MCP Server Spinnaker

🚀 Spinnaker's MCP Server

This package provides an implementation of the Model Context Protocol (MCP) server for Spinnaker integration. It enables AI models to interact with Spinnaker's deployments, pipelines, and applications through the standardized MCP interface.

🚀 Quick Start

This MCP server demonstrates how Anthropic's new AI model, Claude, can be directly integrated into and enhance the software deployment process, adhering to the MCP standard. Claude can access rich context information, understand the status of Spinnaker applications, pipelines, and deployments, and actively manage them through well - defined tools.

Let's explore some exciting possibilities of AI - driven CI/CD:

1. Intelligent Deployment Decisions: With a comprehensive understanding of application and pipeline status, AI models can analyze and make informed decisions on when and where to deploy. For example, Claude can review test coverage, code changes, and historical success rates to determine the optimal deployment time and target environment.
2. Proactive Problem Detection and Automatic Repair: AI continuously monitors the CI/CD process to detect potential issues in advance. Imagine Claude discovering a known vulnerability in a new dependency version and automatically creating a pull request for an update, or noticing a long - running deployment and proactively adding resources to prevent timeouts.
3. Continuous Process Optimization: After each deployment, AI can learn and adapt to continuously optimize the CI/CD process. Claude can analyze build and deployment logs, identify bottlenecks, and propose improvement suggestions.
4. Context Update: The server automatically maintains context information about Spinnaker deployments, including the list of applications and their statuses, pipeline statuses, current deployments, and recent pipeline executions.

📦 Installation

Install using npm or yarn:

npm install @airjesus17/mcp-server-spinnaker
# Or
yarn add @airjesus17/mcp-server-spinnaker

💻 Usage Examples

Basic Usage

import { SpinnakerMCPServer } from '@airjesus17/mcp-server-spinnaker';

// Create an MCP server instance
const server = new SpinnakerMCPServer({
  gateUrl: 'http://localhost:8050', // Required: Spinnaker Gate service URL
  port: 3000, // Optional: MCP server port, default is 3000
});

// Start the server
await server.start();

Advanced Usage

The following tools are provided:

• getApplications(): Get all Spinnaker applications.
• getPipelines(applicationId: string): Get all pipelines of a specified application.
• getDeployments(environment: string): Get all deployments in a specified environment.

📚 Documentation

Environment Variable Configuration

The server can be configured through the following environment variables:

• GATE_URL: Spinnaker Gate service URL
• MCP_PORT: Port on which the MCP server runs (Default: 3000)
• REFRESH_INTERVAL: Context refresh interval (seconds, Default: 30)

Type Definitions

The package exports TypeScript types for use:

import type {
  SpinnakerApplication,
  SpinnakerPipeline,
  SpinnakerDeployment,
  SpinnakerExecution
} from '@airjesus17/mcp-server-spinnaker';

Development

To contribute code:

1. Clone the repository.
2. Install dependencies: yarn install
3. Build the project: yarn build
4. Run tests: yarn test

📄 License

MIT License - See LICENSE for details.