project-governance

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides a local validation script (scripts/validate-governance.mjs) to be executed via Node.js for structural validation of governance documents. This command checks for duplicate requirement IDs and broken local Markdown links within a specified directory.
  • [PROMPT_INJECTION]: The skill is designed to process external project documentation, creating an attack surface for indirect prompt injection.
  • Ingestion points: Reads repository instructions and existing documentation (SKILL.md).
  • Boundary markers: Explicit instructions to establish context, preserve project terminology, and resolve conflicts through user decisions.
  • Capability inventory: Execution of local structural validation scripts.
  • Sanitization: The validation logic is limited to static regex matching of document IDs and links, avoiding dynamic execution of document content.
  • [SAFE]: The skill defines a clear document authority hierarchy and lifecycle, focusing on durable decisions and avoiding duplication. All script-based operations are local and deterministic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 03:07 AM
Security Audit — agent-trust-hub — project-governance