The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches content from external sources to update project documentation.
Downloads version-specific documentation from Vercel's servers using the official @next/codemod package.
Fetches the README and best practices guide from the zhanymkanov/fastapi-best-practices repository on GitHub via network tools like curl.
[COMMAND_EXECUTION]: The skill performs shell-based operations to detect project environments and apply changes.
Executes npx @next/codemod@canary agents-md to process Next.js documentation.
Reads project configuration files including package.json, pyproject.toml, and requirements.txt to auto-detect frameworks.
Writes and appends content to local files such as CLAUDE.md, AGENTS.md, and .gitignore.
[INDIRECT_PROMPT_INJECTION]: The skill implements a workflow that ingests data from untrusted or external sources and writes it to a control file (CLAUDE.md) used to instruct AI agents.
Ingestion points: Fetches content from Vercel's external servers and a third-party GitHub repository.
Boundary markers: The instructions do not specify the use of clear delimiters or 'ignore embedded instructions' warnings for the injected content.
Capability inventory: The skill has the capability to execute shell commands via npx and perform file-system writes across the project root.
Sanitization: There is no explicit logic described for sanitizing or escaping the fetched documentation before it is appended to the project's primary instruction file.

inject-docs