Local-first AI agent workspace for coding, writing, design, research, and automation — one runtime for desktop GUI and TUI.
Terminal score: 0–1000 raw, weighted across 4 dimensions. Public score: 0–10 normalized (shown in the 30-day stars chart above).