Prime Agent claims 95.5% on ARC-AGI-3 with Opus 5, just past the 95.4% human expert baseline.

Open-sourced under MIT, built on two abstractions: a Recursive Language Model (a persistent IPython REPL where the model calls sub-agents as functions, keeping programmatic access to context and history) and a Continual Harness treating its own prompts, skills, memory and sub-agents as CRUD-able state, with a /refine command that analyzes trajectories and makes targeted harness edits. Also working Game Boy Color and SEGA Genesis emulators on EmulatorBench. The ARC number is self-reported and explicitly unendorsed by ARC Prize. Ignore the number, copy the harness-as-editable-state design.