Bil0000/Hill-Climb
Make your coding agent prove "done". Each change is kept only if a locked test or benchmark scores it better. Claude Code and Codex skill.
GitHub repository with 137 stars and 6 forks.
Language: Python
Topics: agent-skills, ai-agents, ai-coding, autonomous-agents, benchmarking, claude-code, claude-code-plugin, claude-skills, cli, codex