Evidence-first LLM-as-judge scoring for AI artifacts — papers, PRs, prompts, cold emails: synthesizes a rubric, collects quoted-evidence citations, scores only against that evidence, and hedges on thin evidence. Every dimension ties to a file:line or quote, with reproducibility receipts. 7 backends.
python citations reproducibility ai-safety ai-agents llm prompt-engineering evals llm-evaluation llm-judge ai-audit ai-reliability structured-scoring evidence-first hermes-labs
-
Updated
Aug 14, 2026 - Python