Constrain access and rehearse rollback

Separate runtime and operator roles for support-answer service comparing retriever_v1 vs reranking retriever_v2. Runtime may only perform the narrow actions that experiment owner measuring groundedness lift with SRM checks needs; operators get audited break-glass with TTL. Encode a negative probe that denies the privilege trick related to reassignment every request → users flicker; metrics uninterpretable.

Redact prompts/PII at collection. Secrets enter via a manager or workload identity — never source, fixtures, or exception strings. Incident AB-SRM-FAIL-27 should be impossible if these controls hold. Output allowlists and schema checks stay in force on error paths.

Rehearse the rollback or kill switch timed against a clock. Record actor, reason, prior revision/secret/flag, and verification query. Invariant reminder: assignment sticky by user_id; analysis gated on sample ratio mismatch (SRM) p>0.001 fail.

Abuse cases unique to this lab include the privilege path implied by reassignment every request → users flicker; metrics uninterpretable. Prove a read-only role cannot mutate. Break-glass tokens expire; leftover tokens fail the drill. Dependency pin/digest story matters when images or models move under you. Document how to rotate the credential that support-answer service comparing retriever_v1 vs reranking retriever_v2 uses without a full outage window longer than your dual-run plan. Security evidence is part of ship, not an appendix nobody reads.

List network egress destinations and justify each. Ensure debug endpoints are off by default in the shipping config. Verify that error responses do not echo secrets or raw stack frames to clients. For multi-tenant paths, add a cross-tenant probe fixture. Time the rollback drill twice — once with the author, once with a peer. Store the drill transcript beside the threat notes for the incident id. In this chapter the product is support-answer service comparing retriever_v1 vs reranking retriever_v2, the human stakeholder is experiment owner measuring groundedness lift with SRM checks, and the incident id you design against is AB-SRM-FAIL-27. Re-state the oracle in your notes — after 20k sessions, v2 groundedness +2.1pp, p95 latency +80ms within budget — and keep the invariant visible: assignment sticky by user_id; analysis gated on sample ratio mismatch (SRM) p>0.001 fail. Track groundedness_lift_pp and srm_pvalue as the scoreboard. Surface under change control: POST /v1/support/answer. If you only have forty minutes, finish the fixture for reassignment every request → users flicker; metrics uninterpretable before polishing UI. Promotion language stays ternary: promote, hold, or roll back based on evidence, not hope.