search_flet_docs

Instructions

Implementation Reference

• src/flet_mcp/server.py:12-21 (handler)

  The async function `search_flet_docs` is decorated with @mcp.tool(), defining the tool. It accepts a `query` string and delegates to `docs_fetcher.search_docs(query)`.

  @mcp.tool()
  async def search_flet_docs(query: str) -> list[str]:
      """
      Search the official Flet documentation index for a specific topic or control.
      Always use this first to find the correct file path before calling get_flet_doc.
      
      Args:
          query: The keyword to search for (e.g., 'dropdown', 'navigation', 'layout').
      """
      return await docs_fetcher.search_docs(query)

• src/flet_mcp/server.py:12-21 (registration)

  The tool is registered via the `@mcp.tool()` decorator on the `search_flet_docs` function in the FastMCP server.

  @mcp.tool()
  async def search_flet_docs(query: str) -> list[str]:
      """
      Search the official Flet documentation index for a specific topic or control.
      Always use this first to find the correct file path before calling get_flet_doc.
      
      Args:
          query: The keyword to search for (e.g., 'dropdown', 'navigation', 'layout').
      """
      return await docs_fetcher.search_docs(query)

• src/flet_mcp/server.py:13-20 (schema)

  Type annotations and docstring define the input schema: a single `query: str` parameter (keyword to search for). Return type is `list[str]`.

  async def search_flet_docs(query: str) -> list[str]:
      """
      Search the official Flet documentation index for a specific topic or control.
      Always use this first to find the correct file path before calling get_flet_doc.
      
      Args:
          query: The keyword to search for (e.g., 'dropdown', 'navigation', 'layout').
      """

• src/flet_mcp/services/github_docs.py:73-81 (helper)

  The `search_docs` method on `FletDocsFetcher` is the helper that performs the actual keyword search. It fetches the full docs tree via `get_docs_tree()`, then filters paths containing the query (case-insensitive).

  async def search_docs(self, query: str) -> list[str]:
      """A keyword search over the available document paths."""
      all_docs = await self.get_docs_tree()
      query_lower = query.lower()
  
      # Filter paths that contain the query string 
      # e.g., querying "dropdown" will match "docs/docs/controls/dropdown.md"
      matches = [path for path in all_docs if query_lower in path.lower()]
      return matches

• src/flet_mcp/services/github_docs.py:47-61 (helper)

  The `get_docs_tree` method fetches the Git tree from the flet-dev/flet GitHub repo using the Trees API, returning a flat list of all Markdown file paths under `website/docs/`.

  async def get_docs_tree(self) -> list[str]:
      """Gets a flat list of all Markdown documentation paths in the Flet repo."""
      # The Tree API is the most efficient way to get all files in a repo at once
      repo_api_url = "https://api.github.com/repos/flet-dev/flet/git/trees/main?recursive=1"
      data = await self._fetch_json(repo_api_url)
  
      if not data or "tree" not in data:
          return []
  
      # Filter out everything except markdown files in the docs folder
      doc_paths = [
          item["path"] for item in data["tree"]
          if item["path"].startswith("website/docs/") and item["path"].endswith(".md")
      ]
      return doc_paths