project-requirements-management
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a bundled Python script
scripts/validate-deliverable.pyto validate the structure of project artifacts. This is a localized tool use pattern. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes external documents which serves as a potential attack surface for indirect prompt injection. The risk is mitigated by the validation script's static analysis approach, which checks for headers rather than interpreting content.
- Ingestion points: Paths to user-created requirements documents in
SKILL.md(Step 4, 5, and 7). - Boundary markers: None explicitly defined in the instructions for reading input files.
- Capability inventory: File reading and local script execution (
python3) to validate Markdown structure. - Sanitization: The validation script uses regular expression escaping for heading checks, which limits the potential for regex-based manipulation.
Audit Metadata