How to Generate AI Images with OpenClaw (Free MCP Tool)

OpenClaw runs locally and talks to you through messaging apps. It’s useful for a wide range of tasks, but generating images isn’t built in. Adding that capability normally means signing up for an image API, getting a key, and writing an integration. With AgentPatch, you skip all of that — it works as a free AI image generator through the credits you get when you sign up.

Why This Matters

AgentPatch includes a Recraft-powered image generation tool that supports realistic photos, illustrations, and vector-style output. When you connect AgentPatch to OpenClaw, your agent can generate images on demand — directly in your Telegram or Discord chat — without any additional setup on your end.

Recraft is a strong image generation API, particularly for illustrations and design assets. Getting access to it through AgentPatch means you don’t need your own Recraft account.

Setup

Add AgentPatch to ~/.openclaw/openclaw.json:

{
  "mcp": {
    "servers": {
      "agentpatch": {
        "transport": "streamable-http",
        "url": "https://agentpatch.ai/mcp",
        "headers": {
          "Authorization": "Bearer YOUR_API_KEY"
        }
      }
    }
  }
}

Replace YOUR_API_KEY with your actual key from the AgentPatch dashboard. Restart OpenClaw and it discovers all AgentPatch tools automatically.

Example

Once connected, send your OpenClaw bot a message like:

“Generate an image of a cozy coffee shop interior with warm lighting and wooden furniture. Make it look like a realistic photo.”

OpenClaw calls the Recraft image generation tool through AgentPatch and returns the image to your chat. The request goes in, the image comes back — no browser, no separate app, no Recraft dashboard.

You can get specific about style:

“Create a flat vector illustration of a mountain landscape with a river in the foreground. Use a limited color palette — greens, blues, and warm oranges.”

OpenClaw handles the style prompt and returns the image. Recraft does well with this kind of stylistic direction.

You can also chain it with other tools in the same request:

“Search for popular hiking spots near Portland, then generate a scenic illustration of one of them.”

OpenClaw uses Google Search (also available through AgentPatch) to find the location, then calls image generation to create the illustration.

Wrapping Up

Image generation is one of many tools available through AgentPatch. Once connected, your OpenClaw agent also has access to web search, email, Google Maps, and more — all from the same configuration. Browse the full catalog at agentpatch.ai.