evaluate-grader skillA
evaluate-grader is agent-read markdown (skill) from ai-analyst-lab/ai-analyst: Compare a narrow model grader with frozen human labels and inspect disagreement, bias probes, and repeated scoring stability. Use when building or changing a model-based evaluator..
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file 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. Keep human labels outside the judge workspace. Use `python3 -m helpers.evals.cli run-isolated-judge` to create a fresh child workspace containing only the named examples and current rubric. Run every revised rubric in a different fresh child. A child must not receive human labels, prior verdicts, captured verdicts, or later rubric versions. Preserve the isolation record with the verdicts. 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. …
Read the whole file at its exact version.
How to install
mdr add ai-analyst-lab/ai-analyst/evaluate-grader@git:20260920.18d2675mdr add ai-analyst-lab/ai-analyst/evaluate-grader@sha256:64fb96fc239f015bPin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_lajytpzdymgzypu5)
1 badge views in 30 days
Versions
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20260920.18d2675 latest | 2026-09-20 | 18d2675 | 1,560 B | A | view · diff |
| git:20260910.c4b91ef | 2026-09-10 | c4b91ef | 1,167 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (1560 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
ai-analyst-lab/ai-analyst · 302 stars · license MIT · pushed 2026-09-22 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_lajytpzdymgzypu5 GET https://markdownregistry.com/api/v1/resolve?ref=ai-analyst-lab/ai-analyst/evaluate-grader GET https://markdownregistry.com/api/v1/blob/64fb96fc239f015beea02a5946bf56bc73f7f05b12a9527a048c9de1966a489c
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.