A³-Skill: Learning Autonomous Management of Executable Tool Skills via Pure Reinforcement Learning

Large language models (LLMs) increasingly rely on external tools to solve complex tasks. Tool skills enable LLMs to reuse procedural knowledge from past tool interactions. Existing approaches, however, often rely on fragmented skill learning, skill-operation supervision, or natural-language skill representations. We study autonomous tool-skill management: jointly acquiring, applying, and adapting executable tool skills within a single policy. We propose A${}^3$-Skill, a two-stage reinforcement learning framework without supervised skill operations. In the first stage, we adopt a progressive curriculum that gradually transitions from reference-guided skill construction to autonomous skill acquisition and refinement. In the second stage, a leave-one-scenario-out strategy trains the policy to retain, revise, or rebuild imperfectly matched transferred skills through successive task interactions. We evaluate A${}^3$-Skill on AppWorld across independent transfer, sequential adaptation, and cold-start acquisition, and additionally assess cross-benchmark generalization on BFCL-v3. On AppWorld, A${}^3$-Skill improves TGC over the strongest competing baseline by 13.63% under sequential adaptation and 15.78% under cold-start acquisition. It also achieves 32.00% accuracy on BFCL-v3, compared with 28.00% for the strongest baseline.