feature-design-doc

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a bundled Python script (scripts/validate_fdd.py) to perform structural validation on generated documents. The script uses only standard Python libraries and does not perform network operations or access sensitive system files.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted sources including codebase content, session history, and external documentation to generate and fact-check designs.
  • Ingestion points: Local files identified via search tools, user-provided documents, and session conversation history.
  • Boundary markers: The skill does not explicitly use delimiters for ingested code, but it employs [NEEDS INPUT] markers to prevent hallucinating decisions when data is missing.
  • Capability inventory: The skill can write files to the docs/FDD/ directory and execute the bundled validation script.
  • Sanitization: No explicit sanitization of ingested content is performed before interpolation into the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 02:59 PM
Security Audit — agent-trust-hub — feature-design-doc