add-docs-page

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • Command Execution: The skill provides instructions for running several local development commands including make, uv run, python, and mint. These tools are used for linting, building the documentation site, and managing redirects. While these commands execute code on the host system, they are standard tools within the documentation workflow described.
  • External Resource Reference: The instructions direct the agent to check documentation on mintlify.com and use tools related to the Mintlify platform. These are well-known services for documentation hosting and are appropriate for the skill's stated purpose.
  • Environment Configuration: The skill mentions running mise trust to allow the execution of local environment configuration hooks. This is a common pattern for development environments to manage tool versions and environment variables.
  • Indirect Prompt Injection: The skill is designed to process and edit MDX (Markdown React) files.
  • Ingestion points: The agent reads and modifies MDX files located in the src/ directory and navigation configuration in src/docs.json.
  • Boundary markers: The skill references specific frontmatter rules and style guides in AGENTS.md to define content structure.
  • Capability inventory: The agent has the ability to run build and linting commands (make, uv, python, mint) which interact with the file system.
  • Sanitization: The process includes automated verification steps like make lint_prose, make check-mdx, and a docs-review skill pass to ensure content follows repository standards.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 05:49 PM