AWS CodePipeline MCP Server

What is AWS CodePipeline MCP Server?

This is a Model Context Protocol (MCP) server that integrates with AWS CodePipeline, allowing you to manage your pipelines through Windsurf and Cascade. It provides a standardized interface for interacting with AWS CodePipeline services, enabling natural language control of pipeline operations.

How to use AWS CodePipeline MCP Server?

Clone the repository, install dependencies with npm install, and configure the .env file with your AWS region and credentials (using AWS profile, SSO, static keys, or IAM role). Build with npm run build, start with npm start, and add the server configuration to your Windsurf MCP config file at ~/.codeium/windsurf/mcp_config.json. Once configured, Cascade can interpret natural language requests like "List all my pipelines" or "Trigger the 'test-build' pipeline".

Key features of AWS CodePipeline MCP Server

• List all CodePipeline pipelines
• Get pipeline state and detailed definitions
• List and trigger pipeline executions
• Approve or reject manual approval actions
• Retry failed stages
• View pipeline execution logs and metrics
• Stop pipeline executions
• Tag pipeline resources
• Create webhooks for automatic triggering

Use cases of AWS CodePipeline MCP Server

• List all pipelines and check their current state
• View execution history and logs for a specific pipeline
• Trigger a new pipeline execution on demand
• Approve or reject a manual approval action in a stage
• Create a GitHub webhook to automatically start a pipeline on pushes

FAQ from AWS CodePipeline MCP Server

What prerequisites are needed to run the server?

Node.js v14 or later, an AWS account with CodePipeline access, AWS credentials with permissions for CodePipeline and CloudWatch (for metrics), and the Windsurf IDE with Cascade AI assistant.

How do I configure AWS credentials?

You can use an AWS profile (recommended for local dev), AWS SSO, static access keys in the .env file, or an IAM role if running on EC2/ECS/Lambda/EKS. Set AWS_PROFILE and AWS_REGION in .env or in the Windsurf MCP env block.

How do I integrate the server with Windsurf?

Add a server entry to ~/.codeium/windsurf/mcp_config.json using the command and args pointing to the built dist/index.js, and set the required AWS environment variables. Then restart Windsurf.

What operations can I perform on pipelines?

You can list pipelines, get pipeline state and details, list and trigger executions, approve/reject manual approvals, retry failed stages, view execution logs, stop executions, tag resources, create webhooks, and get pipeline metrics.

How can I troubleshoot common issues?

Ensure the server is running, check AWS credentials (verify with aws sts get-caller-identity), confirm the mcp_config.json file format is correct, and restart Windsurf after making changes. Use DEBUG=* npm start for verbose logging.