Fabric is Daniel Miessler’s open-source framework for augmenting humans with AI, with 44,300+ stars on GitHub — a modular, crowdsourced library of expert prompts called Patterns that each solve one specific problem well.
Most people improvise a fresh prompt for every task and get mediocre results. Fabric takes the opposite approach: for each recurring job — summarizing a video, extracting the wisdom from an article, analyzing a claim, rating content, writing an essay, turning a meeting into actions — the community maintains one refined, versioned Pattern, a system prompt tuned for exactly that outcome. Patterns run anywhere: pipe a YouTube transcript, an article or a file through the Fabric CLI, use them in chat applications, or call them from your own scripts. Fabric is model-agnostic, working with OpenAI, Anthropic and local models, and you can keep your own pattern directory to extend the library with your workflows. Behind it is a simple philosophy: use AI to amplify human thinking across work and life, one well-scoped problem at a time. Fabric is MIT-licensed and one of the most starred AI projects on GitHub.
Fabric is completely free and open source under the MIT license. Your only cost is whatever model you choose to run the patterns on — including free local models.
Writers, analysts, researchers and knowledge workers who ask AI for the same kinds of thinking every day and want proven, repeatable prompts instead of ad-hoc ones. (Listed on Tooloso with its GitHub home, the project’s canonical address.)