Open-source AI Agent platform for teams. Your agents don't just chat — they read files, run code, call APIs, and deliver results.
Terminal score: 0–1000 raw, weighted across 4 dimensions. Public score: 0–10 normalized (shown in the 30-day stars chart above).