OpenClaw x Nano Banana

You have access to RunAPI task tools for Nano Banana.

Available Nano Banana models:
- nano-banana: 1K 分辨率；快速基线版本（2025 年 8 月）
  endpoints: /api/v1/nano_banana/text_to_image
  request fields: model, prompt, aspect_ratio
- nano-banana-2: 基于 Gemini 3.1 Flash；最佳多角色一致性，4–8 秒
  endpoints: /api/v1/nano_banana/text_to_image
  request fields: model, prompt, aspect_ratio
- nano-banana-2-lite: 最快、成本最低的 Nano Banana 图像模型，适合快速生成与迭代
  endpoints: /api/v1/nano_banana/edit_image, /api/v1/nano_banana/text_to_image
  request fields: model, prompt, aspect_ratio
- nano-banana-edit: 从现有图片进行提示词驱动的编辑
  endpoints: /api/v1/nano_banana/edit_image
  request fields: model, prompt, source_image_urls
- nano-banana-pro: 基于 Gemini 3 Pro；最高 4K 分辨率，最佳文字渲染，10–20 秒
  endpoints: /api/v1/nano_banana/text_to_image
  request fields: model, prompt, aspect_ratio

Use the model ID and endpoint that match the user's request.
Example prompt: 创建一张励志海报，粗体衬线字写上 'Never Give Up'，背景是日落山脉。
/api/v1/nano_banana/text_to_image: Submit the task, poll for its status, then verify the completed output. POST /api/v1/nano_banana/text_to_image，并保存返回的 Task ID。
GET /api/v1/nano_banana/text_to_image/<task-id>，直到状态为 completed 或 failed。
确认 completed 响应包含 API 参考中记录的端点专属输出。
/api/v1/nano_banana/edit_image: Submit the task, poll for its status, then verify the completed output. POST /api/v1/nano_banana/edit_image，并保存返回的 Task ID。
GET /api/v1/nano_banana/edit_image/<task-id>，直到状态为 completed 或 failed。
确认 completed 响应包含 API 参考中记录的端点专属输出。