DarwinX: Evolving Agent Harnesses Through Natural Selection

Abstract. An LLM agent’s capability depends not only on model weights but on its harness: prompts, tools, skills, and control flow. Self-improvement loops already edit harnesses, yet single-lineage search is path-dependent and local wins often regress other tasks. We introduce DarwinX, which treats self-evolution as selection over a population of harnesses with the model frozen: a preserve-and-extend contract admits only variants that extend coverage without regressing, an archive keeps alternative lineages for recombination, and failure-, teacher-, and self-derived evidence share one edit interface. Fitness comes from each benchmark’s own verifier: no gold solutions, no hand-picked winners. Across four benchmarks that progressively separate the evolution signal from the test, one loop adds about 17 points on average: Terminal-Bench 2.1 rises +7.7 to 83.2% on a matched base and to the verified frontier at 84.7% on a stronger one; TerminalWorld’s held-out split reaches 68.3%, ahead of every off-the-shelf agent; WebArena-Infinity real-task pass@1 rises from 43.5% to 93.0% audit-clean; and a Terminal-Bench 2.1 harness transfers unchanged to SWE-bench Verified.

Introduction. The capability of a modern LLM agent is determined as much by its harness (the prompts, tools, memory, and control flow that mediate the model) as by the underlying model [11, 56]. A growing body of work makes this harness self-improving:1 the agent edits its own procedure and keeps changes that help. These systems span an axis of what they edit (prompts [12, 49, 54], skill documents [51], workflows [16, 58, 61], and finally the agent’s own source code, as in SICA [37] and the Darwin Gödel Machine (DGM) [56]) and an axis of how they search (single-lineage keep-best vs. population/archive; Table 1). Strikingly, almost all recent work converges on the 1The natural selection of our title is meant literally, not as a metaphor: no gold labels and no hand-picked winners, only survival of the fitter variant under measured fitness, with the model itself left unchanged (§2). same inner optimization loop: batch rollouts, reflect, propose a bounded edit, gate against a held-out/regression signal. Yang et al.

Discussion / Conclusion. Scope of the evidence. The strongest matched-model evidence comes from TB2.1 (75.5 → 83.2%) and WAI (43.5 →93.0% audit-clean), both with GPT-5.5 frozen; TerminalWorld adds a disjoint held-out task set, where the evolved harness reaches 28/41. Cross-benchmark transfer is measured in one direction only: a TB2.1-evolved harness reaches 84.2% on SWE-bench Verified, ahead of the fix-skill reference but inside a narrow band around it, so the transferred gain is far smaller than the in-domain ones (§7). DarwinX improves a frozen-model agent by selecting over a population of harness variants rather than training new weights. Three parts compose it: a preserve-and-extend contract that promotes a child only when it improves on some task without giving up what its parent solved; an archive that keeps alternative lineages so complementary specialists can be inherited and recombined; and a signal interface that turns failure-, teacher-, and self-derived evidence into harness edits.