Skills
Navigate
skills/thorinside/skills/test-quality-review/Gen Agent Trust Hub

test-quality-review

Pass

Audited by Gen Agent Trust Hub on Oct 31, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill reads and processes test source code from the project environment, which presents a surface for indirect prompt injection. \n
  • Ingestion points: scripts/review.py reads test file contents into the agent context for analysis. \n
  • Boundary markers: Explicit instructions in SKILL.md and scripts/questions.json command the agent to treat the source code as untrusted evidence rather than instructions. \n
  • Capability inventory: The skill is capable of reading and writing files within the project root and can be used to send data to an external decision API. \n
  • Sanitization: Data is structured into JSON packets for review. \n- [COMMAND_EXECUTION]: The integration test suite in scripts/test_review.py uses subprocess.run to execute the skill's own CLI tool. This execution is limited to internal verification of the skill's logic within the testing environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 31, 2026, 08:36 PM