Skip to content

Pages

  • Getting started
  • Models & providers
  • Sessions
  • Customization
  • Automation

Getting started

Install e, connect a model, and run your first task from the terminal.

Install

The installer supports macOS and Linux on ARM64 and x86-64. It downloads the release for your platform, verifies its checksum, and installs the binary to ~/.local/bin.

Terminal
curl -fsSL https://raw.githubusercontent.com/intuitums/e/main/install.sh | sh

Prefer to install manually? Use the release archives. To build from source, install Rust and run:

Build from source
cargo install --git https://github.com/intuitums/e

Open a project

Start e in the directory you want it to work in.

Terminal
cd your-project
e

Review the directory trust prompt. Trust allows e to load the repository's instructions and resources.

Connect a model

Type /login and follow the sign-in or API-key flow for your provider. Then open /models and select a model.

For a local model, start Ollama or LM Studio before opening the model picker. See Models & providers for endpoints and custom configuration.

Run your first task

Enter a question or describe a change. e can read files, edit code, and run shell commands.

Example prompt
Find the authentication entry point.
Explain how a request reaches the session check.

For an edit, name the behavior you want and the checks that should pass. Inspect the diff and test output before committing.

Ctrl + O opens the transcript reader. Press Esc to return to the composer. Ctrl + C cancels active work.

Troubleshooting

Command not found
Add ~/.local/binto your shell's PATH, then reopen the terminal.
No models listed
Connect a provider with /login. For a local provider, check that its server is running, then reopen /models.
Need a format reference?
Run e docs to list the guides bundled with the binary.