Decision information
Verified: 8/14/2026- Ecosystem category
- Research & Benchmarks
- Provider
- AIcell
- Content type
- Research
- Pricing
- Open Source
- Open source
- Yes
- License
- MIT
- Maintenance
- Active
- Install methods
- pip, Docker, Source
- Platforms
- macOS, Linux, Docker
- Interfaces
- Terminal, MCP, OpenClaw plugin
Security signals (informational, not a security certification)
Source availableRequires shellRequires credentials
Verification source: https://arxiv.org/abs/2604.18543 · arXiv:2604.18543
Overview
An open-source toolkit for generating, executing, and grading environments for Claw-like agents, paired with the Auto-ClawEval benchmark of 1,040 environments across 24 categories.
What it provides
- Generates agent environments from natural-language specifications
- Uses auditable mock services and deterministic verification
- Evaluates multiple Claw harnesses through plugin, MCP, and shell adapters
AllClaw classification
This project supports the Claw ecosystem as research benchmarks; it is not presented as a standalone Claw agent.
Provider: AIcell. Primary source: https://arxiv.org/abs/2604.18543.
The classification and core claims were checked against the linked primary source on August 14, 2026.
Tags
benchmarkevaluationenvironment-generationdatasetagent-harness