golden-file-conventions
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides documentation on established snapshot testing practices and does not contain malicious code or dangerous instructions.
- [SAFE]: The documentation emphasizes sanitizing sensitive and volatile data like timestamps, UUIDs, and absolute file paths, which enhances both the security and stability of the testing pipeline.
- [INDIRECT_PROMPT_INJECTION]: The skill involves processing snapshot files (ingestion point: fs.readFileSync in test files), which constitutes a data ingestion surface. The agent executes tests (capability: node --test) on these files. However, the skill provides extensive sanitization rules and the provided task data contains no malicious instructions. The risk is minimized by the skill's own best practice requirements.
- [SAFE]: No prompt injection, data exfiltration, persistence mechanisms, or obfuscation patterns were detected in the skill files or evaluation tasks.
Audit Metadata