Fabric

An open-source framework for augmenting humans with AI, one composable prompt at a time.

Free· Free and open-source (MIT). Users bring their own API keys and pay each LLM provider directly; local models via Ollama or LM Studio incur no per-token cost.AgentsModel-agnostic: OpenAI GPT-4o/GPT-4.1, Anthropic Claude (incl. Opus 4.7), Google Gemini, Azure OpenAI, Bedrock, Vertex AI, plus local Ollama and LM Studio models.

Visit website →

Best for

Developers, security analysts and terminal-native power users who want a scriptable, provider-agnostic way to apply reusable AI prompts to files, pipelines and personal knowledge work.

Skip if

Non-technical users who want a chat UI, teams needing hosted infrastructure with billing and SSO, or anyone who needs managed guardrails, evals and audit logging out of the box.

Fabric is an open-source, command-line-first framework for turning AI models into a library of reusable, chainable prompts. Created by security researcher Daniel Miessler, it addresses what he calls AI's real bottleneck: not model capability, but integration into everyday workflows. Instead of pasting prompts into a chat window, you install a Go binary, pick a 'Pattern' (a curated Markdown prompt such as extract_wisdom, summarize, analyze_claim, create_summary, or write_essay), pipe text into it, and get a structured result on stdout. Patterns can be chained through Unix pipes so a YouTube transcript can flow into extract_wisdom, then into create_summary, then into a social-post generator in a single one-liner. Fabric natively integrates with OpenAI, Anthropic Claude, Google Gemini, Azure OpenAI, Amazon Bedrock, Vertex AI, Ollama, LM Studio and Perplexity, plus many OpenAI-compatible endpoints (Groq, DeepSeek, OpenRouter, Together, Mistral, Cerebras, GitHub Models and others). A --serve mode exposes a REST API with chat-completion, pattern-management, context and session endpoints, and includes Ollama-compatible routes so existing tooling can point at Fabric as a drop-in backend. The audience is developers, sysadmins, security analysts, researchers and power users who already live in the terminal and want AI to behave like just another Unix tool: composable, scriptable, and versioned in Git alongside their own custom patterns.

Editor's take

Fabric is the closest thing the LLM world has to a Unix philosophy: small, composable prompt tools you can pipe together. For anyone comfortable at a shell, it turns 'go paste this into ChatGPT' into a reproducible one-liner, and the ability to swap between Claude, GPT and a local Ollama model with a single flag is genuinely liberating. It is not a product so much as a toolkit, and that is exactly its charm.

— The AI Tool Bible editorial team

Pros

✅ Fully open source (MIT) with a large, actively curated pattern library covering summarisation, analysis, extraction and writing tasks.
✅ Provider-agnostic: works with commercial APIs (OpenAI, Anthropic, Gemini, Bedrock, Vertex) and local models (Ollama, LM Studio) from one CLI.
✅ Composable through Unix pipes, making it trivial to chain patterns and integrate with existing shell scripts and cron jobs.
✅ Built-in helpers like YouTube transcript extraction, streaming output, dry-run cost preview and multi-language support.
✅ REST API server mode with Ollama-compatible endpoints, so it can act as a backend for other apps.
✅ Custom patterns are just Markdown files you can version, share and fork, making prompt engineering reviewable in Git.

Cons

⚠️ CLI-first design has a steep learning curve for non-technical users; there is no polished consumer GUI.
⚠️ You must bring and manage your own LLM API keys and pay provider costs; Fabric itself is unhosted.
⚠️ Pattern quality varies and community contributions are not always benchmarked, so output consistency depends on which pattern you pick.
⚠️ Windows support exists but the ergonomics still favour macOS and Linux terminal workflows.
⚠️ No built-in evaluation, guardrails or observability layer, so production use requires bolting on your own logging and safety tooling.

Use cases

YouTube video summarisation and wisdom extractionLong-form article and PDF summarisationSecurity report and threat-intel analysisMeeting-transcript distillationBlog post and essay drafting from notesClaim and argument analysisSocial media post generationShell-scripted batch processing of documentsLocal, private LLM workflows via OllamaDrop-in Ollama-compatible API backend for other apps