documentation-generation

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by design as it parses and processes content from external, potentially untrusted codebases.
  • Ingestion points: SKILL.md specifies that the agent extracts information from code, configurations, and comments. resources/implementation-playbook.md provides code examples for parsing Python files and Pydantic schemas.
  • Boundary markers: The instructions lack explicit delimiters or guidance to the agent to disregard instructions found within the processed source material.
  • Capability inventory: The skill is focused on documentation generation (Markdown, HTML, YAML) and does not demonstrate capabilities for executing the analyzed content on the agent host.
  • Sanitization: There is no evidence of input validation or sanitization to prevent malicious instructions in the code from affecting the agent's behavior.
  • [EXTERNAL_DOWNLOADS]: Templates within the resource files reference assets and tools from well-known services.
  • Evidence: resources/implementation-playbook.md includes HTML templates that load Swagger UI components from the jsDelivr CDN and references official GitHub Actions for environment setup and deployment.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:27 AM