pagefind-mcp
@tkellogg
About pagefind-mcp
An MCP server for searching static sites that are indexed with pagefind
Basic information
Config
No standard config provided
This server doesn't expose a parseable MCP config block in its README. See the repository for install instructions.
RepositoryTools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is pagefind-mcp?
pagefind-mcp is an MCP (Model Context Protocol) server that enables searching static sites that have been indexed with the Pagefind search tool. It is designed for developers and users who want to integrate Pagefind-powered search into MCP-compatible applications or workflows.
How to use pagefind-mcp?
Run the server using npx pagefind-mcp --host <base-url> [--no-resources]. The --host parameter sets the base site URL to index (prefixing with http:// or https:// is optional; https is assumed when omitted). The optional --no-resources flag skips pushing full page resources and instead fetches each result page to provide a short, markup-free snippet.
Key features of pagefind-mcp
- Searches static sites indexed with Pagefind.
- Configurable base site URL via
--host. - Optional snippet-only results via
--no-resources. - Fetches each result page for short snippets.
- Integrates with MCP-based clients.
Use cases of pagefind-mcp
- Search static sites that already have a Pagefind index.
- Retrieve search results with short, markup-free snippets instead of full page resources.
FAQ from pagefind-mcp
What does pagefind-mcp do?
It is an MCP server for searching static sites indexed with Pagefind, returning search results that can include full page resources or short snippets.
How do I run pagefind-mcp?
Use the command npx pagefind-mcp --host <base-url>; optionally add --no-resources to skip full page pushes.
What does the --host parameter do?
It sets the base site URL that the server will index for search queries. If no protocol is provided, https is assumed.
What does the --no-resources flag do?
It skips pushing full page resources and instead fetches each result page to return a short, markup-free snippet.
Comments