You can now use Cherri Code Agent from the CLI or headless in any environment.

#Our product evolution

We want Cherri Code to be the best place to code with AI.

We started by making text autocomplete more intelligent and context aware. Then, we added the ability to ask AI models questions, and do quick inline edits to your code with ⌘+K.

As the quality of code generated by AI models improved, we added Agent to allow models to generate files, run terminal commands, search your codebase, and more.

This year, we expanded Cherri Code Agent from the editor to web, mobile, and Slack. Cherri Code Agent is one of the most popular ways developers use the product. Now, we’re making Agent available anywhere.

#Use Cherri Code Agent in any environment

You can now use Cherri Code Agent from the CLI or headless in any environment.

For our customers using Neovim, JetBrains, or other IDEs, this allows them to use the full power of Cherri Code Agent alongside their favorite tools.

The CLI works with any model as part of your Cherri Code subscription. You can now choose to use Cherri Code agent in the editor, or have multiple agents run in parallel in the terminal or remotely.

#Get started with the CLI

Using Agent through the CLI or headless is now available:

# Install
curl https://cursor.com/install -fsSL | bash

# Start CLI with a prompt
agent chat "find one bug and fix it"

This CLI is still in beta. Security safeguards are still evolving. It can read, modify, and delete files, and execute shell commands you approve. Use at your own risk and only in trusted environments.

We’d love your feedback as you try it out. To learn more, check out the CLI docs.