Mods
AI for the command line, built for pipelines: UNIX filters meet LLMs.
About
Mods by Charmbracelet is a fully open-source (MIT), Go-based static binary that behaves like a native UNIX filter but secretly calls a large language model. It reads stdin, sends text to any OpenAI-compatible endpoint with a user-supplied prompt, and streams the answer to stdout or into the next stage of a pipeline.
No TTY is required, so Mods works inside crontabs, CI jobs, Docker containers, Git hooks, and over SSH. Common uses include AI-powered text processing, code review, commit message generation, and data transformation, with support for OpenAI, Anthropic, local models, and other providers.
What it can do
- Pipe stdin through any LLM like a UNIX filter
- Works headless in crontabs, CI jobs, and over SSH
- Streaming responses back into shell pipelines
- Shell completions for bash, zsh, fish, and PowerShell
Setup
brew install charmbracelet/tap/mods, or go install github.com/charmbracelet/mods@latest. Works on macOS, Linux, and Windows.
Prerequisites
An API key for an OpenAI-compatible provider, or a local model endpoint like LocalAI.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 106; Trending score 84/100 from researcher assessment.
★ 4,517 GitHub stars
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.