methodology-system

Pass

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches configuration schemas from a trusted repository to validate custom methodology definitions.\n- [COMMAND_EXECUTION]: Documentation provides instructions for standard shell commands (e.g., directory creation) and suggests git operations for workflow checkpoints.\n- [PROMPT_INJECTION]: The skill processes data from local methodology files to generate AI guidance, creating an indirect prompt injection surface.\n
  • Ingestion points: .standards/methodologies/*.yaml and .standards/manifest.json files.\n
  • Boundary markers: No explicit delimiters for interpolated data within guidance templates.\n
  • Capability inventory: File system operations and generation of natural language guidance.\n
  • Sanitization: No explicit validation of natural language instruction strings within the YAML methodology files beyond structural schema validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 20, 2026, 03:01 AM
Security Audit — agent-trust-hub — methodology-system