Agent Skills by enkez
Showing 11 vetted skills indexed across 1 GitHub repositories.
EnkeZmodel-integration
Integrate model architectures into the Relax training framework with Megatron and FSDP backends.
EnkeZgit-commit
Generate and enforce Conventional Commits with Markdown descriptions for git commit messages.
EnkeZcreating-skills
Create Claude Code skills following Anthropic's official best practices.
EnkeZredaccel-to-relax
Convert RedAccel RL training algorithms to the Relax framework.
EnkeZdoc-writer
Create and maintain bilingual English and Chinese documentation for the Relax project.
EnkeZverl-to-relax
Migrate RL training recipes from verl to Relax frameworks.
EnkeZcode-review
Automate expert code reviews for the Relax reinforcement learning framework.
EnkeZdebug-hang
Diagnose Ray distributed training hang issues by analyzing cluster status and blocking patterns.
EnkeZperf-doctor
Analyze Relax training scripts for performance and memory misconfigurations.
EnkeZssh-ray-cluster
Execute commands and inspect logs on a Ray cluster head node via SSH.
EnkeZrelax-dev-debug
Automate development and debugging for Relax reinforcement learning on remote Ray clusters.