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

# Harbor docs for AI agents

> Connect your coding agent to the Harbor documentation with MCP.

## Claude Code

```bash theme={"system"}
claude mcp add --transport http harbor-docs \
  https://docs.harborframework.com/mcp
```

## Codex

```bash theme={"system"}
codex mcp add harbor-docs --url https://docs.harborframework.com/mcp
```

## Cursor

Add this to `~/.cursor/mcp.json`. Cursor CLI reads the same configuration.

```json theme={"system"}
{
  "mcpServers": {
    "harbor-docs": {
      "url": "https://docs.harborframework.com/mcp"
    }
  }
}
```

For other agents, check their documentation for how to configure MCP servers.

## One-time query (Not recommended)

For a one-time query, paste this URL into your coding agent's chat and ask it
to query the Harbor docs. An agent with network access can communicate with it
over MCP's JSON-RPC protocol.

```text theme={"system"}
https://docs.harborframework.com/mcp
```

For regular use, configure your MCP client as shown above.

## Example prompt

```text theme={"system"}
Use the Harbor docs to explain how to run an evaluation.
```