The Agent Skills Directory

[PROMPT_INJECTION]: The references/upgrade.md file defines a multi-step workflow for the agent to fetch external documentation and execute maintenance commands, which constitutes an indirect prompt injection surface.
Ingestion points: The agent is directed to read package.json for version detection and fetch external migration documentation from nextjs.org.
Boundary markers: No explicit instructions are provided to the agent to delimit or ignore potential instructions embedded within the fetched documentation content.
Capability inventory: The skill encourages the execution of npx codemods and describes interaction with a Model Context Protocol (MCP) server for tool-based debugging.
Sanitization: There is no mention of sanitizing or validating the content of the external documentation before it influences the agent's actions.
[EXTERNAL_DOWNLOADS]: The skill provides instructions for fetching documentation and configuration guidelines from official Next.js and Vercel-managed domains.
[REMOTE_CODE_EXECUTION]: Guidance is provided for running npx @next/codemod@latest, which downloads and executes remote scripts from the official Next.js registry.
[COMMAND_EXECUTION]: Contains numerous examples of CLI commands for development and deployment, including npm install, docker build, and pm2 management.

next-patterns