Renn F
2fb63fed1f
Build a complete user-facing documentation site (MkDocs Material) under docs/, served at roboco.dev/docs via a new gh-pages deploy workflow. - Sections: Get Started, The Company, the Tour, Operating the Panel, Choosing & Running Models, Cost & Observability, Optional Subsystems, Configure & Deploy, API Reference, Troubleshooting & Security (55 pages). - mkdocs.yml (Material theme; excludes the agent-facing rag/ corpus, internal scratch, and orphaned stub trees) and .github/workflows/docs.yml (mkdocs gh-deploy to gh-pages). - Retire the stale root usage.md and deployment.md to redirect stubs into the site. - Fix the docs tooling: add the pymarkdownlnt dependency + .pymarkdown.json, run serve-docs/lint-docs/fix-docs under the docs extra, add a build-docs strict gate. - Fix the roboco console-script entry point (cli, not the un-awaited async main). - README: correct the project-structure tree (optimal.py, alembic) and link the docs site.
3.1 KiB
Agents & work sessions
The Agents page (/agents) is your view of the workforce: who's running, who's idle, who's stuck, and what each one is costing in tokens. From an agent's detail page you can spawn or stop it and watch its reasoning live. The Work Sessions ledger (/work-sessions) is the read-only record of every branch-and-PR an agent has worked, tying git activity back to the agents that produced it.
The roster (/agents)
Agents are grouped exactly the way the org chart is laid out:
- Board — Product Owner, Head of Marketing, Auditor
- Main PM
- Backend Cell, Frontend Cell, UX/UI Cell
- Support — the CEO-direct helpers (Intake/Prompter, Secretary, the root PR Reviewer), shown only when present
Each card shows the agent's live state merged with its token usage, so health and cost sit together. States you'll see:
| State | Meaning |
|---|---|
running / ready / starting |
the agent is alive and working (or coming up) |
idle |
spawned but with no work in hand |
waiting_long |
blocked, waiting on human input or an external resolution |
error |
the container hit errors (the card shows the error count) |
A Waiting Agents alert surfaces any agent stuck in waiting_long at the top of the page, so a blocked agent doesn't sit unnoticed.
Agent detail (/agents/[id])
Open an agent to control it and watch it think. The page gives you:
- Spawn — bring the agent's container up.
- Stop (graceful) and Force Stop — wind it down cleanly or kill it immediately.
- Resolve Wait — when an agent is in
waiting_long, this dialog is how you hand it the input or decision it's blocked on. - Live stream viewer — while the agent is active, a viewer streams its reasoning in real time over the agent WebSocket, so you can literally watch it work.
!!! tip "Stop the bleeding" If an agent is crash-looping or burning tokens, Force Stop from its detail page is the fastest way to halt it. For provider rate-limits and overloads you don't need to intervene — those park and auto-resume; you'll see an amber banner instead (see resilience).
Per-agent token spend rolls up into the Metrics page for cost analysis, and what agents say and learn is in Communications & journals.
Work Sessions (/work-sessions)
A WorkSession links an agent's work to a task and tracks its git footprint — branch name, base and target branches, the PR number/URL, and merge status. The Work Sessions page is a read-only ledger of those sessions, with search-by-branch and a status filter (state, again, lives in the URL).
!!! note "No sidebar link"
Work Sessions has no entry in the sidebar nav. Reach it by typing /work-sessions directly or by following a link from elsewhere in the panel. It's a reference ledger, not a daily-driver page.
For the lifecycle of a branch from cut to merge, see the merge model.
Next
→ Git to operate on those branches directly, or Tasks & Kanban to see the work the agents are running.