skills/wihlarkop/proofcraft/finish/Gen Agent Trust Hub

finish

Pass

Audited by Gen Agent Trust Hub on Sep 27, 2026

Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [NO_CODE]: The skill consists entirely of Markdown documents and a JSON evaluation file. It does not include any scripts, executables, or code that could be run by the agent.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that involves reading project artifacts. Ingestion points: references/_shared/orchestration/reality-check.md (inspects branch, HEAD, working tree, etc.) and references/_shared/orchestration/handoff.md (records PR references, verification evidence). Boundary markers: Absent. Capability inventory: The skill uses general agent capabilities for file reading and git operations. Sanitization: Absent. This surface is inherent to the skill's purpose of software verification.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 27, 2026, 12:57 PM