Runs in any shell on macOS, Linux, Windows, WSL, Docker, SSH sessions, inside tmux. One install, everywhere.
CLI
AI coding in your terminal.
Same 18 models, same agent capabilities, zero GUI overhead. Works in any shell, any OS, any CI runner.
Screenshot placeholder: terminal session showing mulu-code running in agent mode, streaming a planning output with 4 tasks and a y/n approval prompt.
What you get
Full agent power. No GUI required.
The CLI isn't a stripped-down companion — it's the same runtime as the desktop app.
Agent Mode, Swarm, Team, Research, multi-model compare, code reviews, codebase indexing — all available.
Type /model, /mode, /swarm, /team, /compare, /permissions without leaving the prompt.
Headless mode exits with proper status codes. Drops into GitHub Actions, GitLab CI, and any runner.
Slash commands
Built-in commands for every workflow.
Type a slash, get a menu. Tab-completion makes every command feel like a shell builtin.
/model sonnetSwitch AI model mid-conversation/mode agentEnable autonomous agent mode/swarm "build X"Launch parallel swarm agents/compare "fix bug"Compare answers from multiple models/research "topic"Deep research with citations/team "build app"Multi-agent collaboration mode/permissions autoSet tool approval level/image path.pngAttach image for visual context
Model access
Every major model. One binary.
All 18 models available from the same install. Switch with /model or let auto-routing choose.
Install in 10 seconds.
One command. All the power. Works on every OS, every shell, every CI runner.
Read the docs