discover-test-patterns

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [NO_CODE]: The skill consists entirely of markdown instructions and does not include any executable scripts, binaries, or logic.
  • [INDIRECT_PROMPT_INJECTION]: The skill specifies a process for reading untrusted data from the project's own test files.
  • Ingestion points: Representative test files (2-3) located using glob patterns like *_test.go or *.test.ts.
  • Boundary markers: No explicit markers or 'ignore embedded instructions' warnings are provided for the agent while processing the file content.
  • Capability inventory: The skill is restricted to file system read access and writing output to thoughts/notes/testing.md; it lacks network access or shell execution capabilities.
  • Sanitization: No content validation or sanitization is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 10:32 PM