Runbooks

Use these commands as the operator baseline for recurring production checks.

How to use this page

1. Start with the core production gate for release readiness.
2. Use weekly evidence generation to keep audit and governance material current.
3. Run replay regression checks after changes to memory, policy, or execution flow.
4. Use consistency checks when a scope appears degraded or inconsistent.

Core production gate

npm run -s gate:core:prod -- --base-url "$BASE_URL" --scope default

Weekly governance evidence

npm run -s evidence:weekly -- --scope default --window-hours 168 --strict

Replay regression checks

npm run -s e2e:replay-learning-fault-smoke
npm run -s e2e:replay-learning-retention-smoke

Scope consistency check

npm run -s job:consistency-check:scope -- --scope default --strict-warnings

Recommended cadence

1. Before release: core production gate
2. After release: replay regression checks on critical workflows
3. Weekly: governance evidence
4. During incidents: scope consistency check and replay commands