M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
B

Buildkite MCP Server

Official
Official Buildkite MCP server exposing Buildkite data to AI tooling.
How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #199 of 438
Stars
★ 55 Popular
Transport
stdio (local process)
Runtime
Docker · Prebuilt binary · Go
Credentials
API key / credential required
License
MIT
Last commit
Tools
7
54FMRS · D

The Buildkite MCP Server is an official MCP integration providing stable and secure access to Buildkite data. It suits teams using Buildkite, but requires familiarity with Buildkite and careful token handling.

Strongest · Setup experience 12/20 Weakest · Reliability 8/20

Reliability
8/20
Security and permissions
10/20
Maintenance
12/20
Documentation
12/20
Setup experience
12/20
Why each score
Reliability 8/20
Limited evidence: no tests or actual CI workflow files provided, but README shows a Buildkite CI badge, implying CI. The server is MIT licensed, Go-based, uses stdio transport, and appears well-structured. Since static review and no execution evidence, reliability cannot exceed 12. Deductions for lack of verifiable test results and error handling details.
Security and permissions 10/20
Evidence shows API token passed via environment variable, optional identity header passthrough in HTTP mode, and documentation clearly states header passthrough is not available in stdio mode, with permission checks for log access. No malicious code or gross overreach apparent. However, token via env var may increase leakage risk, and no least-privilege guidance. Overall risk controlled, no red lines crossed, score in 10-14 range; given security design but missing detailed permissions documentation, assign 10.
Maintenance 12/20
Repo has 52 stars, 25 open issues, not archived. README includes build badge indicating CI. Version 0.7.0, MIT license, active development. But no details on contributing guidelines or dependency update strategy. Deductions for unclear issue response and release cadence.
Documentation 12/20
README provides installation and configuration instructions, including environment variables and Docker run examples, and links to full docs. But lacks detailed tool parameter descriptions, limitations, and troubleshooting. Deductions for insufficient documentation depth.
Setup experience 12/20
Clear installation steps provided, including Docker image and command-line run examples, and environment variable configuration. But no client configuration examples and relies on manual API token creation. Static review limits execution verification, so not above 15. Deductions for missing concrete client setup examples.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • Development teams using Buildkite for CI/CD
  • Teams wanting to integrate AI tools into their development workflow
  • DevOps engineers needing quick access to build information

Not for

  • Users not using Buildkite
  • Those needing a graphical interface (MCP server does not provide UI)
  • Scenarios requiring non-official data sources

Required permissions

  • Requires BUILDKITE_API_TOKEN to authenticate with Buildkite API
  • Token permissions determine the scope of accessible data (e.g., read pipelines, builds, logs)

Risks and side effects

  • API token leakage: keep token confidential and only pass in secure environments
  • Job logs may contain sensitive information; ensure caller permissions are correct
  • In HTTP mode, if header passthrough is configured, ensure the reverse proxy authenticates properly

Setup

Before you start

Runtime：Docker · Prebuilt binary · Go

BUILDKITE_API_TOKEN requiredsecret Buildkite API token for authentication, required; get one at https://buildkite.com/user/api-access-tokens.
Other optional settings (2)
BUILDKITE_PASSTHROUGH_HTTP_HEADERS optional Optional, comma-separated names of HTTP headers allowed to pass through to the Buildkite API in HTTP mode (e.g. Authorization).
BUILDKITE_BASE_URL optional Optional, sets the Buildkite API origin; passthrough headers are only forwarded to this origin.
  1. Obtain a Buildkite API token: create one at https://buildkite.com/user/api-access-tokens.
  2. Run the container (recommended): docker run -i --rm -e BUILDKITE_API_TOKEN=your_token ghcr.io/buildkite/buildkite-mcp-server:0.7.0.
  3. Configure your MCP client to use the above command.

Check that it works

Check that tools like list_pipelines appear in your client's tool list, then ask "List my Buildkite pipelines" — returning pipeline names confirms the connection.

Troubleshooting

  1. Ensure BUILDKITE_API_TOKEN is set correctly and valid
  2. Verify the token has the required permissions (e.g., read pipeline, read logs)
  3. When running in a container, ensure `-i` and `--rm` flags are used and environment variables are passed correctly
  4. Check network connectivity to the Buildkite API
  5. In HTTP mode, check that the reverse proxy forwards required headers properly

Things to try

Once connected, you can ask your AI assistant things like:

  • List all pipelines in my Buildkite organization
  • Show the recent build status for a pipeline
  • Get the log for the failed job in a build
  • Show the test results for this pipeline

Tools 7

list_pipelines read-only
List all pipelines in the account
get_pipeline read-only
Get details of a specific pipeline
list_builds read-only
List builds for a pipeline or account
get_build read-only
Get details and status of a specific build
get_job_log read-only
Get the log output for a job
list_jobs read-only
List jobs in a build
list_tests read-only
List test results in a build

Use cases

Let AI assistants query recent build statuses and logs
Quickly identify failing jobs and analyze logs
Monitor build trends across multiple pipelines
Automate build reports

Supported clients

Claude Desktop

Listed from the project's documentation, not tested by this site.

Overview

The Buildkite MCP Server is an official Model Context Protocol (MCP) server that exposes Buildkite API data (pipelines, builds, jobs, tests) to AI tooling and editors. It runs over stdio via a Docker container and requires a BUILDKITE_API_TOKEN for authentication. This server is ideal for development teams who want to integrate AI assistants to query and monitor their CI/CD pipelines securely. It provides a set of tools to list pipelines, builds, jobs, and tests, as well as retrieve job logs.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 62.9k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision cb4c2dc98c40 Data synced 2026-10-12 Read the FMRS scoring method