跳到正文
cccskills
免费在 GitHub 公开

unblind

Routes images to Mimo/OpenAI vision API for text-only models. Use this skill when the user sends an image, asks "what's in this picture", says "analyze this screenshot", requests "OCR" or "extract text", reviews UI designs, reads charts, or uses Chinese triggers like 识别图片/看图. Self-healing setup on first run. Does NOT handle video, audio, or PDFs.

中文简介正在准备中。当前显示原文说明。

Santazuki/unblind星标 24(GitHub 24 + cccskills 0)
集成工具OpenAI

实测评分 未实测

查看安装方法

包含的文件（75）

  • SKILL.md3.3 KB
  • .github/workflows/ci.yml747 B
  • .github/workflows/publish.yml377 B
  • .gitignore509 B
  • display/2026-05-30-provider-v3-protocol-driven-design.md9.1 KB
  • display/provider-optimization.md18.6 KB
  • docs/test-results/gpt-rebuttal-report.md8.1 KB
  • docs/test-results/step1-logger-errorHandler.md696 B
  • docs/test-results/step10-cache.md906 B
  • docs/test-results/step11-health-config.md3.6 KB
  • docs/test-results/step12-cache-granularity.md1.1 KB
  • docs/test-results/step13-openai-provider.md1.9 KB
  • docs/test-results/step14-phase0-silent.md12.2 KB
  • docs/test-results/step15-validation-tests.md2.9 KB
  • docs/test-results/step16-ci-circuitbreaker-fix.md326 B
  • docs/test-results/step17-phase3-chain-registry.md18.0 KB
  • docs/test-results/step18-phase5-test.md7.7 KB
  • docs/test-results/step19-gemini-provider.md2.8 KB
  • docs/test-results/step2-config.md2.4 KB
  • docs/test-results/step3-credentialManager.md3.0 KB
  • docs/test-results/step4-retry.md4.8 KB
  • docs/test-results/step5-imageProcessor.md5.7 KB
  • docs/test-results/step6-provider-mimo.md6.9 KB
  • docs/test-results/step7-orchestrator.md3.4 KB
  • docs/test-results/step8-cli.md13.2 KB
  • docs/test-results/step9-regression.md13.2 KB
  • install.sh1.8 KB
  • LICENSE1.0 KB
  • package.json632 B
  • README.md8.0 KB
  • resources/best_practices.md1.8 KB
  • resources/phase0-details.md2.3 KB
  • resources/troubleshooting.md3.3 KB
  • scripts/install.js3.1 KB
  • scripts/lib/cache.js3.7 KB
  • scripts/lib/config.js3.7 KB
  • scripts/lib/credentialManager.js1.7 KB
  • scripts/lib/errorHandler.js2.3 KB
  • scripts/lib/httpClient.js2.8 KB
  • scripts/lib/imageProcessor.js4.2 KB
  • scripts/lib/logger.js722 B
  • scripts/lib/orchestrator.js7.0 KB
  • scripts/lib/providers/generic-provider.js5.2 KB
  • scripts/lib/providers/protocols.js5.4 KB
  • scripts/lib/providers/provider.js2.8 KB
  • scripts/lib/providers/registry.js3.8 KB
  • scripts/lib/retry.js3.4 KB
  • scripts/unblind.mjs5.3 KB
  • settings.example.json199 B
  • templates/chain_of_thought.md1.1 KB
  • templates/output_formats/csv.md510 B
  • templates/output_formats/json.md989 B
  • templates/output_formats/yaml.md715 B
  • TEST.md12.6 KB
  • tests/evals.json2.4 KB
  • tests/sample_images/README.md1.5 KB
  • tests/scenarios.md2.5 KB
  • tests/test-cache.js2.4 KB
  • tests/test-cli.js5.7 KB
  • tests/test-config.js2.6 KB
  • tests/test-credentialManager.js1.4 KB
  • tests/test-docs.js4.5 KB
  • tests/test-errorHandler.js1.4 KB
  • tests/test-generic-provider.js9.6 KB
  • tests/test-imageProcessor.js3.0 KB
  • tests/test-install.js3.1 KB
  • tests/test-logger.js1.5 KB
  • tests/test-mimo.js2.6 KB
  • tests/test-multi-image.js9.6 KB
  • tests/test-orchestrator.js2.0 KB
  • tests/test-protocols.js11.6 KB
  • tests/test-registry.js4.2 KB
  • tests/test-retry.js3.7 KB
  • tests/test-security.js10.4 KB
  • tests/test-version.js1.3 KB

SKILL.md（原文）

安装前可以查看交给 AI 代理的指令内容。

<!-- LEVEL 1: Metadata above (~180 tokens, always loaded) --> <!-- LEVEL 2: Instructions below (~550 tokens, loaded on trigger) -->

Unblind

Route images to vision API. Never pretend to see. Never Read/Edit settings.json.

Iron Rules

  1. Phase 0 mandatory every invocation
  2. NEVER Read/Edit ~/.claude/settings.json
  3. Config via CLI (--config, --set-model) or node -e, never via tools
  4. Never preamble. Never hallucinate. Always invoke bundled script.
  5. Tool reads API key from env automatically

Quick Start

User sends image → Unblind routes to Mimo/OpenAI → returns text. OCR → extracts all text. ui-review → critiques layout/UX. describe → detailed description.

Phase 0: Self-Healing

0.1 Health check

node -e "const fs=require('fs');const os=require('os');const p=require('path').join(os.homedir(),'.claude','settings.json');const s=JSON.parse(fs.readFileSync(p,'utf8'));const issues=[];if(!s.env?.MIMO_API_KEY) issues.push('KEY_MISSING');if(!s.env?.MIMO_VISION_MODEL||s.env.MIMO_VISION_MODEL==='mimo-v2.5-pro') issues.push('MODEL_MISSING');const a=s.permissions?.allow||[];if(!a.some(x=>x.includes('unblind'))) issues.push('PERM_MISSING');if(issues.length) console.log(issues.join(' '));" 2>/dev/null
  • Empty → healthy, skip to Phase 1
  • KEY_MISSING → 0.3 | MODEL_MISSING → 0.5 | PERM_MISSING → 0.4

0.2 Version check

cd ~/.claude/skills/unblind && git fetch 2>/dev/null && behind=$(git rev-list HEAD..origin/master --count 2>/dev/null) && [ "$behind" -gt 0 ] && echo "NEEDS_UPDATE $behind"
  • NEEDS_UPDATE <N> → 🐌 unblind 有 N 个更新没拿... git pull 一下嘛~（不阻塞）

0.3-0.9 Repair → Phase 1

See resources/phase0-details.md. All healthy → Phase 1.

Phase 1-4: Analyze

  1. Detect image path from [Image: source: <path>]. Must be absolute, supported ext, no shell metacharacters.
  2. Classify mode: describe (default), ocr, ui-review, chart-data, object-detect, compare.
  3. Execute: node ~/.claude/skills/unblind/scripts/unblind.mjs '<path>' <mode> — no preamble.
  4. Report: print stdout. API key error → back to 0.3.
<!-- LEVEL 3: Resources (on-demand) -->

Resources

  • resources/phase0-details.md — 完整 Phase 0 流程、CLI 参考、边缘场景
  • resources/troubleshooting.md — 修复命令、常见错误、Node.js 配置
  • resources/best_practices.md — 模型选择、token 优化、调试
  • README.md — 安装指南、安全验证

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护，适用 Google 的隐私权政策和服务条款。

报告此技能的问题