CI/CD insights in your AI assistant

Ask Claude, Cursor, or Windsurf about your pipelines and get answers grounded in real data. No context-switching, no dashboards — just ask.

How it works

Three steps to give your AI assistant access to your CI/CD data.

1

Create an API key

Generate a read-only API key from your CI/CD Watch dashboard in Settings.

2

Add the MCP config

Paste a small JSON snippet into your AI client's MCP configuration file.

3

Start asking questions

Your assistant now has tools to query pipeline runs, metrics, costs, and performance.

One config, any MCP client

Works with Claude Desktop, Claude Code, Cursor, Windsurf, and any MCP-compatible tool.

mcp config
{
  "mcpServers": {
    "cicd-watch": {
      "type": "http",
      "url": "https://mcp.cicd.watch",
      "headers": {
        "Authorization": "Bearer cw_your_key_here"
      }
    }
  }
}

That's it. No install, no server to run. Just an API key and a URL.

Five tools, full visibility

Your AI assistant gets access to the same data you see in the dashboard.

list-runs

Recent pipeline runs with status, duration, and job details

get-dora-metrics

Deployment Frequency, Lead Time, Change Failure Rate, MTTR

get-costs

Compute costs, developer wait time, and waste breakdown

get-performance

Duration stats, trends, bottlenecks, and optimization suggestions

list-connections

Your connected CI/CD providers and accounts

Just ask

No API calls to write, no dashboards to navigate. Ask in natural language.

>What are my DORA metrics for the last 30 days?
>Which pipelines are the most expensive?
>Show me recent failed builds
>How is our deployment frequency trending?
>What optimization opportunities do we have?

Give your AI assistant CI/CD superpowers

Set up in under two minutes. All you need is an API key.