> ## Documentation Index
> Fetch the complete documentation index at: https://docs.githits.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Quick Start

> Connect your AI coding tool to indexed public open-source code, package docs, metadata, vulnerabilities, changelogs, and examples.

GitHits works by running a local MCP server that your AI coding tool starts on demand.

GitHits gives AI coding agents version-aware access to indexed public open-source code, package docs, package metadata, vulnerabilities, changelogs, dependency graphs, upgrade evidence, and real implementation examples.

<span className="githits-beta-note">Beta: tool names, setup flows, and result formats may change as GitHits evolves.</span>

<Steps>
  <Step title="Check prerequisites">
    GitHits requires **Node.js 20 or later**. To check your version, run:

    ```bash theme={null}
    node --version
    ```

    If you need to install or upgrade Node.js, visit [nodejs.org](https://nodejs.org).
  </Step>

  <Step title="Run the hybrid CLI">
    From your terminal, run:

    ```bash theme={null}
    npx githits@latest init
    ```

    This command:

    1. Detects supported AI coding tools installed on your machine
    2. Configures each detected tool with the GitHits MCP server
    3. Handles authentication automatically for common local setup

    `init` supports automatic install for Claude Code, Cursor, Windsurf, VS Code / Copilot, Cline, Claude Desktop, Codex CLI, Pi, Gemini CLI, Google Antigravity, and OpenCode.

    <Note>
      If you use any MCP-capable client that supports remote servers, connect it to GitHits manually. See [Manual remote MCP setup](/installation/manual-setup).
    </Note>
  </Step>

  <Step title="Use your agent">
    Open your AI coding tool and start a coding task as normal. Many agents call GitHits tools on their own when they need indexed source, docs, package metadata, or implementation examples.

    Some harnesses are more conservative. If the agent starts guessing or does not call GitHits, ask for it directly:

    ```text theme={null}
    Use GitHits Code Navigation to inspect the package version used in this repo. Search for the relevant API or error, read the source lines that explain the behavior, then summarize the finding before editing code.
    ```

    For example, if you ask your agent to debug Express middleware behavior, it may search the indexed package and read the relevant source:

    ```
    Tool call: search
    Query: "middleware error handling"
    Target: "npm:express"

    Tool call: code_read
    Target: "npm:express"
    Path: "lib/router/index.js"
    ```

    The agent receives indexed source context and uses it to explain the dependency behavior before changing your code.
  </Step>
</Steps>

## What's next

* Learn where GitHits fits in an [agentic coding workflow](/guides/agentic-workflow)
* Learn how to [trigger GitHits manually](/guides/trigger-githits)
* Review [Automatic local MCP setup](/installation/automatic-setup) for the local MCP install path
* Add GitHits to any MCP-capable client that supports remote servers with [Manual remote MCP setup](/installation/manual-setup)
* Use GitHits through [CLI-backed Skills](/installation/skills) if you prefer Skills over MCP tools
* Explore [code navigation tools](/tools/code-navigation) for source reading
* Explore [documentation access tools](/tools/documentation-access) for hosted and repo-backed docs
* Explore [package inspection tools](/tools/package-inspection) for vulnerabilities, dependencies, and changelogs
* Browse the [code examples tools](/tools/code-examples) for prior art and implementation-pattern research