Vectra MCP Server
@theVuArena
An MCP server providing tools to manage and query a Vectra knowledge base, enabling integration with MCP clients via a backend API.
概要
What is Vectra MCP Server?
Vectra MCP Server is a TypeScript-based MCP server that provides tools to manage and query a Vectra knowledge base, enabling integration with MCP-compatible clients. It interacts with a backend Vectra API.
How to use Vectra MCP Server?
Install dependencies with npm install, build with npm run build, then run the server with node build/index.js. For development, use npm run watch for auto-rebuild. The server communicates over stdio.
Key features of Vectra MCP Server
- Create and list Vectra collections.
- Embed texts in batch with optional metadata.
- Embed local files into Vectra.
- Query collections using hybrid and graph search.
- Add, list, and delete files in collections.
- Fetch ArangoDB nodes directly by key.
Use cases of Vectra MCP Server
- Integrate a Vectra knowledge base with MCP-compatible AI assistants.
- Embed large batches of documents and query with hybrid search.
- Manage collections and files programmatically through an MCP interface.
FAQ from Vectra MCP Server
What tools does Vectra MCP Server expose?
It exposes tools for collection management, embedding text and files, querying collections, file operations, and direct ArangoDB node access.
How do I build and run the server?
Run npm install, then npm run build, then node build/index.js. For development, use npm run watch for auto-rebuilding.
Does the query tool support hybrid search?
Yes, query_collection always uses hybrid search (vector + keyword) and enables graph search enhancement by default.
Can I fetch data directly from the underlying database?
Yes, the get_arangodb_node tool fetches a specific node from the underlying ArangoDB database by its key.
