ClawHub

Apidot Nano Banana 2 Api

Other

Use APIDot for Nano Banana 2 API workflows, including Gemini 3.1 Flash Image API, nano-banana-2, nano-banana-2-edit, text-to-image API, image editing API, readable text image generation, async task submission, task_id handling, polling, task status, and webhook integration based on APIDot docs.

Install

openclaw skills install @jiehao71727/apidot-nano-banana-2-api

APIDot Nano Banana 2 API

Use APIDot as a Nano Banana 2-focused API surface for prompt-based image generation, image editing, readable text image workflows, polling, and webhook delivery.

This skill is for routing Nano Banana 2 questions to the right APIDot docs, examples, and async integration pattern. It is documentation-only: it includes no scripts, makes no network requests, and does not store credentials.

This release contains only SKILL.md. It includes no executable files, install-time automation, review automation helpers, shell automation, bundled API clients, automatic network calls, or stored credentials.

When To Use

Use this skill when the user asks to:

  • Build a Nano Banana 2 API integration with APIDot.
  • Generate images from prompts, product briefs, ad concepts, UI labels, posters, or design directions.
  • Edit or transform source images with reference-guided workflows.
  • Use Gemini 3.1 Flash Image, Nano Banana 2, nano-banana-2, or nano-banana-2-edit through APIDot.
  • Implement APIDot async image jobs with task submission, task status polling, or webhook callbacks.
  • Find APIDot Nano Banana 2 docs, model pages, or runnable examples.

Security Rules

  • Treat APIDOT_API_KEY as a secret.
  • Keep APIDot API keys in server-side environment variables or a backend secret manager.
  • Never place an API key in browser code, frontend bundles, public repos, logs, screenshots, or chat output.
  • Do not make live API calls unless the user explicitly asks and provides a safe server-side environment.
  • Do not invent API facts, commercial terms, model availability, reliability claims, or competitor comparisons.
  • Use current APIDot docs and model pages for model-specific request fields and current product details.

Nano Banana 2 Workflow

APIDot Nano Banana 2 generation usually follows an async task pattern:

  1. Choose the Nano Banana 2 request mode from the current APIDot docs.
  2. Submit the generation or edit request through the documented APIDot async generation flow.
  3. Save the returned data.task_id immediately.
  4. Poll task status with the documented task status API for local tests.
  5. Use callback_url webhook delivery for production queues or user workflows that may outlive the current page.
  6. Store final image URLs only after the task reaches a terminal success state.

Do not guess model-specific payload fields. If the user needs copyable request examples, point them to the current APIDot docs or the matching APIDot GitHub example.

Model Routing

Start from the user's Nano Banana 2 task, then open the matching APIDot source:

User GoalStart Here
Browse APIDot Nano Banana 2 model pagehttps://apidot.ai/models/nano-banana-2
Read APIDot API docshttps://apidot.ai/docs
Learn APIDot quickstart flowhttps://apidot.ai/docs/quickstart
Implement webhookshttps://apidot.ai/docs/webhooks
Build with Nano Banana 2https://apidot.ai/docs/nano-banana-2
Use runnable Nano Banana 2 exampleshttps://github.com/APIDotAI/nano-banana-2-api
Use general APIDot exampleshttps://github.com/APIDotAI/apidot-examples

For Nano Banana 2 request modes, prefer the live APIDot model page and docs page. Do not copy request fields from another image model family unless the APIDot docs show the same field.

Integration Guidance

  • Use apidot-image-generation-api when the user needs broad image generation guidance across several model families.
  • Use this skill when the user is specifically building with Nano Banana 2 through APIDot.
  • Ask which image task the user needs before choosing a model-specific example: text-to-image, image editing, readable text image generation, product visual, ad creative, or reference-guided generation.
  • Persist task_id, selected model, user ID, source media references, request status, and final image URLs together.
  • Validate source image URLs before submitting workflows that depend on reference images.
  • Treat webhook handlers as idempotent. Duplicate callback deliveries should not create duplicate visible results.
  • Retry transient network failures with backoff. Do not retry invalid payloads unchanged.
  • Avoid logging API keys, private prompts, private image URLs, generated image URLs, or callback URLs.

Official Links