XRepoSkill: Learning Transferable Skills for Software Engineering Agents

2026-09-29

AI Summary

This study addresses the challenge that skills acquired by software engineering agents are difficult to transfer across repositories, as agents often conflate effective behaviors with incidental habits. To overcome this limitation, this work proposes a novel trajectory-difference-based rule extraction mechanism that mines generalizable rules by contrasting successful and failed trajectories. Executable predicates are subsequently introduced to validate these rules, which are then consolidated into transferable skills to guide agents in solving tasks within unseen codebases. The primary contribution lies in achieving the crystallization and generalization of cross-repository skills. The proposed approach attains state-of-the-art performance on both the SWE-bench Pro and DeepSWE benchmarks, demonstrating a 10.3 percentage point improvement over the skill-free baseline on DeepSWE.

Abstract

Software engineering agents increasingly use reusable skills distilled from prior experience to resolve repository-level issues, yet such skills often fail to transfer across repositories. A central challenge is that a behavior appearing in a successful trajectory is not necessarily responsible for the successful outcome: it may be genuinely useful, merely incidental, or simply a recurring habit of the model. We introduce XRepoSkill, a trajectory-based approach for learning transferable skills. We represent a skill as a collection of rules, each specifying what action to take and when to take it during issue resolution. XRepoSkill first contrasts successful and failed trajectories of the same agent on the same issue and derives candidate rules from where their execution paths diverge. Each rule is paired with an executable predicate that enables its prescribed behavior to be evaluated systematically on other trajectories. A rule is verified based on its association with successful issue resolution and retained only when its prescribed behavior recurs across multiple repositories; repository-specific variants of the same behavior are then consolidated into transferable rules. For a new issue, XRepoSkill selects relevant rules to guide the agent. We learn skills from publicly released trajectories on the official SWE-bench Verified leaderboard and evaluate them on SWE-bench Pro and DeepSWE using three backbone LLMs from different vendors; none of the evaluation repositories appears in the skill-learning trajectory pool. Against three recent skill learning methods, XRepoSkill achieves the highest issue resolution rate in all six benchmark--LLM combinations. In particular, on the challenging long-horizon DeepSWE benchmark, XRepoSkill improves issue resolution by 10.3 percentage points over the same agent without learned skills and by 5.0 points over the strongest skill-learning baseline.

Problem

• software engineering agents
• transferable skills
• cross-repository
• trajectory analysis
• skill learning

Innovation

• transferable skills
• trajectory contrast
• software engineering agents
• executable predicates
• cross-repository