What problem does it solve? Fetching library documentation through MCP normally dumps large raw responses into the AI context, consuming hundreds of tokens per query. This Skill filters documentation through shell pipelines so only code examples, API signatures, and key notes reach the model, cutting token usage by roughly 77%. ## Core Features & Use Cases - Token-Efficient Fetching: Wraps Context7 MCP calls in a shell pipeline that extracts code blocks, API signatures, and important notes using awk, grep, and sed. - Library Resolution: Resolves library names like React, Next.js, or Prisma to Context7-compatible IDs, or accepts direct IDs for faster lookups. - Flexible Modes and Pagination: Supports code mode for examples, info mode for conceptual explanations, and pagination across up to 10 pages. - Use Case: When a user asks "Show me React useState examples", the Skill fetches the full documentation in a subprocess and returns only about 205 filtered tokens of examples and signatures instead of the full 934-token response. ## Quick Start Ask the assistant to fetch documentation for a library and topic, for example: "Show me React hooks examples using the context7-efficient skill."
