Fabric — 100+ AI Prompt Patterns for Everything
Fabric organizes 100+ AI prompt patterns for real-world tasks. 40.3K+ GitHub stars. 20+ providers, CLI + REST API, custom patterns. MIT.
Installation avec revue préalable
Cet actif nécessite une revue. Le prompt copié demande un dry-run, affiche les écritures, puis continue seulement après confirmation.
npx -y tokrepo@latest install 64bf132d-10f5-4c1a-adcc-9fb19af8cb3b --target codexDry-run d'abord, confirmez les écritures, puis lancez cette commande.
What it is
Fabric is a CLI tool that organizes 100+ AI prompt patterns for real-world tasks like summarizing articles, explaining code, extracting wisdom from podcasts, and writing essays. Each pattern is a carefully crafted system prompt that turns a general LLM into a specialized tool for a specific task. Fabric supports 20+ LLM providers and works via both CLI and REST API.
The project targets anyone who uses AI regularly and wants consistent, high-quality outputs without reinventing prompts. Developers, researchers, writers, and analysts all benefit from the curated pattern library. The project has 40.3K+ GitHub stars.
How it saves time or tokens
Fabric eliminates the prompt engineering cycle for common tasks. Instead of spending 10 minutes crafting a summarization prompt, you pipe text through fabric --pattern summarize and get a structured result immediately. Patterns are optimized for output quality, meaning fewer retry attempts and less wasted tokens on poorly structured prompts. Custom patterns let you build on the library for domain-specific tasks.
How to use
- Install Fabric:
brew install fabric-ai
# Or: curl -fsSL https://github.com/danielmiessler/fabric/releases/latest/install.sh | bash
- Configure your AI provider:
fabric --setup
- Use a pattern:
echo 'Long article text...' | fabric --pattern summarize
cat code.py | fabric --pattern explain_code
fabric --pattern write_essay 'AI in 2026'
Example
Extract key insights from a YouTube transcript:
# Download transcript and extract wisdom
yt --transcript 'https://youtube.com/watch?v=...' | fabric --pattern extract_wisdom
# Output includes:
# - SUMMARY: One-sentence overview
# - IDEAS: Key concepts discussed
# - INSIGHTS: Non-obvious takeaways
# - QUOTES: Notable statements
# - RECOMMENDATIONS: Actionable advice
Create a custom pattern for your specific domain:
# Create a custom pattern directory
mkdir -p ~/.config/fabric/patterns/review_pr
# Add system.md with your prompt template
Related on TokRepo
- Prompt Library — Browse curated prompts organized by task type
- AI Tools for Automation — Automation tools that work with Fabric patterns
Common pitfalls
- Fabric patterns are system prompts, not magic. Output quality still depends on the underlying LLM capability. Complex patterns may underperform on smaller models.
- The
--setupwizard stores API keys locally. Ensure your key file has restricted permissions on shared machines. - Custom patterns require following the exact directory structure. Missing the
system.mdfile in your pattern directory causes silent fallback to default behavior. - Always check the official documentation for the latest version-specific changes and migration guides before upgrading in production environments.
Questions fréquentes
Fabric includes 100+ curated patterns covering summarization, code explanation, writing, analysis, extraction, and many other tasks. The library is actively maintained and new patterns are added regularly by the community.
Fabric supports 20+ providers including OpenAI, Anthropic, Google, Ollama for local models, and others. The setup wizard lets you configure your preferred provider with API key authentication.
Yes. Custom patterns are directories containing a system.md file with your prompt template. Place them in ~/.config/fabric/patterns/ and they become available alongside built-in patterns via the CLI.
Yes. Fabric supports Ollama for local model inference. This means you can use all patterns with locally running models for privacy-sensitive tasks or offline use.
Fabric is an executable tool, not just a collection of prompts. It handles piping input, selecting providers, formatting output, and chaining patterns. A prompt library gives you templates to copy; Fabric gives you a command-line workflow.
Sources citées (3)
- Fabric GitHub— Fabric organizes 100+ AI prompt patterns with 40.3K+ stars
- Anthropic Prompt Engineering— Prompt pattern design for LLM applications
- Fabric Documentation— CLI tool design for AI workflows
En lien sur TokRepo
Fil de discussion
Actifs similaires
Mr. Ranedeer — Personalized AI Tutor Prompt
GPT-4 system prompt that creates a personalized AI tutor. Adapts to your learning level, language, and style. Used by 30K+ learners.
LangSmith — Prompt Debugging and LLM Observability
Debug, test, and monitor LLM applications in production. LangSmith provides trace visualization, prompt playground, dataset evaluation, and regression testing for AI.
Multi-Platform Social Media Growth Skill for Developer Communities
Production-tested Claude Code skill for automating authentic developer engagement across 10+ platforms (Twitter, Bluesky, Dev.to, HN, Reddit, GitHub, Product Hunt, Hashnode, Medium, Quora). Includes rotation queue, per-platform cooldowns, action diversity rules, Chrome MCP toolchain with JS injection patterns, anti-ban safety, and continuous optimization framework. Battle-tested across 30+ cycles.
Ant Design — Enterprise-Class React UI Library
Ant Design is an enterprise-class UI design language and React component library created by Alibaba. 60+ production-ready components with comprehensive patterns, internationalization for 30+ languages, and a rich theme system. The go-to choice for admin dashboards.