SkillMOO: Multi-Objective Optimization of Agent Skills for Software Engineering
Jingzhi Gong, Ruizhen Gu, Zhiwei Fei, Yazhuo Cao, Lukas Twist + 5 more
TLDR
SkillMOO optimizes LLM agent skill bundles for software engineering, boosting pass rates by 131% and cutting costs by 32% through automated evolution.
Key contributions
- Introduces SkillMOO, a multi-objective optimization framework for LLM agent skill bundles.
- Automates skill bundle evolution using LLM-proposed edits and NSGA-II survivor selection.
- Improves LLM agent pass rates by up to 131% and reduces costs by 32% on coding tasks.
- Reveals that minimal, focused skill bundles, achieved via pruning, are most effective.
Why it matters
Manually optimizing LLM agent skills is challenging. SkillMOO offers an automated, efficient solution to balance performance and cost for coding agents. This research provides a critical step towards more autonomous and effective LLM-powered software development tools.
Original Abstract
Agent skills provide modular, task-specific guidance for LLM- based coding agents, but manually tuning skill bundles to balance success rate, cost, and runtime is expensive and fragile. We present SkillMOO, a multi-objective optimization framework that automatically evolves skill bundles using LLM-proposed edits and NSGA-II survivor selection: a solver agent evaluates candidate skill bundles on coding tasks and an optimizer agent proposes bundle edits based on failure analysis. On three SkillsBench software engineering tasks, SkillMOO improves pass rate by up to 131% while reducing cost up to 32% relative to the best baseline per task at low optimization overhead. Pattern analysis reveals pruning and substitution as primary drivers of improvement, suggesting effective bundles favor minimal, focused content over accumulated instructions.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.