ln-523-auto-test-planner

Pass

Audited by Gen Agent Trust Hub on Apr 27, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches shared reference files and environment state contracts from the author's official GitHub repository (github.com/levnikolaevich/claude-code-skills) when local copies are unavailable.\n- [DATA_EXFILTRATION]: Accesses story descriptions and manual test results from the Linear task management platform or local filesystem to gather context for test planning.\n- [PROMPT_INJECTION]: The skill processes untrusted input from story comments and manual test results, creating a surface for potential indirect prompt injection.\n
  • Ingestion points: linear list_comments, Read story.md, and local glob patterns for task files.\n
  • Boundary markers: Not explicitly defined for the isolation of untrusted data.\n
  • Capability inventory: Writes validation artifacts to .hex-skills/ and delegates complex task creation to sub-skills (ln-301, ln-302).\n
  • Sanitization: Includes structural validation using regex to verify the completeness and state of manual test results.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 27, 2026, 05:41 PM