Constrain access and rehearse rollback

Separate runtime and operator roles for support API: versioned policy retrieval + scoped order tool + cited answer or escalate. Runtime may only perform the narrow actions that customer asking where order A12 is without leaking other customers' orders needs; operators get audited break-glass with TTL. Encode a negative probe that denies the privilege trick related to model passes attacker order_id and tool executes.

Redact prompts/PII at collection. Secrets enter via a manager or workload identity — never source, fixtures, or exception strings. Incident CAP-SUPPORT-TOOL-09 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: tool args bind to session customer_id; refunds never auto-executed.

Abuse cases unique to this lab include the privilege path implied by model passes attacker order_id and tool executes. 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 API: versioned policy retrieval + scoped order tool + cited answer or escalate 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 API: versioned policy retrieval + scoped order tool + cited answer or escalate, the human stakeholder is customer asking where order A12 is without leaking other customers' orders, and the incident id you design against is CAP-SUPPORT-TOOL-09. Re-state the oracle in your notes — order lookup returns status+policy cite; cross-customer order_id → deny+escalate — and keep the invariant visible: tool args bind to session customer_id; refunds never auto-executed. Track unauthorized_tool_calls == 0 and citation_rate ≥ 0.9 as the scoreboard. Surface under change control: POST /v1/support/turn. If you only have forty minutes, finish the fixture for model passes attacker order_id and tool executes before polishing UI. Promotion language stays ternary: promote, hold, or roll back based on evidence, not hope.