agents-md-generator

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill processes local data from knowledge-base/*.md and CHANGES.md to populate instructions in the generated AGENTS.md and CLAUDE.md files, creating a surface for indirect prompt injection. This is the core functionality of the skill. Evidence:
  • Ingestion points: Reads local project documentation and roadmap files.
  • Boundary markers: Absent in the output generation instructions.
  • Capability inventory: Performs file read operations and generates documentation files.
  • Sanitization: Not explicitly defined; however, the skill incorporates a mandatory interactive checkpoint (AskUserQuestion) for the user to review and edit rules before they are finalized.
  • [COMMAND_EXECUTION]: The skill uses a limited set of standard shell commands for project introspection and synchronization checks, including ls, cat, test, and diff. These operations are benign and restricted to verifying the local project state.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 02:28 PM
Security Audit — agent-trust-hub — agents-md-generator