code-documentation

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted codebase files to generate summaries and diagrams.\n
  • Ingestion points: The agent reads source code (Go, TypeScript, React), Docker configs, and database schemas from the target directory to extract documentation details.\n
  • Boundary markers: The skill mandates the use of structured YAML frontmatter, Markdown tables for file listings, and specific Mermaid blocks for diagrams, which help define content boundaries.\n
  • Capability inventory: The agent requires permissions to read files within the codebase and write/modify README.md files to perform the documentation tasks.\n
  • Sanitization: The instructions do not include steps for sanitizing or escaping content extracted from source files, which could potentially contain malicious instructions meant to influence the agent's output.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 05:50 AM