PyAutoBrain

Start-of-day wake-up routine for the PyAuto workspace — sync every repo to main and clean generated cruft (local), then a gh-API status glance (overnight scheduled-run conclusions, version-pin drift, resume context) plus /health and /hygiene, ending in one prioritized digest. Runs on the CLI and on mobile Claude Code chat / Codex (auto-skips local-only steps when there is no workspace). Use when starting the day or asked for a morning status/cleanup pass.

Coordinate PyAuto build, deploy, or release execution through the PyAutoBrain Build Agent, including its vitals consultation and PyAutoHands delegation. Use when the user asks to build, publish, deploy, or release.

Use the single PyAuto health door for readiness assessment, validation sweeps, active-work status, release-run status, or worktree status through the Brain and Heart-owned procedures.

Periodic hygiene sweep across PyAuto repos and worktrees — audit stale feature branches, tracking refs, stashes, and dirty checkouts, then execute per-bucket cleanups after confirmation. Use for end-of-week tidying or unwieldy branch lists.

Ship PyAutoLabs source-library changes — run tests, commit, push, open pending-release PRs, analyze downstream workspace impact, and update the issue and PyAutoMind task state.

Analyze how a mature PyAuto domain assistant should be reproduced as an exact clone, sibling, or seed through the PyAutoBrain Clone Agent, and birth a lightweight seed once a human has answered the clone-mode question. Use for assistant-cell cloning decisions and seed births.

Run the visualization review loop through the PyAutoBrain Eyes Agent — survey a visualization workspace's figure surface, render via its gallery harness, review figures with the human, and route accepted critiques to intake/start_dev. Use when the user wants to look at, judge, or update the organism's plots and figures.

Hear and answer the PyAuto community through the PyAutoBrain Community Agent — scan every repo for user-filed issues awaiting a response, triage one issue's context sufficiency, then draft human-approved replies and route actionable work via start_dev_for_user. Use when responding to external users' GitHub issues or checking what the community is waiting on.

Turn raw PyAutoLabs ideas, bug reports, or loose requirements into classified and sized PyAutoMind prompt files through the PyAutoBrain Intake Agent. Use before start-dev when intent is not yet formalized.

Register and execute a resumable sequence of PyAutoLabs tasks using PyAutoMind state and the PyAutoBrain autonomous queue contract. Use for ordered multi-task series rather than a single development task.

Ship PyAutoLabs workspace and tutorial changes — validate scripts, commit, push, open pending-release PRs behind the library-first merge gate, and update the issue and PyAutoMind task state.

Trial a new non-linear sampler through the ingest → prototype → profile → promote pipeline — point it at a sampler's GitHub repo, get it running on the standard problem and on a likelihood the user owns (MLTracker diagnostics, benchmark comparison), then (if warranted) the full PyAutoFit implementation. Use when the user wants to try, benchmark, or promote a sampler / search / MCMC / nested-sampling / HMC method, or gives a sampler repo URL.

Run a named PyAutoBrain conductor or faculty directly through bin/pyauto-brain. Use for low-level agent invocation, agent help, or debugging the Brain router.

Infer the appropriate PyAutoBrain command or development work type from a free-form PyAutoLabs request. Use when the request is actionable but does not name intake, feature, bug, refactor, health, build, docs, or research explicitly.

Plan workspace or HowTo example authorship through the PyAutoBrain Workspace Agent — target repo, audience register, placement, sibling to mirror, format checklist — or survey a workspace repo's example catalogue. Decision-only; authoring itself runs through start_dev → start_workspace.

Default entry point for PyAutoLabs development tasks — implement, fix, add, change, refactor, migrate, optimize, document, test, ship, or open a PR — even when the user does not say start_dev. Not for pure questions, reviews, status checks, or explicit workflow opt-outs.

Audit and prioritise the organism's code-quality upkeep — slow tests/scripts/imports, CLI noise, dependency-cap drift, stale API docs, git debris — through the PyAutoBrain Hygiene Agent, which reasons and delegates fixes without editing source itself.

Start a PyAutoLabs development task for another contributor, including plan, issue, branch, and handoff preparation without claiming the local implementation worktree.

Set up PyAutoLabs source-library development after start_dev — create or resume task worktrees for the PyAuto* libraries, register claimed repos in PyAutoMind, and prepare for source edits.

Set up PyAutoLabs workspace or tutorial development — attach or create workspace worktrees, inspect upstream library API changes, and register workspace repos in PyAutoMind.

Classify and plan PyAutoLabs bugs, regressions, failing tests, incorrect output, or PyAutoHeart findings through the PyAutoBrain Bug Agent before starting development.

Start PyAutoLabs documentation, tutorial, notebook, example, or narrative-prose work through the Brain development workflow with the docs work type fixed.

Select, size, phase, or plan a PyAutoLabs feature through the PyAutoBrain Feature Agent, then route an approved task into start-dev. Use for new capabilities and next-task selection.

Recall cited PyAuto scientific, architectural, assistant, or task-history context through the read-only PyAutoBrain Memory Faculty. Use before substantial planning when prior decisions or domain knowledge could change the approach.

Plan PyAutoLens performance campaigns, ingest fresh profiling results, or triage profiling drift through the PyAutoBrain Profiling Agent without running sweeps directly.

Plan behavior-preserving PyAutoLabs restructuring through the PyAutoBrain Refactor Agent, including invariant checks, candidate mining, and safe-autonomy routing.

Coordinate PyAuto release readiness, rehearsal, validation, nightly-driver inspection, or a human-authorized release through the PyAutoBrain Release Agent. Use for release operations while preserving Heart gates and human-required manual release approval.

Start exploratory scientific, algorithmic, architectural, or prior-art investigation through the PyAutoBrain development workflow with the research work type fixed.

Prepare a PyAuto feature branch ReviewSurface and apply an independent CLEAN, FINDINGS, or BLOCKED judgment for the autonomous ship gate. Use for dev-workflow branch review, not release readiness or implementation fixes.

Run the generic PyAutoBrain task queue with explicit --auto activation, per-task autonomy caps, checkpoint-and-continue behavior, and durable PyAutoMind state.

Inspect PyAuto sampler script tiers, PyAutoFit search coverage, benchmark records, and tier gaps through the read-only PyAutoBrain Samplers Faculty. Use for sampler selection or pipeline planning without running samplers.

Post a progress update to a GitHub issue from the current CLI session — commits, summary, and remaining work.

Read and explain the authoritative PyAutoHeart readiness verdict through the read-only PyAutoBrain Vitals Faculty. Use when a conductor or user needs current GREEN, STALE, YELLOW, or RED reasons without dispatching work.