Buildkite MCP server for AI apps

What your AI apps can do with Buildkite

Find pipelines and builds
Read job logs and test results

Buildkite MCP tools

56 tools, taken from the publisher's documentation on 31 October 2026. The server decides what it offers and can add or rename tools at any time; after you connect, Toolstile shows the current list.

Tool	What it does
`access_token`	Returns the current API token's details, such as scopes and UUID.
`current_user`	Returns profile details for the user who owns the token.
`user_token_organization`	Returns the organization tied to the user's token.
`list_clusters`	Lists the organization's clusters with summary details.
`get_cluster`	Gets full details for one cluster.
`create_cluster`	Creates a cluster in the organization.
`update_cluster`	Updates a cluster's name, description, emoji, color or default queue.
`list_cluster_queues`	Lists a cluster's queues with dispatch status and agent settings.
`get_cluster_queue`	Gets one queue's details, including hosted agent configuration.
`create_cluster_queue`	Creates a self-hosted queue in a cluster.
`update_cluster_queue`	Updates a queue's description or retry agent affinity.
`pause_cluster_queue_dispatch`	Stops dispatching new jobs to agents on a queue.
`resume_cluster_queue_dispatch`	Resumes job dispatch on a paused queue.
`list_cluster_secrets`	Lists a cluster's secrets and metadata without values.
`get_cluster_secret`	Gets metadata for one cluster secret.
`create_cluster_secret`	Creates a cluster secret, optionally with a description and access policy.
`list_cache_registries`	Lists a cluster's cache registries with cursor pagination.
`get_cache_registry`	Gets one cache registry, including its policy and default status.
`create_cache_registry`	Creates a cache registry, optionally with metadata and a policy.
`update_cache_registry`	Updates a cache registry's name, description, emoji, color or policy.
`set_default_cache_registry`	Makes another registry the cluster's default.
`delete_cache_registry`	Deletes a non-default cache registry and its cache metadata.
`list_agents`	Lists agents with connection state, host, version, current job and pause status.
`get_agent`	Gets one agent's details, including metadata.
`list_pipelines`	Lists the organization's pipelines with build counts and status.
`get_pipeline`	Gets a pipeline's configuration, steps, environment variables and build stats.
`create_pipeline`	Creates a YAML pipeline with a repository and an assigned cluster.
`update_pipeline`	Updates a pipeline's configuration, repository, settings or metadata.
`list_pipeline_schedules`	Lists a pipeline's schedules with cron, branch, enabled state and next run.
`get_pipeline_schedule`	Gets one pipeline schedule, including last failure and next build time.
`create_pipeline_schedule`	Creates a schedule that triggers builds on a cron interval.
`update_pipeline_schedule`	Updates a schedule's cron, branch, environment variables or enabled state.
`list_builds`	Lists lightweight build summaries for the organization or one pipeline, without jobs.
`get_build`	Gets one build's metadata, without jobs.
`get_build_test_engine_runs`	Gets the Test Engine runs linked to a build.
`create_build`	Creates a build for a commit and branch with optional environment, metadata and author.
`cancel_build`	Cancels a running build.
`rebuild_build`	Reruns an entire build.
`list_jobs`	Lists a build's jobs with optional state filtering and cursor pagination.
`get_job`	Gets a job by UUID, scoped to a build or the organization.
`unblock_job`	Unblocks a blocked job so the build can continue.
`retry_job`	Retries a failed or timed-out job.
`get_job_env`	Gets a job's environment variables.
`search_logs`	Searches job log entries with regex patterns, optionally with surrounding lines.
`tail_logs`	Returns the last N log entries of a job.
`read_logs`	Reads job log entries, optionally from a given row.
`list_artifacts_for_build`	Lists artifacts across all of a build's jobs with paths, sizes, MIME types and download URLs.
`list_artifacts_for_job`	Lists the artifacts for one job.
`get_artifact`	Downloads an artifact's content as base64.
`list_annotations`	Lists the annotations on a build or job.
`create_annotation`	Creates an annotation on a build or job.
`list_tests`	Lists a test suite's tests with metrics over a time range, with filtering, sorting and pagination.
`get_test`	Gets one test, with extra metadata for failed executions.
`list_test_runs`	Lists the runs in a test suite.
`get_test_run`	Gets one test run.
`get_failed_executions`	Returns a run's failed executions, optionally with full errors and stack traces.

Suggested access for each AI app

The documentation this list comes from doesn't mark Buildkite's tools as read-only or destructive. Read each tool's description and decide per app before allowing it.

Coding agent in Claude Code or Cursor. Allow the read tools so the agent can look things up without stopping, and require approval for tools that change shared data.

Chat or research app in Claude Desktop. Allow reading and searching. Allow writes only where you want the app to act for you.

Unattended agent, such as a scheduled Codex job. No one is watching the session, so keep it narrow: allow what the job needs, require approval for changes and block the rest.

Blocked tools are removed from the tool list an app sees, so it won't try them. Because rules apply per app, the same Buildkite connection can be broad in one app and read-only in another, and the credential stays in Toolstile instead of each app's config file.

How it works with Toolstile

Connect once. Add Buildkite to a workspace (browser sign-in). Toolstile keeps the credential, so no agent config holds it.
Choose access per app. Give Claude Code every Buildkite tool, give Cursor only the read tools, or keep Buildkite away from an app entirely.
Share approved accounts. Teammates use the connection without copying tokens, and you can remove access in one place.
Review activity. See which app called which tool and when, so you can tighten access with evidence.

Add the Buildkite MCP server to an AI app

Through Toolstile, every AI app uses one gateway entry, whichever servers you connect. In Claude Code:

Shell

claude mcp add --transport http toolstile 'https://mcp.toolstile.com/mcp'

Then connect Buildkite in your workspace and choose which of its tools that app may use. Client setup covers Codex, Cursor and Claude Desktop.

To connect directly instead, add the Buildkite endpoint to each app separately. In Claude Code:

Shell

claude mcp add --transport http buildkite 'https://mcp.buildkite.com/mcp'

In Cursor's mcp.json:

JSON

{
  "mcpServers": {
    "buildkite": {
      "url": "https://mcp.buildkite.com/mcp"
    }
  }
}

Some servers accept only clients they have approved, or need a token header. Check the Buildkite MCP documentation for what your client needs.

Before you connect

A Buildkite account. Sign-in uses your Buildkite permissions.

Server details

Detail	Value
Publisher	Buildkite
Endpoint	`https://mcp.buildkite.com/mcp`
Authentication	Browser sign-in
Categories	Development
Tool list	56 tools, observed 31 October 2026 (docs)

Buildkite MCP documentation

Buildkite MCP server for Claude, Codex and Cursor