ai-model-wechat

1## Sibling skills (local only)
2 
3Sibling CloudBase skills ship beside this skill. Use local relative paths such as `../auth-tool-cloudbase/SKILL.md`.
4 
5If a referenced sibling skill file is missing from this environment, ask the user to install the full CloudBase plugin (or the missing skill). Do **not** HTTP-fetch remote skill or protocol markdown into the agent context.
6 
7## When to use this skill
8 
9Use this skill for **calling AI models in WeChat Mini Program** using `wx.cloud.extend.AI`.
10 
11**Use it when you need to:**
12 
13- Integrate AI text generation in a Mini Program
14- Stream AI responses with callback support
15- Read text and fields from an image (OCR, receipts, forms) with a vision-capable model
16- Call Hunyuan models from the WeChat environment
17 
18**Do NOT use for:**
19 
20- Browser/Web apps → use `ai-model-web` skill
21- Node.js backend or cloud functions → use `ai-model-nodejs` skill
22- Image generation → use `ai-model-nodejs` skill (not available in Mini Program)
23- Runtimes without a CloudBase SDK (native apps, Python, etc.) → use `http-api-cloudbase` skill (it now includes the `ai_model` OpenAPI spec for direct HTTP calls)
24 
25---
26 
27## ⛔ STOP — `wx.cloud.extend.AI.createModel(provider)` argument is **not** a vendor / model name
28 
29Read this before writing any `createModel(...)` line. Agents frequently hallucinate this argument. There are **exactly four** legal shapes. Anything else is a bug.
30 
31| ✅ Legal `createModel(provider)` argument | When to use it |
32…