mcp-azure-tfvc

MCP (Model Context Protocol) server for Azure DevOps Team Foundation Version Control — read-only access.

Lets AI assistants (Claude Desktop, Gemini, etc.) browse TFVC source trees, read file contents, and inspect changesets via structured tools.

Setup

1. Install dependencies

npm install

2. Configure environment

Copy .env.example to .env and fill in your values:

AZURE_DEVOPS_ORG_URL=https://dev.azure.com/your-org
AZURE_DEVOPS_PAT=your_personal_access_token
AZURE_DEVOPS_PROJECT=YourProjectName

PAT Scope required: Code → Read only.

3. Run the server

node src/index.js

Registering with Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "tfvc": {
      "command": "npx",
      "args": [
        "mcp-azure-tfvc"
      ],
      "env": {
        "AZURE_DEVOPS_ORG_URL": "https://dev.azure.com/your-org",
        "AZURE_DEVOPS_PAT": "your_pat",
        "AZURE_DEVOPS_PROJECT": "YourProject"
      }
    }
  }
}

Testing with MCP Inspector

npx @modelcontextprotocol/inspector node src/index.js

Available Tools

Example Tool Calls

List root folders:

{ "tool": "tfvc_list_items", "arguments": { "scopePath": "$/", "recursionLevel": "OneLevel" } }

Read a file:

{ "tool": "tfvc_get_file_content", "arguments": { "path": "$/ProjectName/filePath" } }

Latest 10 changesets:

{ "tool": "tfvc_list_changesets", "arguments": { "maxCount": 10 } }

Changeset details:

{ "tool": "tfvc_get_changeset", "arguments": { "changesetId": 12345 } }

File history:

{ "tool": "tfvc_get_item_versions", "arguments": { "itemPath": "$/ProjectName/filePath" } }