Metraipolis › archive › dc_5223532bdf5c

Coding Agents are Strong Prompt Optimizers — arXiv:2609.26261

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.26261
License (as archived)
CC0-1.0
Archived
2026-10-08T13:45:42.531Z
SHA-256 of the archived text
149145143d3aa34bc97a78232fe88af6e5c482f98fedee183f1c3535a5d2ebc3
Stable URL of this page
https://metraipolis.com/documento/dc_5223532bdf5c

How to cite

Coding Agents are Strong Prompt Optimizers — arXiv:2609.26261 — arXiv (en), https://arxiv.org/abs/2609.26261 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5223532bdf5c (Metraipolis archive dc_5223532bdf5c, sha256 149145143d3aa34bc97a78232fe88af6e5c482f98fedee183f1c3535a5d2ebc3)
BibTeX
@misc{dc5223532bdf5c,
  title = {Coding Agents are Strong Prompt Optimizers — arXiv:2609.26261},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.26261},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5223532bdf5c; sha256 149145143d3aa34bc97a78232fe88af6e5c482f98fedee183f1c3535a5d2ebc3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.26261. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Search-based prompt optimizers improve prompts through iterative search: they propose edits, execute fresh rollouts, score the resulting trajectories, and retain only edits that improve a validation metric. We show that this optimization loop is unnecessary. Given only a static corpus of agent trajectories, an off-the-shelf coding agent can directly synthesize an optimized prompt, requiring neither environment access nor validation data. We call this approach \textit{Coding-Agent Skill Distillation} (CASD). The key insight is reflection scope. Rather than reasoning over a small batch of trajectories at each optimization step, the coding agent writes and executes analysis code to compute corpus-wide statistics, identifies systematic failure modes, inspects representative episodes, and distills the resulting insights into behavioral rules. Across four agentic benchmarks (ALFWorld, $\tau^2$-bench retail and telecom, and SpreadsheetBench-Verified), under matched data access, a single CASD pass outperforms GEPA, a state-of-the-art reflective prompt optimizer, on three of four benchmarks and outperforms validation-gated reflective search (SkillOpt) on all four, improving the unoptimized baseline by 16.6 percentage points on average versus 10.9 for GEPA and 5.3 for SkillOpt. Because CASD performs a single offline analysis pass rather than iterative search, producing an optimized prompt costs approximately \$1.60---over $22\times$ cheaper than validation-gated search. Even when competing methods are granted additional validation data and unrestricted environment access, CASD remains ahead on two of four benchmarks. These results suggest that corpus-scale statistical reflection is a viable alternative to iterative search for prompt optimization.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-23 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.26261 — 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 2026-10-08T13:18:38.162Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).