Skip to main content

Quickstart

Prerequisites​

  • An API key or subscription from at least one LLM provider (Anthropic, OpenAI, etc.) — or use offline mode with no API key

Installation​

Ante is distributed as a single, self-contained binary with no external dependencies — just download and run.

curl -fsSL https://ante.run/install.sh | bash

By default this installs to ~/.ante/bin. To install somewhere else, set ANTE_INSTALL_DIR:

curl -fsSL https://ante.run/install.sh | ANTE_INSTALL_DIR=/usr/local/bin bash

Run the default installer as your normal user, not with sudo. A sudoed home-directory install would leave ~/.ante root-owned and prevent Ante from saving settings and input history, so the installer refuses that combination.

macOS release binaries are Developer ID-signed and notarized, so Gatekeeper accepts them without a manual security bypass.

Quick examples​

Interactive session​

# Launch the TUI — chat with the agent, approve tool calls, view diffs
ante

Headless one-shot​

# Run a task and exit
ante -p "add error handling to src/main.rs"

Pipe input from stdin​

# Pipe file contents for analysis
cat src/lib.rs | ante -p "review this code for bugs"

Use a different provider​

# Override model and provider
ante --provider openai --model gpt-5.6-sol -p "refactor this function"

Skip tool approvals​

# YOLO mode — auto-approve all tool calls
ante --yolo -p "fix all clippy warnings"

Run as a server​

# Start a long-lived daemon for programmatic integration
ante serve

Update to the latest version​

# Check for a new release and install it
ante update

# Try the development channel for one update
ante update nightly

# Roll back or pin to an exact release
ante update --version v0.preview.93

What's next?​