OpenSkill: Open-World Self-Evolution for LLM Agents
This work addresses the problem of enabling LLM agents to autonomously improve after deployment in real-world scenarios where no curated learning signals are available, which is a critical bottleneck for practical autonomous agents.
OpenSkill enables LLM agents to self-evolve in open-world deployments without any target-task supervision, achieving the best automated pass rate across three benchmarks by bootstrapping skills and verification signals from open-world resources.
Self-evolving agents requires adaptation after deployment, but existing approaches assume a usable learning loop, such as curated skills, successful trajectories, or verifier signals. Real open-world deployments may provide none of these, offering only a task prompt. In this work, we study open-world self-evolution, where an agent must build both its skills and its own verification signals from scratch, using open-world resources but no target-task supervision. We propose OpenSkill, a framework that bootstraps this loop: it acquires grounded knowledge and verification anchors from documentation, repositories, and the web, synthesizes them into transferable skills, and refines those skills against self-built virtual tasks grounded in the anchors rather than in target answers. The open world thus supplies both the knowledge to be learned and a supervision-independent practice environment, with target-task supervision reserved for final evaluation. Across three benchmarks and two target agents, OpenSkill attains the best automated pass rate while satisfying the no-supervision constraint. Analysis shows its skills transfer across models without model-specific adaptation, and its self-built verifier aligns with ground-truth outcomes despite never accessing them.