounded selection of saved conversation context and retrieval references.
- Workspace file changes are captured in hidden Git checkpoint refs, and the Terms describe T3 displaying file changes…
Sources
Open the T3 Code page
Temboresearch correctionfreshness recheck42 findings · 22 sources
Findings today
- Tembo offers web/desktop, device CLI and documented iOS/Android phone-browser PWA.
- Public API and sandbox tools expose agents, sessions, artifacts, schedules and run state; pairing CLI is separate.
- Tembo planning mode produces a plan for review and a shareable artifact.
- Tembo documents sessions consuming worker capacity and self-hosted pools running separate VM sessions.
- Tembo capacity queues integration bursts and self-hosted hosts claim sessions.
- Scheduled and event-triggered agents run in the background in Tembo sandboxes; capacity and spend constraints are…
- Tembo exposes planning/questions, live steering, recovery controls and reset/retry.
- Cloud VMs and licensed/self-hosted machines have distinct responsibility and resource boundaries.
- Dated release notes report automatic recovery; self-hosted same-host memory and replacement-host disk restores differ.
- Project environments cache setup, and setupScript reruns only on fresh builds.
- Session admission respects allowance/overage caps; existing runs can continue and setup hook failures are non-blocking.
- Public API reports run status, duration and errors; Billing separates usage categories.
- Self-hosted host controller sizes pools from running/waiting memory and bounded host counts.
- Tembo documents enterprise SSO, admin/reviewer/operator roles and team/repository permissions.
- Vendor documents centralized audit trails for every agent session, with prompt-to-PR traceability, configurable…
- Tembo supports file artifacts created for sessions, preview service artifacts and PR-linked diffs.
- Current plans pool gateway/compute usage with optional capped overages.
- Compute/model units, overage admission and custom Enterprise pricing are documented.
- Tembo manages MCP integrations and repository/workspace skills.
- External issues, chat and Git events supply agent intake.
- Repo/project rules and managed skills provide repeatable workspace context.
- Tembo publishes model/compute usage accounting with separate Billing categories.
- Setup hooks are not delivery gates; prePush is explicitly not executed.
- Session sandbox cleanup differs from Cloud Git mirrors, audit retention and provider inference.
- Vendor reports SOC 2 Type II and ISO certifications; report scope, issuer and audit period remain unverified.
- A seven-day trial is followed by paid continuation when time or allowance expires.
- Agents have selectable harness and model defaults.
- Tembo documents BYOK and separate agent-subscription routes.
- Self-hosting uses a licensed machine image and browser installer.
- Rule files can carry architecture conventions; enforcement was not demonstrated.
- Agents can update external issue records through connected services.
- Tembo supports planning, session forking and reusable agent instruction revisions.
- Cloud sessions have dedicated VMs; device commands use host-user permissions.
- Workspace visibility can be set per session.
- Connected devices are restricted to owner-started human sessions.
- Tembo documents expandable task diffs and line comments.
- PR actions support multiple repositories and Git providers.
- Documentation leaves review and merge to the user.
- PRs retain links to the starting issue, error or session.
- Tembo is offered under vendor-authored service/license terms.
- The licensed self-hosted platform remains dependent on vendor release licensing.
- Desktop and Devices have explicit release/access restrictions.
Sources
Open the Tembo page
Unpeelfreshness recheckresearch correction39 findings · 24 sources
Findings today
Sources
Open the Unpeel page
Verdentfreshness recheckresearch correction41 findings · 32 sources
Findings today
Sources
Open the Verdent page
Vibe Kanbanresearch correctionfreshness recheck43 findings · 40 sources
Findings today
- Local Codex uses native login or API keys and detects custom configuration directories.
- Local plan approval and workspace notes coexist with explicit archive retention.
- Local sessions share files and reusable prompts while keeping conversations separate.
- Local chat provides stop, queue/retry and history controls.
- Local users can try a different approach by duplicating a workspace, which keeps the repository/branch configuration…
- Branches/worktrees separate workspace files; sessions within one workspace share files/Git.
- Local archive preserves history/files and missing worktrees rebuild from last commit with uncommitted changes lost.
- Repository setup/dev/cleanup scripts prepare environments with user-defined lifecycle behavior.
- Local logs include expandable tool command/search results and execution states.
- Local conversation history persists across session switches and archive, but message retry can replace later visible…
- Local review UI provides diff modes, filtered files and submitted GitHub comments.
- Local archive retains conversation notes/files, while diffs and previews remain separate review views.
- Local telemetry has a settings toggle and a released source default of enabled, but a successful sign-in turns…
- Local merge is documented.
- The pinned release ships an explicit orchestrator MCP mode.
- The continuing local product provides browser workspaces, a product CLI and Tauri desktop packages.
- Released source integrates multiple coding-agent adapters; actual mixed-harness execution was not tested.
- The documented local flow is npx, agent authentication and repository selection.
- Local MCP provides workspace start, session creation/follow-up and execution inspection.
- Vibe Kanban manages per-harness MCP configuration; connections and permissions execute through the selected harness.
- The desktop distribution uses Tauri webviews and a web UI.
- Local prompt-driven workspaces continue independently of the retired hosted issue tracker.
- Local chat accepts file mentions and images; architecture-aware retrieval and enforcement remain untested.
- The vendor-hosted issue board is retired.
- Docs describe independent sessions that can run simultaneously and multiple workspaces.
- In the pinned local continuation, work starts through explicit session/workspace requests.
- The local workspace list supplies session status context, but a 12-state fleet/epic fixture was not evaluated.
- Local agent execution continues.
- Approval timeouts are documented; enforced time, token and spend budgets remain unverified.
- The documented context gauge is not a verified cost ledger.
- Resource limits, check reuse and fleet backpressure remain unknown.
- The shutdown announcement retired vendor-hosted issue discussion/comments.
- The vendor-hosted organisation/project service was named for retirement.
- Harness-specific plan/permission controls appear in local configuration.
- Repository scripts can run checks; docs advise review and CI before merge.
- Local workspace docs cover branches/worktrees, PR creation, push, merge and rebase/conflict handling.
- Git PR workflow is documented, but completion-to-revision evidence, rollback and stale-approval invalidation were…
- Single-workspace review is documented; six-output batch review, dependency queues and exact-revision approvals…
- Pinned repository distribution has an Apache-2.0 license.
- Local workspaces continue without vendor sign-in.
- Commercial services were sunset while the local open-source project continues.
- The local open-source project remains available without vendor sign-in.
- Commercial subscriptions were terminated, so historical Vibe Kanban tariffs do not describe a current service.
Sources
Open the Vibe Kanban page
Warpresearch correctionfreshness recheck39 findings · 25 sources
Findings today
- Warp Terminal downloads are offered for macOS 10.14+, Windows 11/10 and Linux.
- Terminal integrations include Claude Code, Codex, OpenCode, Cursor CLI, Gemini CLI, Pi and Grok Build, plus Amp,…
- A third-party Claude Code session uses the user’s paid Claude plan or API credits through Claude login, separate…
- Install the CLI, start warp, approve its browser device-code login, then send a first prompt.
- warp starts the standalone interactive Warp Agent CLI; /status reports version, conversation ID and account.
- Warp local agents can use configured command, Streamable HTTP or SSE MCP servers.
- /plan or a natural-language request creates a structured plan.
- Global Rules and repository AGENTS.md or backward-compatible WARP.md provide reusable instructions.
- Without indexed context agents can still navigate with grep and sed.
- Agents automatically create step lists for complex requests.
- The agent automatically decomposes complex requests into task lists.
- Plans persist as reusable documents in Warp Drive.
- A native parent agent spawns and coordinates child agents.
- /orchestrate accepts a natural-language task and proposed child boundaries.
- Warp runs agents on local worktrees and via SSH.
- Each worktree has its own checked-out branch and Git status.
- Finished command blocks can be shared by permalink or embed; this does not establish co-prompting.
- Single-team workspaces have Owner, Admin and Member roles.
- Command permissions, allowlists and denylists control approval prompts.
- The product page describes reviewing changes, commenting and sending changes back to agents.
- A collaborator with edit access can continue prompting the same shared session.
- Warp detects worktrees for branch-aware review and indexing.
- Line/block comments collect concrete requested changes.
- Warp's privacy documentation says its client source is open source under AGPL v3 and links warpdotdev/warp.
- Terminal and customer execution are local/customer-controlled, while orchestration, observability and inference…
- With telemetry disabled no console interactions are persisted on Warp servers, according to the privacy documentation.
- Warp states SOC 2 Type II certification.
- Ctrl+C stops a streaming agent response or running command.
- CLI conversation history saves and syncs with the Warp account, including after exiting.
- Every CLI conversation is saved automatically as work proceeds.
- Client implementation: Native Rust desktop app on Warp's own GPU-rendered warpui framework: AppKit and Metal on…
- Warp documents running several Warp Agent conversations in parallel, tracked in the Agent Management Panel, and…
- Fleet and epic management UI: Vertical tabs list every tab or pane with Git branch, working directory, diff stats,…
- Tab indicators can mark commands that exit with errors; full task-state consistency was not tested.
- Warp says the user's own provider bills customer-supplied inference separately.
- An integrated editing and review interface is advertised, without an artifact persistence test.
- Build is $20/month monthly or $18/month annually.
- Free is $0/month with core terminal features, limited cloud access and limited conversation storage; additional…
- Platform usage applies to cloud runs and to local Business or Enterprise runs with customer-supplied inference;…
Sources
Open the Warp page
Archived event links
Something wrong or missing?