teamcity
Model Context Protocol (MCP) server for JetBrains TeamCity: control builds, tests, agents and configs from AI coding assistants.
claude mcp add --transport stdio daghis-teamcity-mcp npx -y @daghis/teamcity-mcp
How to use
The TeamCity MCP Server provides a bridge between AI coding assistants and a JetBrains TeamCity CI/CD server. It exposes TeamCity operations as MCP tools that can be invoked from supported AI assistants, enabling actions such as triggering builds, fetching build logs, inspecting test failures, and managing configurations and parameters. The server supports two runtime modes: Dev (default) for safe CI/CD operations with a subset of tools, and Full for complete infrastructure management. You can switch modes at runtime using get_mcp_mode and set_mcp_mode when supported by the client, and the tool list may update dynamically. To use, configure the MCP server (via environment variables or client-provided arguments) and start the server; then use your MCP-enabled client to call tools like triggering a build, listing projects, or querying queues, all backed by the TeamCity REST API with token-based authentication and redacted sensitive values in logs.
How to install
Prerequisites:
- Node.js 20.x (LTS versions 20, 22, 24 tested)
- Access to a TeamCity Server 2020.1+ with REST API and an authentication token
Install steps:
-
Clone the repository git clone https://github.com/Daghis/teamcity-mcp.git cd teamcity-mcp
-
Install dependencies npm install
-
Configure environment (example): cp .env.example .env
Edit .env to set TEAMCITY_URL and TEAMCITY_TOKEN, plus optional MCP_MODE
-
Run in development mode npm run dev
-
Run via npx (one-off) with required env vars (inline): TEAMCITY_URL="https://teamcity.example.com"
TEAMCITY_TOKEN="<your_token>"
MCP_MODE=dev
npx -y @daghis/teamcity-mcp -
Alternatively, rely on a local .env file: npx -y @daghis/teamcity-mcp
Additional notes
Tips and common considerations:
- Ensure TeamCity_URL and TEAMCITY_TOKEN are correctly set; token-based authentication is required for API access.
- You can configure MCP_MODE to switch between Dev (default, ~14k tokens) and Full (~26k tokens) tool sets; runtime switching is supported in newer MCP clients.
- If you encounter environment-variable merging issues on Windows with Claude Code, prefer using explicit CLI arguments for tool configuration.
- The server uses a centralized environment validation (via Zod) and supports a variety of optional advanced TeamCity options (timeouts, retries, pagination, circuit breaker).
- For security, avoid exposing tokens in process listings; prefer using a .env file or config-based invocation when possible.
Related MCP Servers
after-effects
MCP Server for Adobe After Effects. Enables remote control (compositions, text, shapes, solids, properties) via the Model Context Protocol using ExtendScript.
mcp-ssh
🔐 SSH MCP Tool - AI-powered SSH management through MCP protocol | 基于MCP协议的SSH工具,为AI提供SSH远程操作能力
architect
A powerful, self-extending MCP server for dynamic AI tool orchestration. Features sandboxed JS execution, capability-based security, automated rate limiting, marketplace integration, and a built-in monitoring dashboard. Built for the Model Context Protocol (MCP).
jenkins -enterprise
The most advanced Jenkins MCP server available - Enterprise debugging, multi-instance management, AI-powered failure analysis, vector search, and configurable diagnostics for complex CI/CD pipelines.
mcp-agent
Lightweight, focused utilities to manage connections and execute MCP tools with minimal integration effort. Use it to directly call tools or build simple agents within your current architecture.
warp-sql
🗄️ Model Context Protocol (MCP) server for SQL Server integration with Warp terminal. Execute queries, explore schemas, export data, and analyze performance with natural language commands.