Vectora

Overview

Vectora is an MCP server that brings advanced AI capabilities to your workflow through retrieval-augmented generation, web search, filesystem access, and persistent memory. It functions as a sub-agent for Claude Code, enabling any MCP-compatible AI assistant to leverage its tools seamlessly during conversations and agent runs.

The server runs locally as a Python package installed via uvx vectora-agent and communicates over stdio. You'll need to provide COHERE_API_KEY and TAVILY_API_KEY as environment variables to enable its search and generation features. The latest release is v0.1.0rc3, with source code available at brunosrz/vectora on GitHub.

Vectora is categorized as AI & Agents on mcp.site and integrates with MCP clients including Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, and others. If you maintain Vectora, claim this listing to verify your ownership, earn a Verified badge, and ensure the information stays accurate.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "vectora": {
      "command": "uvx",
      "args": [
        "vectora-agent"
      ],
      "env": {
        "COHERE_API_KEY": "<cohere-api-key>",
        "TAVILY_API_KEY": "<tavily-api-key>"
      }
    }
  }
}

Requires COHERE_API_KEY, TAVILY_API_KEY