AI Agent Hub
Back to skills
All-in-One Image Editing icon

All-in-One Image Editing

Design & Media Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please follow https://skillhub.cn/install/skillhub.md and install @ixhlink/ixhlink-skills-image-edit-pro.

About this skill

Problem

Image-editing Agents need to handle reference-image input, async task polling, and paid retry. This skill wraps the image_edit capability and keeps request, task status, and result URL in one call path.

Workflow

  • Submit POST /api/v1/llm/invoke with prompt and images; images needs at least 1 item, using either b64 or url.
  • On success, read data.task.id and poll GET /api/v1/llm/tasks/{task_id} until succeeded or failed.
  • Output images are in data.result.data[].url; poll every 2–5 seconds as a starting interval.
  • If the first call returns 402, keep the JSON body unchanged and retry with WeixinPay-Required and X-Payment-Id headers.

Boundaries

  • Paid calls require the Agent to have WeChat Pay capability; terminate and notify the user if it is missing.
  • 400, 404, and 503 usually point to parameter, model, or service issues; check data.error_message when a task fails.

Use Cases

  • An Agent edits a supplied product image according to a text prompt and returns the result image URL.
  • A service submits a Base64 reference image and edit prompt, then polls the async task for output.
  • A client receives a 402 response, keeps the request body unchanged, and retries with payment headers.
  • An MCP/Web client checks WeChat Pay capability before confirming the paid image-editing flow.

Best For

  • Platform engineers adding visual editing to Agents need to connect reference images, prompts, and result URLs.
  • MCP developers integrating multimodal tools need to handle async polling and 402 payment retries.
  • Backend engineers building paid creation products need per-call pricing, order confirmation, and body-preserving retries.
  • Technical leads for image-editing products need clear Base64/URL input and task-status handling.