Hi all! We built HALO, an open-source local tool for debugging and improving agent harnesses from AI agent traces. It ingests Langfuse, Arize/OpenInference, or JSONL traces, uses an RLM-based engine to find recurring harness-level failure modes, and produces concrete reports you can feed into C...
Source: [Hacker News](https://github.com/context-labs/halo)