blueprint-work-order

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute standard development utilities including git (status, log), gh (issue view, create, comment, edit), and jq for JSON manipulation. These operations are restricted to the local repository and the authenticated GitHub account, aligning with its documented purpose.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes external content from GitHub issues and local design documents (PRPs) to generate new work-orders. While it lacks explicit sanitization for these inputs, this is a common characteristic of workflow automation skills and does not represent an active threat in a controlled developer environment.
  • Ingestion points: Reads content from docs/prps/*.md and fetches issue data via gh issue view.
  • Boundary markers: None explicitly used during content interpolation.
  • Capability inventory: Uses Bash for CLI operations and Write for file modifications in the docs/blueprint/ directory.
  • Sanitization: Content is extracted and reformatted into Markdown templates without specific filtering for embedded instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 07:34 AM
Security Audit — agent-trust-hub — blueprint-work-order