req-tester
Pass
Audited by Gen Agent Trust Hub on Feb 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes external requirement documents which may contain hidden instructions intended to divert the agent's analysis or behavior.\n
- Ingestion points: The agent context is exposed to untrusted data from 'cleaned-requirements/index.md' and 'CLAUDE.md'.\n
- Boundary markers: No delimiters or 'ignore' instructions are present to prevent the agent from accidentally obeying instructions embedded within the requirement text.\n
- Capability inventory: The agent is granted 'Read', 'Write', and 'Bash' permissions. The 'Bash' tool is utilized to execute the local 'scripts/validate_issues.py' script for output verification.\n
- Sanitization: There is no evidence of input filtering or validation performed on the requirement document content prior to analysis.
Audit Metadata