The Agent Skills Directory

[Command Execution Patterns]: The skill uses local commands like grep and jq to search through documentation and process model lists. It also uses curl to fetch a list of available AI models from a centralized gateway. These are standard operations for development tools aimed at providing up-to-date configuration data.
[External Package Management]: The instructions guide users to install standard libraries from the AI SDK ecosystem (e.g., ai, @ai-sdk/react) via established package managers like pnpm. This is a routine part of setting up a development environment for the SDK.
[Data Ingestion Surface]: The skill is designed to search through local node_modules and fetch markdown documentation from official project domains. While this represents an ingestion of external content into the agent's context, it is the primary purpose of the skill to provide accurate, documentation-based answers.
[Environment Configuration]: Reference files mention the use of .env.local for managing API keys, which aligns with security best practices for local secret management during development.

ai-sdk