ac-qa-reviewer

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION] (LOW): The skill is vulnerable to Indirect Prompt Injection (Category 8) because it processes implementation files that may contain untrusted data.
  • Ingestion points: The review_feature method and 'SCAN' workflow step ingest source code from feature implementations.
  • Boundary markers: The documentation does not define markers or instructions to prevent the model from following commands hidden in code comments or strings.
  • Capability inventory: The skill can execute subprocesses (tests, linting) and write to the filesystem (auto-fix capabilities).
  • Sanitization: No sanitization logic for external code content is described.
  • [COMMAND_EXECUTION] (SAFE): The skill describes executing tests and static analysis tools.
  • While executing arbitrary code from feature branches is high-risk, it is the stated primary purpose of the skill. The risk is considered acceptable within the context of a QA tool provided it is used in a sandboxed environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:30 PM