sonht1109/.github/skills/docs-seeker

docs-seeker Skill

Prerequisites

• Setup CONTEXT7_API_KEY in .env file with your API key from context7.com.
• Setup context7-mcp server running locally.

Steps

1. Detect User Intent: Only use if the user query indicates they are looking for documentation or code examples related to a specific library, framework, or API. Look for keywords like "how to use", "integrate", "setup", "documentation", "API reference", etc. If yes, proceed to step 2. If no, skip this skill.

2. Prepare query: From the user query, extract the library/framework name. Also, update the query to specifically ask for code examples if not already included. Output:

• prepared_query
• library_name

3. Extract library id: Use #tool:context7-mcp/resolve-library-id to get exact library id.

• input:

{
  "query": "<prepared_query>",
  "library": "<library_name>"
}

• output: output will be a list of result object. Eg:

[
  {
    "id": "/facebook/react",
    "name": "React",
    "description": "A JavaScript library for building user interfaces",
    "totalSnippets": 1250,
    "trustScore": 95,
    "benchmarkScore": 88,
    "versions": ["v18.2.0", "v17.0.2"]
  }
]

Get the top 3 results with highest trustScore and benchmarkScore. For each result, extract the id field as library_id to use in next step until you get relevant documentation in step 4.

4. Fetch documentation: Use #tool:context7-mcp/query-docs to get the relevant documentation and code examples.

• input:

{
  "libraryId": "<library_id>",
  "query": "<prepared_query>",
  "type": "json", // or txt
}

5. Format and return results: Format the output from the previous step to return relevant documentation snippets and code examples to the user. Ensure the response is concise and directly addresses the user's query.

Notes

• If fails to use mcp server, inform user to solve issue.
• If fails to get docs or docs return empty, inform user that no relevant documentation was found for their query.