kangarooking

kangarooking / system-prompt-skills

Public

从 165 个顶级 AI 产品系统提示词中蒸馏出的 15 个可执行 Agent skill

agent-design agent-skills ai-agents cangjie-skill context-management
14
3
69% credibility
Found May 05, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
AI Summary

A pack of 15 reusable design patterns for crafting effective system prompts for AI agents, distilled from prompts used by major AI products.

How It Works

1
🔍 Discover the skill pack

You stumble upon a handy collection of ready-made tips for making AI assistants smarter and safer.

2
📖 Browse the guide

You open the main index to see simple maps of all the tips grouped by what they help with.

3
Pick your path
🆕
New AI product

Grab basics for role, safety, and neat replies.

🤖
Smart agent

Add tools, flow control, and team-up skills.

🔎
Search helper

Mix search triggers, sources, and proof links.

4
🧩 Mix your tips

You pick the best matching tips and blend them into your own instructions for the AI.

5
✨ Craft your prompt

You build a powerful set of starting rules that make your AI reliable and clever right away.

6
🚀 Launch your AI

You add these rules to your AI chat or app and watch it come alive with better behavior.

🎉 Smarter AI achieved

Your assistant now handles talks safely, uses tools wisely, and gives clear, trustworthy answers.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is system-prompt-skills?

This repo delivers 15 ready-to-use agent skills system prompt patterns, distilled from 165 system prompts of top AI products like OpenAI and Anthropic. It solves the chaos of prompt engineering by giving you battle-tested blueprints for agent identity, tools, safety, memory, and output control—think skills vs system prompt debates settled with plug-and-play modules for stable AI agents. Developers grab composable designs for voice, mobile, or coding agents without starting from scratch.

Why is it gaining traction?

Unlike raw prompt leaks or generic templates, it organizes skills into core architecture, interaction, engineering, and scenario layers, making agent 165 builds faster and safer—reducing injection risks and sloppy outputs. The hook is its origin from real-world 165 euro job agentur für arbeit style prompts, plus guides for mixing them into custom agents, pulling devs from prompt trial-and-error to production-ready systems.

Who should use this?

AI builders turning chatbots into tool-using agents, prompt engineers crafting RAG or search integrations, and indie devs targeting mobile or voice apps. Perfect for teams handling coefficient 165 agent de maitrise complexity without full-time safety experts, or thanhtran 165 github fans experimenting with system prompt skills.

Verdict

Solid starter pack for agent prompt design at low maturity—14 stars and 0.699999988079071% credibility score mean it's early but docs are clear and MIT-licensed. Try it if you're prototyping; skip for mission-critical unless you validate the patterns yourself. (198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.