Skip to main content
Glama
README.md
# Dovetail MCP Server

A Model Context Protocol (MCP) server for connecting AI tools to the Dovetail API.

> Note: This is the npm package distribution of the official Dovetail MCP server. The canonical source lives in the [Dovetail MCP GitHub repository](https://github.com/dovetail/dovetail-mcp).

## Disclaimer

This package may not always reflect the very latest changes from the upstream repository. As of now, it corresponds to upstream Release 0.3. For the most up-to-date code and releases, please refer to the repository linked above.

## Prerequisites

- Node.js 22 or higher (download Node from [nodejs.org](https://nodejs.org/))
- A Dovetail API token (see [Dovetail API Documentation](https://developers.dovetail.com/docs/introduction) for details on obtaining one)

## Get a Dovetail API token

1. In Dovetail, go to Settings → Account → Personal API keys.
2. Generate a new token and copy it.
3. Set it in your environment as `DOVETAIL_API_TOKEN`.

For details, see Dovetail’s MCP server docs: [docs.dovetail.com/integrations/mcp-server](https://docs.dovetail.com/integrations/mcp-server).

## Configuration

Use this configuration style in MCP-compatible clients (e.g., Claude Desktop, Cursor). It invokes the server via `npx` and passes your API token via environment variable.

```
{
  "mcpServers": {
    "dovetail-mcp-server": {
      "command": "npx",
      "args": ["-y", "dovetail-mcp-server"],
      "env": {
        "DOVETAIL_API_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}
```

## License

MIT


TDQS

B3.4/5.0

Scored across 7 tools

Disambiguation4/5

Most tools have distinct purposes, but get_project_data and list_project_data could be confused since both involve project data; descriptions clarify but potential ambiguity remains.

Naming Consistency5/5

All tools use consistent snake_case with 'get_' for single items and 'list_' for collections, following a predictable pattern.

Tool Count5/5

7 tools is well-scoped for a research platform, covering projects, data, insights, and export without being too few or too many.

Completeness4/5

Covers core read operations (list, get, export) but lacks a global search across projects or data; minor gaps that agents can work around.

Maintenance

ActivityInactive
ResponsivenessNo issues