Full-stack AI agent for coding, files, terminals, tools, research, content, multimodal work, and multi-agent workflows.
Terminal score: 0–1000 raw, weighted across 4 dimensions. Public score: 0–10 normalized (shown in the 30-day stars chart above).