siliconflow-vision
图片识别与分析工具。使用视觉大模型识别图片内容,输出详细客观的识别结果供主模型分析。当用户发图片时,主模型必须直接调用此 skill,然后基于识别结果进行分析和回答。支持 SiliconFlow(默认)、OpenAI、Anthropic 等多服务商。
Install
mkdir -p .claude/skills/siliconflow-vision && curl -L -o skill.zip "https://mcp.directory/api/skills/download/5962" && unzip -o skill.zip -d .claude/skills/siliconflow-vision && rm skill.zipInstalls to .claude/skills/siliconflow-vision
About this skill
图片识别与分析 Skill
工作流程
用户发图片 → 主模型直接调用 skill → skill 识别图片 → skill 输出详细结果 → 主模型分析+网络搜索 → 给出准确回答
核心原则(重要)
主模型必须遵守:
- 必须调用 skill:用户发图片时,主模型必须调用此 skill
- 禁止直接回答:不要用 OpenClaw 的
image工具,不要跳过 skill - skill 只识别:skill 只做客观识别,不做分析解读
- 主模型负责思考:分析、联想、回答由主模型完成
正确流程:
用户: [图片] 这个 meme 笑点在哪?
主模型: python scripts/analyze_image.py meme.jpg
↓
Skill 输出: 详细识别结果(文字+元素)
↓
主模型: 基于识别结果进行分析
- 如果需要背景知识 → 网络搜索
- 如果需要验证 → 网络搜索
↓
主模型回答: 结合事实的准确解析
错误示范:
❌ 直接调用 image 工具回答
❌ 跳过 skill 自己猜测
❌ skill 做过多分析解读
❌ 不验证信息就回答
使用方式
脚本调用
# 基本用法(推荐)
python scripts/analyze_image.py /path/to/image.jpg
# 指定自定义问题
python scripts/analyze_image.py image.jpg -q "只提取文字"
# 智能模式(更精准,适合复杂图片)
python scripts/analyze_image.py meme.png -m smart
# 简短输出
python scripts/analyze_image.py screenshot.png -s
# 指定服务商
python scripts/analyze_image.py photo.jpg --provider openai
脚本参数
| 参数 | 说明 | 示例 |
|---|---|---|
image | 图片路径或 URL | /path/to/image.jpg |
-q, --question | 自定义问题 | -q "提取所有文字" |
-m, --model | 模型选择 | -m smart |
-s, --short | 简短模式 | -s |
--provider | 指定服务商 | --provider openai |
-c, --compress | 压缩图片 | -c |
支持的服务商
| 服务商 | 默认模型 | 特点 | 配置键 |
|---|---|---|---|
| SiliconFlow | deepseek-ai/deepseek-vl2 | 默认,快速稳定 | siliconflow_api_key |
| OpenAI | gpt-4o | 通用强大 | openai_api_key |
| Anthropic | claude-sonnet-4 | 推理能力强 | anthropic_api_key |
模型选择
| 模式 | 模型 | 速度 | 适用场景 |
|---|---|---|---|
| fast | deepseek-ai/deepseek-vl2 | ~5秒 | 默认,详细识别日常图片 |
| smart | Qwen/Qwen2.5-VL-72B-Instruct | ~2分钟 | 复杂图片、图表、需要精准分析 |
| balanced | deepseek-ai/deepseek-vl 待测试 | 2-turbo | 平衡速度与精度 |
配置说明
文件: config/default.json
{
"provider": "siliconflow",
"api_key": "sk-xxx",
"model": "fast"
}
也可通过环境变量:
SILICONFLOW_API_KEYOPENAI_API_KEYANTHROPIC_API_KEY
Skill 输出格式(优化版)
当识别图片时,skill 会输出:
1. 图片类型:截图/表情包/聊天记录等
2. 清晰文字:完整提取所有文字
3. 画面元素:列出所有视觉元素
4. 整体布局:图片结构
5. 风格氛围:简约/搞笑/暗黑等
6. 其他细节:值得注意的元素
重要原则:
- 只做客观识别,不做分析解读
- 只做简单描述,不过度思考
- 文字必须完整准确
- 让主模型负责思考分析
错误处理
| 错误 | 解决方案 |
|---|---|
| API key 失效 | 检查配置或环境变量 |
| 图片不存在 | 检查路径是否正确 |
| 超时 | 切换到 fast 模式重试 |
| 服务商不支持 | 切换到其他服务商 |
支持的图片类型
- 📸 截图:代码错误、聊天记录、网页
- 🎭 表情包/Meme:搞笑图片、网络梗图
- 📄 文档:表格、合同、发票、名片
- 📊 图表:数据可视化、流程图
- 🖼️ 照片:风景、产品、人物
优化历史
2026-02-04 优化:
- ✅ 修改默认提示词为详细识别模式
- ✅ 要求完整提取所有清晰文字
- ✅ 要求描述画面元素和布局
- ✅ 明确禁止 skill 做分析解读
- ✅ 强调主模型负责思考分析
- ✅ 强制主模型使用 skill(禁止用 image 工具)
2026-02-06 整合:
- ✅ 合并 image-understand 功能
- ✅ 支持多服务商(SiliconFlow、OpenAI、Anthropic)
- ✅ 支持多种图片格式
- ✅ 支持图片 URL 和本地路径
More by openclaw
View all skills by openclaw →You might also like
flutter-development
aj-geddes
Build beautiful cross-platform mobile apps with Flutter and Dart. Covers widgets, state management with Provider/BLoC, navigation, API integration, and material design.
drawio-diagrams-enhanced
jgtolentino
Create professional draw.io (diagrams.net) diagrams in XML format (.drawio files) with integrated PMP/PMBOK methodologies, extensive visual asset libraries, and industry-standard professional templates. Use this skill when users ask to create flowcharts, swimlane diagrams, cross-functional flowcharts, org charts, network diagrams, UML diagrams, BPMN, project management diagrams (WBS, Gantt, PERT, RACI), risk matrices, stakeholder maps, or any other visual diagram in draw.io format. This skill includes access to custom shape libraries for icons, clipart, and professional symbols.
ui-ux-pro-max
nextlevelbuilder
"UI/UX design intelligence. 50 styles, 21 palettes, 50 font pairings, 20 charts, 8 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app, .html, .tsx, .vue, .svelte. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient."
godot
bfollington
This skill should be used when working on Godot Engine projects. It provides specialized knowledge of Godot's file formats (.gd, .tscn, .tres), architecture patterns (component-based, signal-driven, resource-based), common pitfalls, validation tools, code templates, and CLI workflows. The `godot` command is available for running the game, validating scripts, importing resources, and exporting builds. Use this skill for tasks involving Godot game development, debugging scene/resource files, implementing game systems, or creating new Godot components.
nano-banana-pro
garg-aayush
Generate and edit images using Google's Nano Banana Pro (Gemini 3 Pro Image) API. Use when the user asks to generate, create, edit, modify, change, alter, or update images. Also use when user references an existing image file and asks to modify it in any way (e.g., "modify this image", "change the background", "replace X with Y"). Supports both text-to-image generation and image-to-image editing with configurable resolution (1K default, 2K, or 4K for high resolution). DO NOT read the image file first - use this skill directly with the --input-image parameter.
fastapi-templates
wshobson
Create production-ready FastAPI projects with async patterns, dependency injection, and comprehensive error handling. Use when building new FastAPI applications or setting up backend API projects.
Stay ahead of the MCP ecosystem
Get weekly updates on new skills and servers.