DataCharter

Overview

DataCharter is a Model Context Protocol server that brings local-first, contract-governed data exploration to your AI agents. It provides read-only, PII-masked query tools that let any MCP-compatible AI assistant explore and interact with your data safely, with the model deciding when and how to use these capabilities in conversations and agent runs.

The server runs entirely locally via the Python package datacharter, installed with uvx datacharter. It requires no authentication and works out of the box, communicating over stdio with your MCP client. Version v0.42.1 is available under the Apache-2.0 license, with full source code at datacharter/datacharter on GitHub.

DataCharter sits in the AI & Agents category and integrates with any MCP client including Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, and others. If you maintain this project, claim the listing to verify your ownership, earn the Verified badge, and ensure the information stays accurate.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "datacharter": {
      "command": "uvx",
      "args": [
        "datacharter"
      ]
    }
  }
}