LiveClawBench: Benchmarking LLM Agents on Complex, Real-World Assistant Tasks

TL;DR

LiveClawBench is a new benchmark designed to evaluate LLM agents on complex, real-world assistant tasks by considering environment complexity, cognitive demand, and adaptability.

Contribution

It introduces a Triple-Axis Complexity Framework and a pilot benchmark to assess LLM agents in realistic, compositional assistant scenarios.

Findings

Benchmark covers real-world tasks with annotated complexity factors.

Framework enables evaluation of LLM agents across multiple difficulty dimensions.

Project page provides ongoing updates and expanded task collections.

Abstract

LLM-based agents are increasingly expected to handle real-world assistant tasks, yet existing benchmarks typically evaluate them under isolated sources of difficulty, such as a single environment or fully specified instructions. This leaves a substantial gap between current evaluation settings and the compositional challenges that arise in practical deployment. To address this gap, we introduce LiveClawBench, a benchmark to evaluate LLM agents on real-world assistant tasks. Based on an analysis of various real OpenClaw usage cases, we derive a Triple-Axis Complexity Framework that characterizes task difficulty along three dimensions: Environment Complexity, Cognitive Demand, and Runtime Adaptability. Guided by this framework, we construct a pilot benchmark with explicit complexity-factor annotations, covering real-world assistant tasks with compositional difficulty. Together, the framework and benchmark provide a principled foundation for evaluating LLM agents in realistic assistant settings, and establish a basis for future expansion across task domains and complexity axes. We are continuing to enrich our case collections to achieve more comprehensive domain and complexity coverage. The project page is at https://github.com/Mosi-AI/LiveClawBench.