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 xyva-yuangui/XyvaClaw --skill vision-readergit clone --depth 1 https://github.com/xyva-yuangui/XyvaClawWrote 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/xyva-yuangui/xyvaclaw/vision-reader)<a href="https://agentmods.dev/skills/xyva-yuangui/xyvaclaw/vision-reader"><img src="https://agentmods.dev/badge/skills/xyva-yuangui/xyvaclaw/vision-reader/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/xyva-yuangui/xyvaclaw/vision-reader"><img src="https://agentmods.dev/badge/skills/xyva-yuangui/xyvaclaw/vision-reader.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00059 | $0.01084 |
| Opus 5.5 | $0.00024 | $0.00434 |
| Sonnet 5.5 | $0.00012 | $0.00217 |
| Haiku 4.5 | $0.00006 | $0.00108 |
Grade A, and why
vision-reader 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Vision Reader
图片识别助手。触发后必须先询问用户确认，再执行分析。
| 触发场景 | 询问确认内容 |
|---|---|
| OCR 提取 | 确认图片路径、识别语言 (中文/英文)、输出格式 |
| 图片分析 | 确认分析深度 (简述/详细)、关注重点 |
| 截图理解 | 确认截图来源、需要提取的信息类型 |
| 文档识别 | 确认文档类型、是否保留格式 |
执行流程: 触发 → 询问用户 → 用户确认 → 执行识别 → 返回结果
Full docs: read SKILL-REFERENCE.md
Analyze images using AI vision models or local OCR.
Capabilities
- AI Vision Analysis — send image to vision model (qwen3.5-plus, qwen3-max) for understanding
- OCR Text Extraction — extract text from images using macOS Vision framework
- Screenshot Analysis — combine with screenshot-sender for screen understanding
- Document Reading — extract text from scanned documents, receipts, etc.
Method 1: AI Vision Model (Recommended)
Use vision-capable models directly. These models accept image input:
bailian/qwen3.5-plus— 1M context, supports["text", "image"]bailian/qwen3-max-2026-01-23— 262K context, supports["text", "image"]qwen-portal/vision-model— 128K context, supports["text", "image"]
When analyzing images, switch to one of these models if the current model doesn't support image input.
Workflow
- User provides image path or URL
- Agent reads the image file
- Agent sends to vision model with analysis prompt
- Return structured analysis
Method 2: Local OCR (macOS Vision Framework)
For pure text extraction without needing an AI model:
python3 {baseDir}/scripts/ocr_extract.py <image_path> [--lang zh-Hans,en] [--json]
Parameters
| Flag | Description | Default |
|---|---|---|
<image_path> |
Path to image file (PNG, JPG, etc.) | required |
--lang |
OCR languages (comma-separated) | zh-Hans,en |
--json |
Output as JSON with bounding boxes | plain text |
--confidence |
Min confidence threshold (0-1) | 0.5 |
Output (plain text mode)
第一行文字
第二行文字
...
Output (JSON mode)
{
"text": "完整提取文本",
"blocks": [
{"text": "第一行", "confidence": 0.98, "bbox": [x, y, w, h]},
...
],
"language": "zh-Hans"
}
What ships with it
4 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 · 129 lines · 59 tokens per session scan A 4b9342d6b94f
vision-reader is a skill published in the GitHub repository xyva-yuangui/XyvaClaw (21 stars, last pushed 2mo ago), licensed MIT. It adds 59 tokens to every session and 1,084 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
deepseek-helper
A guide for using the DeepSeek API, a service that lets programs send requests to DeepSeek language models. It covers model choice, example requests, prompt improvements, cost estimates, and common API errors.
image-understanding
Enable image understanding capabilities for describing/analyzing images, reading text, and answering questions about visual content.
mastra
Comprehensive Mastra framework guide. Teaches how to find current documentation, verify API signatures, and build agents and workflows. Covers documentation lookup strategies (embedded docs, remote docs), core concepts (agents vs workflows, tools, memory, RAG), TypeScript requirements, and common patterns. Use this…
claude-api
Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude…
claude-api-in-prototypes
Call Claude from your HTML artifacts via window.claude.complete.
clawrouter
Hosted-gateway LLM router — save 84% on inference costs. A local proxy that forwards each request to the blockrun.ai gateway, which routes to the cheapest capable model across 90 models from OpenAI, Anthropic, Google, DeepSeek, xAI, Z.AI, and more. Free open-weight models included. Also exposes realtime market data…