Vectorize MCP Server is a Model Context Protocol (MCP) server implementation, officially maintained by Vectorize, used to integrate Vectorize's vector retrieval and text extraction capabilities into MCP-compatible clients.
Purpose The server is aimed at workflows requiring "retrieval + document processing + research generation," exposing Vectorize's related API capabilities as MCP tools for clients to call.
Main capabilities According to the repository description, it provides the following capabilities:
- Retrieve documents: perform vector search and retrieve documents (corresponding to the official API).
- Text extraction and chunking (convert any file to Markdown): extract text from documents and chunk it into Markdown format (corresponding to the official API).
- Generate private deep research: generate Private Deep Research from a user pipeline (corresponding to the official API).
Usage
- Run with npx.
- VS Code installation: provides a one-click install button; it can also be installed manually by adding a JSON configuration block to VS Code's user settings (JSON), which can be opened via Ctrl + Shift + P and entering Preferences: Open User Settings (JSON); optionally, the configuration can be added to the .vscode/mcp.json file in the workspace to share it with others.
- Configuration on Claude/Windsurf/Cursor/Cline: the repository provides corresponding configuration instructions.
- Development-related: modify the package.json version, then you can fork the repository, create a feature branch, submit a pull request, and so on.
Applicable scenarios Applicable to scenarios in editors or clients that support MCP, where Vectorize needs to be called for vector retrieval, document-to-Markdown conversion and chunking, and generation of private deep research.
Limitations The materials do not provide more details about fees, pricing, platform support scope, performance metrics, or specific usage steps; the content not mentioned above is not currently provided.
