jenkins-mcp
MCP server for Jenkins to inspect builds, control jobs, and manage pipeline configuration.
README
jenkins-mcp
A Model Context Protocol (MCP) server for Jenkins. Inspect builds, control jobs, and manage pipeline configuration — designed to pair with atlassian-mcp so you can ask "did the build pass?" alongside Jira and Bitbucket questions.
Tools
Inspection
| Tool | Description |
|---|---|
jenkins_get_build |
Status of a build by number, by SHA (auto-finds the build for a commit), or the latest. Includes Pipeline stage breakdown, parameters, changeset, and artifact URLs. |
jenkins_get_log |
Console log for a build. Pass stageId for just one Pipeline stage; defaults to a 200KB tail (use full=true to opt in to the entire log). |
jenkins_get_tests |
Test report — failing test names, classes, and stack traces. Far more useful than grepping the console log when a build fails. |
jenkins_list_builds |
Recent builds for a job, with result + duration + commit. |
jenkins_list_jobs |
Browse jobs at the root or inside a folder. |
jenkins_get_queue |
Items currently waiting in the build queue. |
Build control
| Tool | Description |
|---|---|
jenkins_trigger_build |
Trigger a new build, optionally with parameters. Returns a queue item URL to track progress. |
jenkins_stop_build |
Abort a running build by number. |
Job configuration
| Tool | Description |
|---|---|
jenkins_get_job_config |
Fetch a job's configuration as structured JSON — pipeline definition, triggers, parameters, and retention settings. |
jenkins_update_job_config |
Patch a job's configuration. Pass only the fields to change; everything else is preserved. Supports description, pipeline script/path, triggers, parameters, and build retention. |
Natural-language examples
- "did CI pass for this commit?" →
jenkins_get_buildwithsha=<HEAD> - "why did build #42 fail?" →
jenkins_get_tests(failing tests + stack traces) orjenkins_get_logwithstageId=<id>for the failing Pipeline stage - "show me the last 10 builds of master" →
jenkins_list_builds - "what jobs exist under platform/api?" →
jenkins_list_jobswithfolder=platform/api - "what's waiting in the queue?" →
jenkins_get_queue - "trigger a build with ENV=production" →
jenkins_trigger_buildwithparameters - "stop the current build" →
jenkins_stop_buildwithbuildNumber - "add a nightly cron trigger to this job" →
jenkins_get_job_configthenjenkins_update_job_configwith updatedtriggers
Setup
1. Generate a Jenkins API token
- In Jenkins, click your username (top-right) → Configure.
- In the API Token section, click Add new token, give it a name, and click Generate.
- Copy the token value — it is only shown once.
2. Create a config file
Create ~/.jenkins-mcp.json:
{
"$schema": "https://raw.githubusercontent.com/stubbedev/jenkins-mcp/master/jenkins-mcp.schema.json",
"url": "https://jenkins.example.com",
"username": "your-jenkins-login",
"token": "your-api-token",
"repoJobMap": {
"stubbedev/atlassian-mcp": "atlassian-mcp/master",
"kontainer/api": ["kontainer-api/master", "kontainer-api/PR-builds"]
}
}
The $schema field is optional but enables editor autocomplete and validation.
repoJobMap keys are matched as case-insensitive substrings against the git remote URL of the current repo, so the server can resolve jobPath automatically when you call jenkins_get_build from inside a repo. Values can be a single job path or an array. Job paths use slashes for nested folders (e.g. folder/sub-folder/job-name).
Alternatively, use environment variables (or a .env file in this directory):
JENKINS_URL=https://jenkins.example.com
JENKINS_USERNAME=your-jenkins-login
JENKINS_TOKEN=your-api-token
Config is resolved in this order: --config <path> CLI arg → JENKINS_MCP_CONFIG env var → ~/.jenkins-mcp.json → .jenkins-mcp.json in cwd → environment variables.
3. Connect to your AI tool
No cloning or building required — point your tool at npx @stubbedev/jenkins-mcp@latest and it will install and run automatically.
Claude Code
claude mcp add jenkins -- npx -y @stubbedev/jenkins-mcp@latest --config ~/.jenkins-mcp.json
Cursor
Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (project-only):
{
"mcpServers": {
"jenkins": {
"command": "npx",
"args": ["-y", "@stubbedev/jenkins-mcp@latest", "--config", "/Users/you/.jenkins-mcp.json"]
}
}
}
Windsurf
Add to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"jenkins": {
"command": "npx",
"args": ["-y", "@stubbedev/jenkins-mcp@latest", "--config", "/Users/you/.jenkins-mcp.json"]
}
}
}
Zed
Add to ~/.config/zed/settings.json:
{
"context_servers": {
"jenkins": {
"command": {
"path": "npx",
"args": ["-y", "@stubbedev/jenkins-mcp@latest", "--config", "/home/you/.jenkins-mcp.json"]
}
}
}
}
OpenCode
Add to opencode.json in your project root (or ~/.config/opencode/opencode.json for global):
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"jenkins": {
"type": "local",
"command": ["npx", "-y", "@stubbedev/jenkins-mcp@latest", "--config", "/home/you/.jenkins-mcp.json"]
}
}
}
Codex CLI
Add to ~/.codex/config.yaml:
mcpServers:
jenkins:
command: npx
args:
- -y
- @stubbedev/jenkins-mcp@latest
- --config
- /home/you/.jenkins-mcp.json
Any other MCP-compatible tool
Most tools that support MCP accept the same JSON format. Use npx as the command with ["-y", "@stubbedev/jenkins-mcp@latest", "--config", "/path/to/config.json"] as the args.
Updating existing installs
If your MCP client is already configured and you want the newest package version:
npx clear-npx-cache
Then restart your MCP client.
Manual install (optional)
If you prefer to clone and run locally:
git clone git@github.com:stubbedev/jenkins-mcp.git
cd jenkins-mcp
npm install
npm run build
Then use node /path/to/jenkins-mcp/dist/index.js instead of the npx command in the configs above.
Notes
- Authentication is HTTP Basic with
username:apitoken(Jenkins's standard mechanism). - Write operations (
jenkins_trigger_build,jenkins_stop_build,jenkins_update_job_config) require the API token user to have the appropriate Jenkins permissions. - CSRF crumb handling is automatic — API tokens bypass the crumb check on most Jenkins installations, but the server fetches and caches a crumb for installations that require it.
jenkins_get_buildwithshascans the last 30 builds of the job for a matchinglastBuiltRevision. IncreasescanLimitfor repos with many builds per commit.- Pipeline stage breakdown comes from Jenkins's
wfapi(workflow plugin). Freestyle jobs don't expose it, and the tool falls back to summary fields only. - Test reports come from the standard Jenkins JUnit / xUnit plugins. Jobs that don't publish results return a friendly 404 message.
Releases (Maintainers)
This package is published to npm as @stubbedev/jenkins-mcp.
Use semantic versioning for releases. Breaking tool-surface changes should bump the minor version while <1.0.0 (for example 0.0.x -> 0.1.0).
Automatic publish is configured in .github/workflows/publish.yml and runs when a new version tag is pushed.
Release flow:
# choose one: patch | minor | major
increment=patch
# bumps package.json + package-lock.json,
# creates a version commit, and creates a git tag (for example v0.1.17)
npm version "$increment"
# push commit and tag to GitHub
git push origin HEAD --follow-tags
GitHub Actions will publish the npm release from that pushed tag.
- The workflow is configured for npm Trusted Publisher (OIDC), so no
NPM_TOKENsecret is required
Required npm setup (one-time):
- In npm package settings, add this GitHub repo/workflow as a Trusted Publisher
Development
# Watch mode — recompiles on file changes
npm run dev
# Run the built server directly
node dist/index.js
# Test the tool list
echo '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}' | node dist/index.js
# Quick release smoke check
npm run smoke
To use a specific config file:
node dist/index.js --config /path/to/config.json
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。