prd-to-issues
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh) to interact with repository data and create issues. These commands (gh repo view, gh label list, gh issue create) are standard for project management automation and are used within the context of the user's repository.
- [PROMPT_INJECTION]: The skill processes potentially untrusted data from PRDs, plan files, and GitHub issues, creating a surface for indirect prompt injection. This risk is addressed by the mandatory human-approval step (Step 4) where the user reviews the proposed breakdown before issue creation. * Ingestion points: PRDs and plans from file paths (e.g., docs/plans/) or GitHub issue references. * Boundary markers: No specific delimiters are used to separate ingested plan text from agent instructions. * Capability inventory: GitHub CLI for creating issues (gh issue create). * Sanitization: No explicit content sanitization is described in the logic.
Audit Metadata