Fabric CLI

A CLI with more than 250 ready prompt patterns: runs them from the terminal on the model you choose, with a REST API and Ollama support

  • 44,243 ★ on the repository
  • Last commit
  • MIT
CLI

Extended access

Why this level

  • Sends the entered text to a third-party model provider
  • Can save notes to a chosen folder
  • Installing by an internet script runs someone else's code
All reasons and checks

danielmiessler/fabric

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add cli/fabric

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Extended accessThe tool runs code or makes network calls and only reads external data.

  • Sends the entered text to a third-party model provider
  • Can save notes to a chosen folder
  • Installing by an internet script runs someone else's code

This is third-party code. Review the repository files before installing.

What it does

Fabric is an open framework with a catalog of prompt patterns for specific tasks. A pattern runs from the command line against the model you choose, for example `yt URL | fabric -p extract_wisdom` pulls the essence out of a YouTube video. The repository has more than 250 patterns, and you can add your own, bind models to them and set aliases. There is a REST API server with Swagger at `/swagger` and an Ollama compatibility mode. Fabric itself is free, and you pay the provider for model calls: it works with keys or with local models (OpenAI, Anthropic, Ollama, Azure and others).

Who it is for. For people who repeat the same requests to AI: summaries of videos and articles, text analysis, notes. It suits developers, analysts and authors who are at home in the terminal.

Good fit when

  • You need to run a text or a video through a proven pattern instead of writing a prompt from scratch
  • You need a set of patterns that can be called from scripts and command-line pipelines
  • You need a single point for working with different models, including local ones through Ollama

Not a fit when

  • You need an agent that performs actions in a project itself: Fabric is not an agent, it runs a pattern and returns the model's answer
  • You cannot send text to a third-party model provider

Example request

Take the meeting transcript, run it through the extract_wisdom pattern and save the result to a note

Limitations

Fabric sends the text you enter to the chosen model provider, so do not pass materials through it that must not leave your machine. Installing through `curl ... | bash` runs a script from the internet: read it before running. Homebrew installs the `fabric-ai` package and the command is called `fabric-ai`, so add an alias `fabric`. After installation run `fabric --setup`: it creates directories and asks for provider keys. The README does not count the patterns, we count by the repository folders and write more than 250.

How to disable. Remove the Fabric executable (in Homebrew: brew uninstall fabric-ai), the settings directory and revoke the provider keys you entered during `fabric --setup`.

FAQ

What is a pattern in Fabric?

A ready prompt for one task, for example analyzing a video or extracting the main ideas. A pattern runs with fabric -p pattern_name.

How many patterns does Fabric have?

More than 250 by the repository folders. The README gives no exact number.

Is Fabric paid?

No, Fabric itself is free. You pay the model provider if you use cloud models. With Ollama you can work locally.

How is Fabric different from agent skills?

The CLI itself runs a pattern: it sends the text to a model and returns the answer. An agent reads a skill and decides which actions to take.

Discussion

No questions or reviews yet.

CLI tools
Editors’ pick

A structured prompt system: templates with role, profile, skills, rules and workflow for repeatable model requests

PromptMinimal access12.6KRepository stars

A visual environment for testing prompts: a graph compares the answers of several models to prompt variations, with evaluators and 188 example flows

PromptExtended access3KRepository stars

Turns any OpenAPI or Swagger API into CLI commands at runtime, with BM25 endpoint search and no codegen

CLIExtended access265Repository stars

A skill that gives text-only models sight: an image by path or from a paste becomes structured JSON with transcribed text, layout and semantics

SkillExtended access4.2KRepository stars

A CLI and Python library for OpenAI, Claude, Gemini, local and other models: prompts, chats, tools, embeddings and a SQLite log

CLIExtended access12.6KRepository stars

A local-first AI agent running on your own machine via llama.cpp that drives a browser, edits files, runs approved commands and remembers context across session

CLIBroad access3.3KRepository stars

Comparison with similar tools

Entries that share the main search query, and how they differ.

EntryOfficialRepository starsMain difference
Fabric (this entry)no44,243
Awesome Claude Promptsno5,519a collection of prompts to read
LangGPTno12,584a structured prompt format

README in short

The README describes Fabric as an open framework for augmenting humans with AI: a catalog of patterns, launching from the command line, a REST API and Ollama. Installation is by script, Homebrew (fabric-ai), Docker or building in Go, and the first setup is the fabric --setup command. MIT license.

README badge

Report a problem

Foxx AIFabric
Foxx AI

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.