Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add iflytek/iFly-Skills --skill iflytek-image-understandinggit clone --depth 1 https://github.com/iflytek/iFly-SkillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/iflytek/ifly-skills/iflytek-image-understanding)<a href="https://agentmods.dev/skills/iflytek/ifly-skills/iflytek-image-understanding"><img src="https://agentmods.dev/badge/skills/iflytek/ifly-skills/iflytek-image-understanding/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/iflytek/ifly-skills/iflytek-image-understanding"><img src="https://agentmods.dev/badge/skills/iflytek/ifly-skills/iflytek-image-understanding.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00061 | $0.01423 |
| Opus 5.5 | $0.00024 | $0.00569 |
| Sonnet 5.5 | $0.00012 | $0.00285 |
| Haiku 4.5 | $0.00006 | $0.00142 |
Grade A, and why
iflytek-image-understanding scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ifly-image-understanding
Analyze images and answer questions about their content using iFlytek's Spark Vision model (图片理解).
Setup
- Create an app at 讯飞控制台 with 图片理解 service enabled
- Set environment variables:
export IFLY_APP_ID="your_app_id" export IFLY_API_KEY="your_api_key" export IFLY_API_SECRET="your_api_secret"
Usage
Describe an image
python3 scripts/image_understanding.py photo.jpg
Ask a question about an image
python3 scripts/image_understanding.py photo.jpg -q "图片里有什么动物？"
Use basic model (lower token cost)
python3 scripts/image_understanding.py photo.jpg --domain general
Options
| Flag | Short | Description |
|---|---|---|
image |
Image file path (.jpg, .jpeg, .png) | |
--question |
-q |
Question about the image (default: describe) |
--domain |
-d |
imagev3 (advanced, default) or general (basic, fixed 273 tokens/image) |
--temperature |
-t |
Sampling temperature (0,1], default 0.5 |
--max-tokens |
Max response tokens 1-8192, default 2048 | |
--raw |
Output raw WebSocket JSON frames |
Examples
# OCR a receipt
python3 scripts/image_understanding.py receipt.png -q "总金额是多少？"
# Identify objects
python3 scripts/image_understanding.py scene.jpg -q "图片中有哪些物体？"
# Low-cost basic model
python3 scripts/image_understanding.py chart.png -q "图表的趋势是什么？" -d general
Notes
- Image formats: .jpg, .jpeg, .png
- Max image size: 4MB
- Max tokens: 8192 (input + output combined)
- Auth: HMAC-SHA256 signed WebSocket URL
- Endpoint:
wss://spark-api.cn-huabei-1.xf-yun.com/v2.1/image - Pure stdlib: No pip dependencies — uses built-in
socket+sslfor WebSocket - Model versions:
imagev3(advanced, dynamic token cost) vsgeneral(basic, fixed 273 tokens/image)
错误码说明 😢
遇到错误先别慌～看看下面找到对应的解决方法吧！✨
| 错误码 | 错误信息 | 解决办法 |
|---|---|---|
| 0 | 🎉 成功 | 恭喜你！请求正常完成啦～ |
| 10003 | 用户的消息格式有错误 | 检查一下你的请求格式是否正确哦～确保发送的是合法的JSON格式呢！ |
| 10004 | 用户数据的schema错误 | 看起来数据结构有点问题～请检查一下字段名称和类型是否正确呀！ |
| 10005 | 用户参数值有错误 | 参数值可能不太对呢～仔细核对一下每个参数的有效范围吧！ |
| 10006 | 用户并发错误：同一用户不能多处同时连接 | 检测到重复连接啦！请确保只有一个客户端在连接同一个用户ID哦～ |
| 10013 | 用户问题涉及敏感信息，审核不通过 | 哎呀，你的问题可能包含了一些不太合适的内容～换个问题试试看吧！ |
| 10022 | 模型生产的图片涉及敏感信息，审核不通过 | 生成的图片没有通过审核呢...很抱歉，换张图片再试一下吧！ |
| 10029 | 图片任何一边的长度超过12800 | 图片尺寸太大啦！请确保图片宽高都不超过12800像素哦～ |
| 10041 | 图片分辨率不符合要求 | 图片尺寸不合适的呢～要求是：50×50 < 图片总像素值 < 6000×6000 哦！ |
| 10907 | Token数量超过上限 | 内容太丰富啦！对话历史+问题的字数太多，需要精简一下输入哦～ |
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 120 lines · 61 tokens per session scan A d505a0d5a2be
iflytek-image-understanding is a skill published in the GitHub repository iflytek/iFly-Skills (212 stars, last pushed 3d ago), licensed Apache-2.0. It adds 61 tokens to every session and 1,423 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
docker-model
Use this skill when running local AI models with Docker Model Runner — the docker model CLI — e.g. "run an LLM locally with Docker", "pull a model from the ai/ namespace", "connect my app to a local model", "use a local model as backend for the Drupal AI module", or when wiring the models: top-level element into a…
image-prompt-builder-nl
Craft high-quality natural-language image prompts for any modern text-to-image or image-edit model that accepts flowing English. Trigger when the user wants help writing, rewriting, improving, or translating an English natural-language image prompt — including "write me an image prompt", "improve this image prompt"…
scammer
A scam-message checker that analyzes suspicious texts, images, or conversation descriptions. It can identify the likely stage of a scam and suggest what the sender may try next.
gpt-image-2
A skill for generating or editing images with GPT Image 2 across local, host-provided, or advisory setups.
sandbase
Access 2,000+ AI models and API tools through one MCP interface for inference, media generation, search, scraping, embeddings, social data, and structured retrieval. Use sandbasediscover before building custom integrations or declaring external data inaccessible; prefer an existing dedicated tool or API key when the…
kb-retriever
A retrieval and question-answering assistant for a local folder of documents, including Markdown, text, PDFs, and spreadsheets.