hesprs/harness

Context7

Workflow

Step 1: Search for the Library

To find the Context7 library ID, use the search script:

bun path/to/skill/search-library.ts LIBRARY_NAME TOPIC

Parameters:

• libraryName (required): The library name to search for (e.g., "react", "nextjs", "fastapi", "axios")
• query (required): A description of the topic for relevance ranking
• --type / -t (optional): json (default) for structured output, txt for readable summary

Response fields:

• id: Library identifier for the context endpoint (e.g., /facebook/react), which will be used in next step
• title: Human-readable library name
• description: Brief description of the library
• totalSnippets: Number of documentation snippets available

Step 2: Fetch Documentation

To retrieve documentation, use the fetch script with library ID from step 1:

bun path/to/skill/fetch-docs.ts LIBRARY_ID TOPIC

Parameters:

• libraryId (required): The library ID from search results
• query (required): The specific topic to retrieve documentation for
• --type / -t (optional): Response format - txt (default) or json

If you don't find desired doc info in one invocation, you can repeat this step many times with varying queries. You can also retry step 1 for a more relevant library.

Examples

Next.js routing documentation:

# Find Next.js library ID
bun path/to/skill/search-library.ts nextjs routing

# Fetch app router documentation
bun path/to/skill/fetch-docs.ts /vercel/next.js "app router"

FastAPI dependency injection:

# Find FastAPI library ID
bun path/to/skill/search-library.ts fastapi dependencies

# Fetch dependency injection documentation
bun path/to/skill/fetch-docs.ts /fastapi/fastapi "dependency injection"

Tips

• Be specific with the query parameter to improve relevance ranking
• If the first search result is not correct, check additional results in the array
• Use quoted args for multi-word queries