Skip to main content
The Requesty CLI is a terminal app that connects the AI coding agents already installed on your machine to Requesty. It detects each harness, writes that harness’s own configuration file so requests go through the Requesty gateway, and shows your spend for the last 30 days. There is no proxy to keep running and no wrapper command to learn. You keep starting claude, codex, opencode, pi or hermes exactly as before, and every request now goes through Requesty. With the CLI you can:
  • Access 300+ models from OpenAI, Anthropic, Google, Mistral and many other providers through one API key.
  • Set up a harness in a few keystrokes instead of hand editing JSON, TOML or YAML.
  • Apply fallback policies, load balancing and latency routing to your coding agents.
  • Keep spend, logs and analytics for every tool in one dashboard.

​
Quick start

The installer puts the binary in ~/.requesty/bin and adds that directory to your PATH. Open a new shell, then run:
Paste a key from the API Keys page when the CLI asks for one, select a harness, pick a model, and restart the harness. Your existing configuration file is backed up before anything is written.

​
Learn more

The CLI is open source under Apache 2.0. The README is the reference for supported harnesses, the files each one writes, backups, upgrades and troubleshooting.

requestyai/cli on GitHub

Source, releases and full documentation for the CLI.
Last modified on August 10, 2026