evaluate-grader

evaluate-grader is a skill for Claude Code from ai-analyst-lab/ai-analyst. It costs 35 tokens per session (224 once invoked), scanned A, original, MIT.

A checker for a model-based grader, which is an AI system that scores other answers. It compares the grader with fixed human labels and examines disagreements, bias checks, and scoring consistency.

In plain words
What is it for?
Use it while building or changing an evaluator to inspect disagreements, test answer order and verbosity effects, handle ambiguous cases, and check repeated scoring stability.
Why use it?
It helps reveal when an automated grader disagrees with people, favours certain answer styles, or gives unstable scores.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

Good fit Use it while building or changing an evaluator to inspect disagreements, test answer order and verbosity effects, handle ambiguous cases, and check repeated scoring stability.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/ai-analyst-lab/ai-analyst/evaluate-grader
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add ai-analyst-lab/ai-analyst --skill evaluate-grader
Clone the repo
git clone --depth 1 https://github.com/ai-analyst-lab/ai-analyst

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for evaluate-grader

README.md
[![agentmods](https://agentmods.dev/badge/skills/ai-analyst-lab/ai-analyst/evaluate-grader/github.svg)](https://agentmods.dev/skills/ai-analyst-lab/ai-analyst/evaluate-grader)
Your own site
<a href="https://agentmods.dev/skills/ai-analyst-lab/ai-analyst/evaluate-grader"><img src="https://agentmods.dev/badge/skills/ai-analyst-lab/ai-analyst/evaluate-grader/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for evaluate-grader

Your own site · 80×15
<a href="https://agentmods.dev/skills/ai-analyst-lab/ai-analyst/evaluate-grader"><img src="https://agentmods.dev/badge/skills/ai-analyst-lab/ai-analyst/evaluate-grader.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 35 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 224 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00035 $0.00224
Opus 5 $0.00017 $0.00112
Sonnet 5 $0.00007 $0.00045
Haiku 4.5 $0.00003 $0.00022

Measured 2d ago against content hash 784609c49b4d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-13, from the pricing page.

Security

Grade A, and why

evaluate-grader scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/evaluate-grader/SKILL.md · 24 lines

What it actually says

Evaluate a grader

Freeze the human labels before running the grader. Use one narrow criterion with a written rubric and structured output. The grader must be able to return unknown or request human review.

Use helpers.evals.judges.evaluate_alignment for the confusion table and disagreement set. Repeat at least one unchanged boundary example and use repeated_label_stability to measure scoring stability.

Inspect:

  • every human and grader disagreement;
  • label imbalance;
  • an answer-order reversal when judging pairs;
  • a verbosity trap where a longer answer is not the better answer;
  • an ambiguous example that should produce unknown; and
  • whether the generator and grader are actually independent contexts.

Revise one rubric criterion at a time and rerun only the working examples. Do not tune on the heldout judge set.

Call the classroom result an alignment check. A small agreeing sample is not completed calibration.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 24 lines · 35 tokens per session scan A 784609c49b4d

Subscribe to this mod's changes

evaluate-grader is a skill published in the GitHub repository ai-analyst-lab/ai-analyst (298 stars, last pushed 3d ago), licensed MIT. It adds 35 tokens to every session and 224 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-12.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

chronicle

Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…

microsoft/vscode · 72 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens