Set EXIF, retention, and prompt-injection boundaries

Separate runtime and operator roles for typed receipt-vision API that extracts total + merchant with uncertainty. Runtime may only perform the narrow actions that expense bot that must refuse blurry or non-receipt images needs; operators get audited break-glass with TTL. Encode a negative probe that denies the privilege trick related to screenshot of a meme parsed as a $9,999 expense.

Redact prompts/PII at collection. Secrets enter via a manager or workload identity — never source, fixtures, or exception strings. Incident VISION-MEME-77 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: image ≤ 4 MiB, MIME allowlisted, model output schema-validated before DB write.

Abuse cases unique to this lab include the privilege path implied by screenshot of a meme parsed as a $9,999 expense. 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 typed receipt-vision API that extracts total + merchant with uncertainty 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 typed receipt-vision API that extracts total + merchant with uncertainty, the human stakeholder is expense bot that must refuse blurry or non-receipt images, and the incident id you design against is VISION-MEME-77. Re-state the oracle in your notes — sharp receipt → total=24.50 merchant=Cafe Nora confidence≥0.7; blurry → abstain — and keep the invariant visible: image ≤ 4 MiB, MIME allowlisted, model output schema-validated before DB write. Track extraction_precision and abstain_on_non_receipt ≥ 0.95 as the scoreboard. Surface under change control: POST /v1/receipts/extract. If you only have forty minutes, finish the fixture for screenshot of a meme parsed as a $9,999 expense before polishing UI. Promotion language stays ternary: promote, hold, or roll back based on evidence, not hope.