ModHarness-RL: Recasting Harnesses as Learnable and Evolving Programs for Agentic RL A new arXiv paper (2611.18533v1), ModHarness-RL, recasts the harness around an agentic model as learnable programs that evolve during reinforcement learning training rather than staying hand-written and fixed. ModHarness-RL records interactions as structured State, a set of typed blocks, and treats two harness components as programs over those blocks: a Compression program that assembles accumulated history into the model's context and a Steering program that supplies guidance when behavior deviates; candidate programs are proposed from recent trajectories, verified by execution, and replayed over past trajectories. Tested on CodeSearch in depth plus ALFWorld and BIRD SQL with models from 1.5B to 8B, the Compression and Steering mechanisms improved held-out performance over the fixed-harness baseline in most settings. arXiv:2611.18533v1 Announce Type: cross Abstract: Agentic reinforcement learning RL increasingly relies on a harness, the scaffold around the model that decides what it sees of its own interaction and what guidance it receives. Most existing approaches train under a hand-written harness that stays fixed throughout. However, an effective harness depends on the domain, the data and the model, and keeps changing as the model trains. We introduce ModHarness-RL, which recasts the harness as programs that are learned from data and evolve as the model trains with RL. ModHarness-RL records the interaction as structured State, a set of typed blocks, and two core components of the harness are programs over these blocks, a Compression program that assembles the accumulated history into the model's context and a Steering program that supplies guidance when behavior deviates. During RL training, candidate programs are proposed from recent trajectories, verified by execution, and replayed over past trajectories, so the harness tracks how the model's behavior shifts. We study CodeSearch in depth and further analyze ALFWorld and BIRD SQL, with models from 1.5B to 8B. The proposed Compression and Steering mechanisms improve held-out performance over the fixed-harness baseline in most settings. These results suggest that a harness for agentic RL can be learned from the task's data and evolved with the model.