← All skills

calibrate-an-llm-judge-against-human-labels

80 / 100 · Low risk

Fine to try.

By jiayaoqijia/cryptoskill (GitHub) · other · 81 stars · scanned at commit 0dc9f60

Use when replacing or supplementing human ratings with an LLM judge. Measures agreement, per-class error, and drift before the judge decides anything.

Works with any agent that supports SKILL.md skills.

What this skill can do

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'calibrate-an-llm-judge-against-human-labels' ] || [ -L ~/.agents/skills/'calibrate-an-llm-judge-against-human-labels' ]; then echo 'A skill folder named calibrate-an-llm-judge-against-human-labels already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/jiayaoqijia/cryptoskill' "$d" && git -C "$d" checkout --quiet '0dc9f60ab1f3fcae24da49c4cef09c46c589f903' -- 'skills/payments/0xklen-calibrate-an-llm-judge-against-human-labels' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/payments/0xklen-calibrate-an-llm-judge-against-human-labels' ~/.agents/skills/'calibrate-an-llm-judge-against-human-labels' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'calibrate-an-llm-judge-against-human-labels' ~/.claude/skills/'calibrate-an-llm-judge-against-human-labels' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for jiayaoqijia/cryptoskill shows the score of its lowest-scoring skill.

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/jiayaoqijia/cryptoskill.svg)](https://zalavijaysinh.github.io/skilltrust/repos/jiayaoqijia/cryptoskill/)

Score history