by suyoumo · Codex Skill · ★ 822
ClawProBench Transparent live-first benchmark harness for evaluating model capability inside the OpenClaw runtime. 102 active scenarios, 164 catalog scenarios, deterministic grading, and OpenClaw-native coverage. Paper now on arXiv: ClawProBench: Trace-Aware Evaluation of AI Agents with Runtime Coverage and Frozen Workplace-Style Holdouts. Dataset available on Hugging Face: xyh110sym/clawprobench. <a href="README.z
| Stars | 822 |
| Forks | 54 |
| Language | Rust |
| Category | Codex Skill |
| License | Apache-2.0 |
| Quality Score | 42.25/100 |
| Last Updated | 2026-08-25 |
| Created | 2025-03-02 |
| Platforms | python |
| Est. Tokens | ~210k |
These tools work well together with ClawProBench for enhanced workflows:
Explore other popular codex skill tools:
ClawProBench is ClawProBench is a live-first benchmark harness for evaluating LLM agents in the OpenClaw runtime with deterministic grading and repeated-trial reliability.. It is categorized as a Codex Skill with 822 GitHub stars.
ClawProBench is primarily written in Rust. It covers topics such as agent, benchmark, evaluation.
You can find installation instructions and usage details in the ClawProBench GitHub repository at github.com/suyoumo/ClawProBench. The project has 822 stars and 54 forks, indicating an active community.
ClawProBench is released under the Apache-2.0 license, making it free to use and modify according to the license terms.