Run any Skill in Manus with one click

evoclawbench

EvoClawBench - Benchmark for evaluating LLM agent skill evolution capabilities. Tests whether agents can identify repeating patterns, create reusable skills, and effectively reuse them across sub-problems.

Run Skill in Manus

Overview

Install command

npx skills add https://github.com/pzy2000/EvoClawBench --skill evoclawbench

Copy and paste this command into Claude Code to install the skill

Source

pzy2000/EvoClawBench

Stars0

Forks0

UpdatedMay 26, 2026 at 10:45

File Explorer

100 files

SKILL.md

readonly

name	evoclawbench
description	EvoClawBench - Benchmark for evaluating LLM agent skill evolution capabilities. Tests whether agents can identify repeating patterns, create reusable skills, and effectively reuse them across sub-problems.

EvoClawBench

An open-source benchmark for evaluating LLM agents' ability to create and reuse skills (auto-evolution) at runtime.

Quick Start

uv run scripts/benchmark.py --model anthropic/claude-sonnet-4 --runtime nanobot --mode both

Key Concepts

Baseline mode: Agent is forbidden from creating skills
Evolution mode: Agent is encouraged to create reusable skills
fail2pass ratio: Measures the benefit of skill creation

Source

pzy2000

pzy2000/EvoClawBench

View GitHub Repository View Creator Repositories

Install command

Download

Run Skill in Manus

Useful forSOC

Software Quality Assurance Analysts and TestersComputer and Mathematical Occupations15-1253L4

Run any Skill with one click