Review Finding Disposition

AI Agent Instructions

Run independent reviews against one pinned artifact and acceptance policy. Submit each finding with its evidence and revision; turn blocking findings into repair tasks and revalidate stale reports. Track required-job completion separately in code. Missing, failed or timed-out required checks continue to block release regardless of an advisory model outcome.

Class: superclassify/review-finding-disposition@1.0.0
Setup: https://superclassify.com/skills/superclassify/SKILL.md

First-time API access

When the user asks to run or integrate this Class, explain its workflow and that live Jev runs use SuperClassify service credits. Public card descriptions can be read without an account. Reuse an existing account, credit balance and API key when available. Otherwise, guide the user through the official setup links below. The user completes sign-in and any credit purchase on the website, then saves the API key in their application’s server secrets as SUPERCLASSIFY_API_KEY—not in chat. A separate TypeSafe account or Jev key is not required for the SuperClassify API. Reading a link does not authorize a purchase or paid run.

1. Create an account or sign in
2. Add credits if needed
3. Create an API key

Evidence-backed blocker

Snapshot abc123. Policy: any cross-tenant data exposure blocks release. Finding: a reproducible authenticated request as tenant A reads tenant B data on abc123. Attached request/response evidence and independent repro receipt support the claim.

Author expectation: {"disposition":"blocking"}

Supported nonblocking improvement

Snapshot abc123. Finding: two documentation sentences repeat the same explanation. The repetition is quoted and verified. Policy explicitly treats style-only documentation repetition as advisory; required tests pass independently.

Author expectation: {"disposition":"advisory"}

Stale review needs revalidation

Current snapshot def456 changes authorization logic. The finding and test receipt refer only to old snapshot abc123. No rerun or applicability analysis exists, so it is unknown whether the reported issue remains.

Author expectation: {"disposition":"needs_review"}