Tencent-Hunyuan/UniRL

UniRL is a Framework for Unified Multimodal Model Reinforcement Learning

Latest release v0.2.0 — UniRL v0.2.0 · 8 Oct 2026

5 files for Claude Code, Codex and OpenCode: add-model-bundle, pr-workflow, UniRL AGENTS.md, pr-review and 1 more — 1,583 tokens loaded in every session.

Open on GitHub unirl-project.github.io

Instructions

UniRL AGENTS.md AGENTS.md A 1,583 tok

Skills

add-model-bundle .claude/skills/development/add-model-bundle/SKILL.md A 62 tok
pr-workflow .claude/skills/development/pr-workflow/SKILL.md A 62 tok
pr-review .claude/skills/development/pr-review/SKILL.md A 60 tok
readme-guide .claude/skills/development/readme-guide/SKILL.md A 59 tok

These files are Tencent-Hunyuan/UniRL's own configuration — they tell Claude Code, Codex and OpenCode how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.