The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the mint documentation CLI tool (including validate, broken-links, dev, and a11y subcommands) and performs git commits. These operations are standard for a documentation audit and maintenance tool. \n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided documentation files (MDX and JSON) which could contain malicious instructions. The presence of file-writing and command-execution capabilities creates a potential injection surface. \n
Ingestion points: The agent reads the documentation navigation file (docs.json) and individual MDX page files from the project directory. \n
Boundary markers: Absent; there are no instructions to use delimiters or specific warnings to ignore embedded instructions within the documentation files being reviewed. \n
Capability inventory: The skill can perform file-system writes (edit command) and execute shell commands (mint CLI, git commit). \n
Sanitization: Absent; the skill does not specify any filtering or escaping of the content read from the documentation files before processing it.

review-mintlify-docs