AgentCompass: A Unified Evaluation Infrastructure for Agent Capabilities
For researchers developing LLM-based agents, AgentCompass reduces engineering overhead and improves reproducibility in agent evaluation.
AgentCompass provides a unified, open-source evaluation infrastructure for LLM-based agents, addressing fragmentation and reproducibility issues by decoupling benchmarks, harnesses, and environments, supporting over 20 benchmarks across five capability dimensions.
As Large Language Models (LLMs) evolve into autonomous agents, the need for unified evaluation infrastructure becomes critical. However, current evaluation pipelines remain highly fragmented and tightly coupled, hindering reproducibility and causing redundant engineering. To address this, we introduce AgentCompass, an open-source, lightweight, and extensible infrastructure for evaluating LLM-based agents. AgentCompass organizes the evaluation process around three independent components, namely Benchmark, Harness, and Environment, thereby enabling flexible configurations without requiring the reimplementation of complex execution logic. Furthermore, it features a fault-tolerant asynchronous runtime and comprehensive trajectory analysis tools to transparently diagnose nuanced failure modes like reward-hacking. Natively supporting over 20 benchmarks across five capability dimensions, AgentCompass provides the community with a scalable and reproducible infrastructure for advancing agent research.