# gha-intel [Health: Active]

**Category:** 🔄 Version Control  
**Repository:** https://github.com/barissozudogru/gha-intel-mcp  
**GitHub Stars:** 2  
**npm Downloads (last month):** 98  
**Views:** 1  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/gha-intel

## Description
Workflow timing analysis, configuration audit and billing insight for GitHub Actions.

## Tools
Capabilities this server exposes over MCP:

- **list_workflow_performance** — Fetch the last N workflow runs and compute job-level timing statistics (avg, min, max, p95) across those runs. Useful for identifying slow jobs and trends.
- **analyze_workflow_config** — Parse a GitHub Actions workflow YAML (provided as a string) and identify optimization opportunities: missing caches, matrix strategy, concurrency controls, slow dependency installs, artifact handling, and more.
- **get_billing_usage** — Retrieve GitHub Actions billing and cache usage statistics for an owner (user or org), optionally scoped to a specific repository.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "gha-intel": {
    "command": "npx",
    "args": ["-y","@barissozudogru/gha-intel-mcp"],
    "env": {
      "GITHUB_TOKEN": ""
    }
  }
}
```

**Requires environment variables:** `GITHUB_TOKEN` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What gha-intel MCP server does

The gha-intel MCP server provides three tools for inspecting GitHub Actions. It accepts a GitHub token and repository or workflow inputs, then returns structured findings that an MCP-compatible client can use in a conversation.

Its performance analysis tool calculates average, minimum, maximum, and p95 durations for recent workflow runs. Results include job and step timing, overall run timing, and recent run conclusions. The configuration tool reads the complete YAML content of a workflow and reports findings by severity, with recommendations for potential improvements. The usage tool retrieves repository cache consumption and recent run estimates where those GitHub endpoints are available.

## How it works

`list_workflow_performance` requires an owner, repository, and workflow file name or numeric workflow ID. The optional `count` parameter controls how many recent runs are examined, defaulting to 10 and allowing up to 100.

`analyze_workflow_config` works from supplied YAML rather than fetching a file itself. Its checks cover dependency caching, matrix behavior and fail-fast settings, concurrency, artifact uploads, checkout depth, job timeouts, runner pinning, Docker layer caching, and trigger path filters.

`get_billing_usage` requires an owner and optionally accepts a repository. With a repository, it can return cache usage and accessible recent run timing. Account-level Actions billing depends on a retired GitHub endpoint, so those requests may fail.

## Setup and configuration

Node.js 18 or newer is required because the package uses the native `fetch` implementation. The required `GITHUB_TOKEN` environment variable must have access to the repositories and GitHub endpoints being queried. Keep the token outside source control.

For local desktop use, run the package through `npx` as a stdio subprocess. The README provides configurations for Claude Desktop, Claude Code, Cursor, Windsurf, VS Code with Copilot, Cline, Continue, Zed, and JetBrains clients.

HTTP mode is enabled with `--http` or by setting `TRANSPORT=http`; `PORT` selects the listening port and defaults to 3000. The MCP endpoint is `/mcp`, and `/health` provides a health check. A Docker image can also be built and run in HTTP mode with the token passed into the container.

## Tools and capabilities

The gha-intel MCP server exposes:

- `list_workflow_performance` for recent run and job timing statistics.
- `analyze_workflow_config` for YAML checks across nine CI configuration areas.
- `get_billing_usage` for repository cache data and accessible run-based estimates.

## Limitations and notes

Timing results describe only the runs fetched for analysis; they are not forecasts of future execution time. Cost estimates use historical runner rates, and cache utilization assumes a fixed 10 GB limit, so neither should be treated as a current account bill or definitive capacity plan.

Account billing uses GitHub’s retired `/settings/billing/actions` endpoint. Broader token permissions cannot restore that API. Repository cache and run timing use separate endpoints and may still work.

HTTP mode listens on all interfaces and has no built-in client authentication. Keep it on a trusted network or place an authenticated gateway in front of it. Client requests use the server’s configured GitHub token.

_Full upstream README: https://allmcps.com/mcp/gha-intel/readme_