prd-to-issues
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs expected project management tasks such as reading PRD files and creating issues in GitHub or Beads. It includes logic to prevent duplicate issue creation by searching for existing IDs first.
- [COMMAND_EXECUTION]: The skill uses standard command-line tools (
gh,bd,grep,cat) for its primary purpose of issue management. These operations are limited to project tracking and are consistent with the skill's stated intent. - [DATA_EXPOSURE_&_EXFILTRATION]: No sensitive data access or network exfiltration patterns were detected. All network operations are directed at legitimate project management platforms (GitHub) via official CLI tools.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted PRD documents. However, it incorporates strict boundaries: it only creates issues, never re-writes the PRD, and never executes instructions found within the requirements text. It explicitly stops if it encounters unresolved clarification tags, reducing the surface for injection.
Audit Metadata