mcp-server-circleci

Official
Local
76
Signed
GitHub Repo

Overview

The mcp-server-circleci is a Model Context Protocol (MCP) server that enables AI assistants and agents to interact directly with CircleCI, the continuous integration and delivery platform. It allows AI-driven workflows to inspect pipelines, jobs, workflows, and build results — bringing CI/CD context directly into conversational and automated AI workflows without switching tools.

This server is especially useful for build monitoring, failure analysis, release workflows, and developer productivity scenarios where understanding CI state and history is critical.

Transport

stdio

Tools

  • get_build_failure_logs
  • find_flaky_tests
  • get_latest_pipeline_status
  • get_job_test_results
  • config_helper
  • create_prompt_template
  • recommend_prompt_template_tests
  • list_followed_projects
  • run_pipeline
  • run_rollback_pipeline
  • rerun_workflow
  • analyze_diff
  • list_component_versions
  • download_usage_api_data
  • find_underused_resource_classes

Key Capabilities

  • Pipeline visibility — Explore pipelines, workflows, and jobs programmatically.
  • Build and job insight — Inspect execution status, timing, and metadata for builds.
  • Failure analysis support — Enable AI assistants to summarize failures and identify patterns across runs.
  • Release and deployment awareness — Reason about CI state as part of broader delivery workflows.
  • Developer productivity — Help assistants answer questions like “why did this build fail?” or “what’s running now?”

How It Works

The mcp-server-circleci runs as an MCP service and connects to CircleCI using an authenticated API token with appropriate scopes. AI clients communicate with the server over the MCP protocol to request CI/CD context as part of broader reasoning workflows.

The server mediates access to CircleCI’s APIs, handling authentication, request execution, and response normalization. Results are returned in structured formats that AI assistants can reason over directly, while respecting CircleCI’s project- and organization-level permissions.

By exposing CircleCI through MCP, the server enables AI-driven workflows such as build inspection, failure triage, and pipeline analysis — bringing CI/CD insight into the same AI-assisted environment used for code, issues, and operations.