keystone

Full dual-flywheel audit — Learning (capture from review) + Pruning (remove dead rules). Periodic charter hygiene.

One-time initial charter scaffold — detect stack, seed state, scaffold idioms, classify sensors, build the globs index.

Run the drift sensor on the current diff — fast pre-verify check that loaded charter rules still match what the code is doing.

Triage the code-debt ledger — review the debt sensor's findings and update corpus/state/code-debt.md.

Regenerate .keystone/INDEX.json — the primitive descriptor index every agent reads at session start.

Capture a learning candidate (surprise, incident, review finding) into .charter/learning/inbox/ for later synthesis.

Switch the charter pacing mode (paired / solo / autopilot).

Scaffold a new lifecycle action (atomic unit of work) at the canonical path.

Scaffold the per-agent adapter triple (lifecycle, sensors, activation) for a new host.

Scaffold a corpus reasoning entry (the *why*) at the canonical path.

Scaffold a new rule guide + paired corpus stub at the canonical path.

Scaffold a new playbook (ordered sequence of actions) at the canonical path.

Scaffold a new policy repo skeleton outside the current project.

Scaffold a new sensor (automated check) at the canonical path.

Enter the planning phase — load codebase state + matching idioms for the touched region, then sketch a plan.

Run the four review sensors on the current diff (functional, security, risk, deployment).

Capture intent + acceptance criteria for a unit of work. First action on any task.

Promote accepted learning-inbox candidates into the right corpus / guide / sensor layer, then refresh INDEX.json.

End-to-end task workflow.

Check vendored policies for drift and the strict cascade for project-layer violations.