The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its core functionality involves processing untrusted codebase content to generate documentation.
Ingestion points: Project source code files, READMEs, and environment configurations.
Boundary markers: While the skill references external role boundaries, it does not define explicit delimiters or instructions to ignore embedded commands within the ingested content.
Capability inventory: Reading file system data and writing documentation or type definition updates.
Sanitization: The provided documentation does not specify sanitization or filtering procedures for codebase content before interpolation into prompts.
[COMMAND_EXECUTION]: Reference guides within the skill suggest using standard local utilities such as grep, find, and npx for auditing tasks and running documentation generators.
[EXTERNAL_DOWNLOADS]: The skill mentions and provides configuration for well-known development tools available through the NPM registry, including typedoc, swagger-jsdoc, type-coverage, and markdown-link-check.

Quill