machine-learningdata-ai
llm-evaluator
LLM-as-a-Judge evaluation system using Langfuse. Score AI outputs on relevance, accuracy, hallucination, and helpfulness. Backfill scoring on historical traces. Uses GPT-5-nano for cost-efficient judging. Use when evaluating AI quality, building evals, or monitoring output accuracy.
maintainer
openclaw
अपडेट किया गया 3/5/2026
स्टार
4001
फोर्क
1095
quick start
Installation and usage
LLM-as-a-Judge evaluation system using Langfuse. Score AI outputs on relevance, accuracy, hallucination, and helpfulness. Backfill scoring on historical traces. Uses GPT-5-nano for cost-efficient judging. Use when evaluating AI quality, building evals, or monitoring output accuracy.
इंस्टॉलेशन
$ install --globalskills.sh
उपयोग
इंस्टॉल करने के बाद, आप टर्मिनल में यह कमांड चलाकर इस स्किल का उपयोग कर सकते हैं:
skills use llm-evaluator