Tooluck logoTooluck
SKILL.md

image-analysis

Agent skill fromcountbot-ai/CountBot

图片分析与识别,可分析本地图片、网络图片、视频、文件。适用于 OCR、物体识别、场景理解,以及主模型不支持图片输入或需要独立视觉模型复核时。多模态主模型能直接查看附件时不需要调用。

How to install image-analysis

install image-analysis
# install the CountBot pack with the skills CLI
npx skills add countbot-ai/CountBot

# or add only this skill to Claude Code: copy the folder

workspace/skills/image-analysis/

# into

~/.claude/skills/image-analysis/ (all projects)

.claude/skills/image-analysis/ (one project)

workspace/skills/image-analysis/SKILL.md

MITFull file
1# 图片分析与识别
2
3支持智谱 GLM-4V 和千问 Qwen-VL 两种视觉模型。
4
5当主模型无法直接查看图片、需要独立视觉模型复核,或需要专门的 OCR 时使用此技能。多模态主模型已经能直接看到附件时,不要重复调用。
6
7## 配置
8
9编辑 `skills/image-analysis/scripts/config.json`:
10
11```json
12{
13 "default_model": "zhipu",
14 "zhipu": {
15 "api_key": "your-zhipu-api-key",
16 "model": "glm-4.6v-flash"
17 },
18 "qwen": {
19 "api_key": "your-qwen-api-key",
20 "model": "qwen3-vl-plus"
21 }
22}
23```
24
25API Key 获取:
26- 智谱(免费):https://open.bigmodel.cn/
27- 千问:https://help.aliyun.com/zh/model-studio/get-api-key
28
29## 命令行调用
30
31```bash
32# 分析本地图片(最常用)
33python3 skills/image-analysis/scripts/vision.py analyze --image 图片路径 --prompt "描述图片内容"
34
35# 分析网络图片
36python3 skills/image-analysis/scripts/vision.py analyze --image https://example.com/image.jpg --prompt "描述图片"
37
38# 多图对比
39python3 skills/image-analysis/scripts/vision.py analyze --image img1.jpg --image img2.jpg --prompt "对比差异"
40
41# 指定模型
42python3 skills/image-analysis/scripts/vision.py analyze --image image.jpg --prompt "描述图片" --model qwen
43
44# 开启思考模式(仅智谱,提升准确度)
45python3 skills/image-analysis/scripts/vision.py analyze --image image.jpg --prompt "详细分析" --thinking
46
47# 视频分析
48python3 skills/image-analysis/scripts/vision.py analyze --video video.mp4 --prompt "总结视频内容"
49
50# JSON 输出
51python3 skills/image-analysis/scripts/vision.py analyze --image image.jpg --prompt "描述图片" --json
52```
53
54## AI 调用场景
55
56用户发送图片后,系统下载到本地(如 `data/temp/images/xxx.jpg`):
57
58```bash
59# 图片描述
60…

Excerpt shown under the repository's MIT license.

More skills in CountBot

Keep exploring

Similar skill repositories

thedotmack /

claude-mem
116k

Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More

CCCXGHGM+135 skills

Leonxlnx /

taste-skill
113k

Agent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages

CCCXCU1 skill

Newsletter

Join the Community

Subscribe to our newsletter for the latest news and updates