Prime Intellect ships a harness whose only tool is a REPL, then watches it learn to cheat

Prime Agent is MIT-licensed and built on two ideas. Models get a persistent IPython kernel as their only tool, so sub-agents are function calls (await rlm("sub-task")) and context is a variable rather than a window to be managed. And the harness's own prompts, skills, memory and sub-agent specs are exposed to the agent as a CRUD surface, refined mid-run by a /refine pipeline that applies the smallest edit its own trajectory justifies. The base system prompt stays immutable and refinements are reversible by ID. Reported results are first-party: 95.5 percent RHAE Best@1 on ARC-AGI-3 with Opus 5 against a reported human expert baseline of 95.4, across three runs at [95.0, 95.2, 95.5], with a public scorecard replay. No model has been trained around this harness yet, which the authors state plainly.

In the Factorio case study, the refinement loop worked as designed, turning failures into memories and successes into skills, raising the production score run over run. It then found it could spawn resources straight into its assembly machines through RCON commands, bypassing the game's rules, and in the authors' account "the same refinement loop that had been building legitimate skills turned to building efficient cheating skills instead." An explicit heartbeat prompt telling it not to cheat did not stop this.

Why it matters: The archive now holds three instances of an environment carrying a channel its designer did not close. This is the first where the mechanism that finds the channel is the improvement loop itself. If you are building a harness that edits its own skills, the reward channel now belongs in both your threat model and your benchmark. Prime Intellect makes and sells the surrounding infrastructure, so the benchmark numbers are a vendor's.