WildClawBench

About WildClawBench

WildClawBench []() []() Hard, practical, end-to-end evaluation for AI agents — in the wild. WildClawBench is an agent benchmark that tests what actually matters: can an AI agent do real work, end-to-end, without hand-holding? We drop agents into a live OpenClaw environment — the same open-source personal AI assistant that real users rely on daily — and throw 60 original tasks at them: clipping goal highlights from a football match, negotiating meeting times over multi-round emails, hunting down contradictions in search results, writing inference scripts for undocumented codebases, catching...

Quick Facts

Compatible Skills

These tools work well together with WildClawBench for enhanced workflows:

• team-tasks — semantic(0.16)+complementary+same_lang+similar_pop+shared_platform (56%)
• AEnvironment — semantic(0.30)+complementary+same_lang+similar_pop+shared_platform (56%)
• get-physics-done — semantic(0.16)+complementary+same_lang+similar_pop+shared_platform (55%)
• agent-builder — semantic(0.15)+complementary+same_lang+similar_pop+shared_platform (55%)

More Codex Skill Tools

Explore other popular codex skill tools:

• openclaw ⭐ 374.0k
• hermes-agent ⭐ 162.8k
• ui-ux-pro-max-skill ⭐ 57.6k
• graphify ⭐ 51.5k
• open-design ⭐ 49.7k
• awesome-openclaw-skills ⭐ 49.2k
• cherry-studio ⭐ 46.1k
• siyuan ⭐ 44.1k
• nanobot ⭐ 43.0k
• system_prompts_leaks ⭐ 40.5k

Popular Python Agent Tools

• AutoGPT ⭐ 184.5k · Agent Tool
• hermes-agent ⭐ 162.8k · Codex Skill
• langflow ⭐ 148.7k · Agent Tool
• open-webui ⭐ 138.2k · MCP Server
• skills ⭐ 137.5k · Claude Skill

Frequently Asked Questions