Metraipolis › archive › dc_f38ca0dd1aaf
Prime Agent: A Self-Improving RLM Harness — arXiv:2608.23552
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.23552
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T07:42:01.443Z
- SHA-256 of the archived text
995826bc39865c20d52d223293eb955492f5457982a6a373f3f71e907f5e42fe- Stable URL of this page
https://metraipolis.com/documento/dc_f38ca0dd1aaf
How to cite
Prime Agent: A Self-Improving RLM Harness — arXiv:2608.23552 — arXiv (en), https://arxiv.org/abs/2608.23552 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f38ca0dd1aaf (Metraipolis archive dc_f38ca0dd1aaf, sha256 995826bc39865c20d52d223293eb955492f5457982a6a373f3f71e907f5e42fe)
BibTeX
@misc{dcf38ca0dd1aaf,
title = {Prime Agent: A Self-Improving RLM Harness — arXiv:2608.23552},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.23552},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f38ca0dd1aaf; sha256 995826bc39865c20d52d223293eb955492f5457982a6a373f3f71e907f5e42fe},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.23552. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Language models are sequential processors, but long-horizon agency requires external information and computation beyond model weights and active context. Prime Agent is an open-source harness for long-horizon evaluation and coding-agent workflows. A persistent IPython REPL follows the Recursive Language Model abstraction for programmatic context processing and test-time compute, while Continual Harness preserves histories, memories, skills, prompts, and subagent specifications across trajectories. Recursive subagents coordinate through direct agent-to-agent communication, and the Agents View lets humans inspect and manage daemon-backed sessions. Prime Agent standardizes execution, recovery, verification, and resource accounting while leaving strategy construction to the model. This low-friction, expressive membrane prevents harness failures from becoming model failures and pushes measurement toward the model's true maximal underlying capability. Prime Agent raises ARC-AGI-3 RHAE Best@1 from 30% to 95.5% and matches or exceeds native and popular harnesses across long-context coding, GPU-kernel generation, emulator construction, and autonomous nanoGPT speedruns. On Factorio, we find refinement allows for continuous technology progression and dedicated subagents enable parallelized work. Code is available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-25 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Software Engineering (cs.SE). Pagina del lavoro: https://arxiv.org/abs/2608.23552 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-18T07:37:52.786Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).