The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install or execute the @mermaid-js/mermaid-cli package via npm or npx from the official NPM registry to perform diagram validation. This is a well-known tool from a recognized service.
[COMMAND_EXECUTION]: The skill executes various shell commands such as find, grep, awk, and git to discover project structures, identify the technology stack, and extract information for documentation templates.
[DYNAMIC_EXECUTION]: In Step 7, the skill employs a shell script that uses awk to extract Mermaid diagram blocks from generated documentation files into temporary .mmd files, which are then processed by the mmdc CLI tool.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the local repository (e.g., README.md, package manifests, and source code) to automatically fill placeholders in documentation templates.
Ingestion points: Local files are read during the exploration phase (Step 1) and template population phase (Step 5).
Boundary markers: The skill includes 'Anti-hallucination guidelines' requiring the agent to verify claims against the codebase and use quotes for names, which helps ground the output.
Capability inventory: The skill can read/write to the file system and execute shell commands and Node.js packages.
Sanitization: The skill relies on specific grep-based token mapping but does not specify sanitization or escaping of the content extracted from the codebase before it is interpolated into the final documentation files.

docs-init