发现优质的 AI Agent 技能
聚合 Claude Skills、LangChain、AutoGPT 等优质资源，助力开发者快速构建智能应用
Example: api-docs-writer (🟡 Trung bình — Complexity 9)
> **Nguồn cảm hứng:** Tổng hợp từ Stripe API docs style + OpenAPI spec. > **Mức độ:** Trung bình — SKILL.md + resources/ + examples/ (không cần scripts). > **Tham khảo:** Anthropic's `doc-coauthoring` skill — tách templates ra resources.
Example: pdf-processor (🟠 Phức tạp — Complexity 18)
> **Nguồn gốc:** Dựa trên Anthropic's official `pdf` skill — 1 trong 16 skills chính thức. > **Điểm hay:** Description cực "pushy", progressive disclosure 3 tầng, reference docs dày. > **Bài học:** Skill phức tạp nhưng SKILL.md vẫn lean nhờ tách logic vào resources.
Example: frontend-design (🟢 Đơn giản nhưng MẠNH — Complexity 5)
> **Nguồn gốc:** Anthropic's official `frontend-design` skill — chỉ 43 dòng! > **Điểm hay:** Chứng minh skill tốt không cần dài. Tone "explain the why" tuyệt vời. > **Bài học:** Lean = AI follow tốt hơn. Viết ngắn nhưng truyền cảm hứng.
Authors and Acknowledgements
This skill was developed collaboratively by [Vijay Samtani](https://orcid.org/0000-0002-4246-6453), Claude Sonnet 4.6 (claude-sonnet-4-6, Anthropic), and Google Gemini 3 (gemini-3-pro-preview, Google), in February 2026.
Claude Agent SDK Corrections
This uses **@anthropic-ai/claude-agent-sdk v0.1.50**.
Claude Agent SDK - Structured Outputs & Error Prevention Guide
**Package**: @anthropic-ai/claude-agent-sdk@0.2.12 **Breaking Changes**: v0.1.45 - Structured outputs (Nov 2025), v0.1.0 - No default system prompt, settingSources required
"TokenWatch"
"Track, analyze, and optimize AI API token costs across 10 providers (Anthropic, OpenAI, Google, Mistral, xAI, Kimi, Qwen, DeepSeek, Meta, MiniMax) — 44 models. Set budgets, get alerts, compare costs, and get optimization suggestions. Zero dependencies, local-only, MIT licensed."
Prompt Engineering Expert - Best Practices Guide
This document synthesizes best practices from Anthropic's official documentation and the Claude Cookbooks to create a comprehensive prompt engineering skill.
Prompt Engineering Expert - Best Practices Guide
This document synthesizes best practices from Anthropic's official documentation and the Claude Cookbooks to create a comprehensive prompt engineering skill.
Anthropic Official Skill Best Practices
> Source: https://platform.claude.com/docs/zh-CN/agents-and-tools/agent-skills/best-practices > Last Updated: 2025-01
Claude Instruction Patterns
What formats actually work when writing instructions for Claude models. Based on real-world patterns from OpenClaw agents, Anthropic documentation, and multi-agent coordination research.
OpenRouter Setup for AgentBox
This skill guides users through configuring OpenRouter as their LLM provider on this AgentBox instance. OpenRouter gives access to models from Anthropic, OpenAI, Google, Meta, and others through a single API key.