Skip to main content
Glama

mcp-azure-devops

CI License: MIT npm

A Model Context Protocol server for Azure DevOps. It lets an MCP-capable client (Claude Desktop, Claude Code, Cursor, Codex, …) work across boards, repos, pipelines, and projects — with a governance layer that keeps an AI agent inside safe boundaries.

What this offers

  • Projects & teams — list projects/teams/members, inspect process templates.

  • Boards / work items — WIQL queries, get work items, create/update work items, list iterations (sprints).

  • Repos / pull requests — list repositories and branches, list/get PRs, open PRs.

  • Pipelines / builds — list pipelines and recent builds, get a build, queue a run.

  • Admin / process — list process templates, create a project, delete a project (guarded).

  • Governance built in — access modes, project allowlists, protected projects, delete gating, typed confirmation for high-impact ops, dry-run, and JSON audit logging.

Related MCP server: Azure DevOps MCP Server

Governance & security model

Concern

Flag

Default

Effect

What can the server do?

AZDO_MODE

read-only

read-only → reads; read-write → work items, PRs, pipeline runs; admin → create/delete project. Tools above the mode are never registered.

Which projects are in scope?

AZDO_PROJECT_ALLOWLIST

(all)

Operations on other projects are refused.

Which projects are read-only forever?

AZDO_PROTECTED_PROJECTS

(none)

Readable, never mutable.

Can it delete?

AZDO_ALLOW_DELETE

false

delete_project needs this and admin mode.

Typed confirmation

AZDO_REQUIRE_CONFIRMATION

true

High-impact ops require confirm to equal the target name — not just a boolean.

Preview

AZDO_DRY_RUN

false

Write/admin tools validate + log intent, then return.

Audit trail

AZDO_AUDIT_LOG

true

JSON line to stderr per guarded operation.

Tools

Read (read-only+): list_projects, get_project, list_teams, list_team_members, list_processes, query_work_items, get_work_item, list_iterations, list_repositories, list_branches, list_pull_requests, get_pull_request, list_pipelines, list_builds, get_build

Write (read-write+): create_work_item, update_work_item, create_pull_request, run_pipeline

Admin (admin): create_project, delete_project (needs AZDO_ALLOW_DELETE + typed confirm)

Quickstart — add to your agent

Published on npm as @dockndevai/mcp-azure-devops. Runs via npx; needs an org and a PAT. See docs/CLIENTS.md for every client and .env.example for all variables.

Claude Code

claude mcp add azure-devops -e AZDO_ORG_URL="https://dev.azure.com/your-org" -e AZDO_PAT="your-pat" -e AZDO_MODE="read-only" -- npx -y @dockndevai/mcp-azure-devops

Claude Desktop · Cursor · Windsurf

{
  "mcpServers": {
    "azure-devops": {
      "command": "npx",
      "args": ["-y", "@dockndevai/mcp-azure-devops"],
      "env": {
        "AZDO_ORG_URL": "https://dev.azure.com/your-org",
        "AZDO_PAT": "your-pat",
        "AZDO_MODE": "read-only"
      }
    }
  }
}

Example prompts

  • "List active pull requests in the Payments project's api repo"

  • "Query work items assigned to me that are still Active"

  • "Show the last 10 builds in the Web project and which failed"

  • "Create a Bug in Payments titled 'Checkout 500 on retry'" (needs read-write)

Run from source (development)

npm install
npm run build
node dist/index.js   # with the environment variables set

Develop

npm run dev
npm test          # governance policy: modes, allowlists, delete gating, confirmation
npm run typecheck

Publishing

Ships a server.json for the official MCP registry and an mcpName for npm ownership validation. See PUBLISHING.md.

License

MIT

A
license - permissive license
Not graded
quality - not tested
C
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

View all related MCP servers

Related MCP Connectors

  • Runtime permission, approval, and audit layer for AI agent tool execution.

  • Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.

  • Operate Linux, macOS and Windows from your LLM. Every action runs through an auditable allowlist.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/dockndevai/mcp-azure-devops'

If you have feedback or need assistance with the MCP directory API, please join our Discord server