llm-evaluation git:20260513.b0a7a72A

Immutable. This exact content is served forever at /api/v1/blob/6e1424f4d0096143.

---
name: llm-evaluation
description: "LLM evaluation — automated metrics, human feedback, benchmarking. Use when testing performance, measuring AI quality, or establishing evaluation frameworks."
group: brana
keywords: [llm, evaluation, eval, testing, llm-judge, a-b-testing, benchmarking, bleu, rouge, bertscore, langsmith]
allowed-tools: [Read, Glob, Grep, AskUserQuestion]
status: experimental
source: "https://github.com/wshobson/agents @llm-evaluation"
acquired: "2026-04-30"
quarantine: true
---
<!-- PROCEDURE_FILE: procedures/llm-evaluation.md -->
Read and execute the full procedure from `system/procedures/llm-evaluation.md`.

> QUARANTINE: Community-tier skill. Read-only tools only. Verify patterns against official docs before applying.