ZJU-REAL/SkillZero

Official code for "Self-Distilled Agentic Reinforcement Learning"

[NeurIPS'26] SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning

Skill training with a corruption and reconstruction loop: train an agent's skill library, not its weights.

Awesome List for Agentic RL

Democratizing Reinforcement Learning for LLMs

EvoSkill — An open-source framework that automatically discovers and synthesizes reusable agent skills from failed trajectories to improve coding agent performance.

SkillX: Automatically Constructing Skill Knowledge Bases for Agents

[ICML-2026] Official implementation of "SEAgent: Self-Evolving Computer Use Agent with Autonomous Learning from Experience"

Official code for Agent-Reflex: Teaching Language Agents to Act Reflectively