mllm-eval

mllm-eval is a skill for Claude Code from Aperivue/medsci-skills. It costs 140 tokens per session (1,729 once invoked), scanned A, original, MIT.

An evaluation-design guide for language models and models that work with both text and images on clinical tasks, such as writing radiology reports or extracting information from medical text.

In plain words
What is it for?
Use it to plan reference standards, clinical metrics, contamination checks, prompt-sensitivity reporting, and reader studies where people assess generated text.
Why use it?
It helps ensure that results reflect clinical accuracy and factual faithfulness, rather than only matching words in a reference answer.

Skill for Claude Code

Written for Claude Code: ${CLAUDE_SKILL_DIR} variable. Also seen: model in frontmatter.

Part of the medsci-modeling plugin — 12 skills shipped together

Good fit Use it to plan reference standards, clinical metrics, contamination checks, prompt-sensitivity reporting, and reader studies where people assess generated text.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/aperivue/medsci-skills/mllm-eval
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 Aperivue/medsci-skills --skill mllm-eval
Clone the repo
git clone --depth 1 https://github.com/Aperivue/medsci-skills

Made for: Claude Code.

Or install medsci-modeling, the plugin that ships this one along with the rest of its 12 skills.

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 mllm-eval

README.md
[![agentmods](https://agentmods.dev/badge/skills/aperivue/medsci-skills/mllm-eval/github.svg)](https://agentmods.dev/skills/aperivue/medsci-skills/mllm-eval)
Your own site
<a href="https://agentmods.dev/skills/aperivue/medsci-skills/mllm-eval"><img src="https://agentmods.dev/badge/skills/aperivue/medsci-skills/mllm-eval/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 mllm-eval

Your own site · 80×15
<a href="https://agentmods.dev/skills/aperivue/medsci-skills/mllm-eval"><img src="https://agentmods.dev/badge/skills/aperivue/medsci-skills/mllm-eval.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 140 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,729 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. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 $0.00140 $0.01729
Opus 5 $0.00070 $0.00864
Sonnet 5 $0.00028 $0.00346
Haiku 4.5 $0.00014 $0.00173

Measured 11d ago against content hash afdc167bfccb, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

mllm-eval 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 11d ago.

The scan reads SKILL.md. This mod also ships 3 executable files (scripts/check_mllm_eval_completeness.py, scripts/mllm_eval_completeness_challenge/verify.sh, tests/test_mllm_eval_completeness.sh), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

skills/mllm-eval/SKILL.md · 117 lines

How it starts

The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MLLM-Eval Skill

Purpose

This skill makes an LLM / MLLM clinical evaluation defensible: a real adjudicated reference standard, faithfulness measured not assumed, clinical-efficacy metrics beyond n-gram overlap, a pretraining- contamination check, prompt-sensitivity disclosed, and a reader study where text is generated. It is model-agnostic — every check applies to a closed API and to open weights — and read-only (an advisory design/audit skill): it audits the evaluation design and specifies and routes the clinical-efficacy metrics (RadGraph-F1 / CheXbert-F1 via their published extractors) rather than running the model or computing the metrics itself.

It is the LLM/MLLM evaluation-design counterpart in the lane — an auditor that hands the specified metrics to their extractors and /analyze-stats, parallel to how /model-validation audits an imaging model's design (the imaging metrics themselves are computed by /model-evaluation). The reviewer-side audit of a finished manuscript uses the mllm_evaluation.md (ME0–ME8) probe via /self-review and /peer-review; this skill is the author-side harness design. It routes the reader study to /design-ai-benchmarking, the sizing to /calc-sample-size, and TRIPOD-LLM / MI-CLEAR-LLM compliance to /check-reporting.

When to use

  • You are designing or auditing an evaluation of an LLM/MLLM on a clinical task and want it to cover the axes a reviewer will check (reference standard, faithfulness, contamination, prompt sensitivity, reader study).

When NOT to use

  • AI-vs-human-expert benchmark with a rated rubric → /design-ai-benchmarking.
  • Imaging prediction/segmentation model → /model-evaluation + /model-validation.
  • Image-to-image generative model → the image_synthesis probe.
  • Training / serving the LLM → out of scope.
  • Item-level TRIPOD-LLM / MI-CLEAR-LLM audit of a finished manuscript → /check-reporting.

Workflow

Phase 1 — Pin the task, model, comparator, decoding (ME0)

State the task (report generation / VQA / extraction-classification), the exact model + version/date (closed API or open-weights id), the decoding settings (temperature, seed, max tokens), and what the outputs are scored against.

Read the full file on GitHub · 117 lines

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. 11d ago First seen · 117 lines · 140 tokens per session scan A afdc167bfccb

Subscribe to this mod's changes

mllm-eval is a skill published in the GitHub repository Aperivue/medsci-skills (291 stars, last pushed 3d ago), licensed MIT. It adds 140 tokens to every session and 1,729 once invoked, about $0.0007 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 2026-08-30.

Related

Other skills, from other repositories

torch-geometric

PyTorch Geometric (PyG) for graph neural networks — node/link/graph classification, message passing (GCN, GAT, GraphSAGE, GIN), heterogeneous graphs, neighbor sampling, and custom datasets. Use when working with torchgeometric, not for general NetworkX analytics or non-graph PyTorch models.

K-Dense-AI/scientific-agent-skills · 71 tokens

bids

Use this skill when working with Brain Imaging Data Structure (BIDS) datasets: organizing neuroscience and biomedical data (MRI, EEG, MEG, iEEG, PET, microscopy, NIRS, motion capture, EMG, MR spectroscopy, behavioral), querying BIDS layouts, validating compliance, converting DICOM to BIDS, writing metadata sidecars…

K-Dense-AI/scientific-agent-skills · 80 tokens

bulk-rnaseq

End-to-end bulk RNA-seq orchestrator — takes raw FASTQ reads through QC and trimming (FastQC, fastp/Trim Galore), alignment and quantification (STAR, Salmon, featureCounts), assembles a gene-level counts matrix, then hands off to differential expression (pydeseq2), pathway/GSEA enrichment (pathway-enrichment), and…

K-Dense-AI/scientific-agent-skills · 218 tokens

aeon

This skill should be used for time series machine learning tasks including classification, regression, clustering, forecasting, anomaly detection, segmentation, and similarity search. Use when working with temporal data, sequential patterns, or time-indexed observations requiring specialized algorithms beyond standard…

K-Dense-AI/scientific-agent-skills · 74 tokens

esm

Use when working directly with the esm Python SDK, ESM3 or ESMC model IDs, Forge/Biohub inference clients, or ESMFold2 folding workflows.

K-Dense-AI/scientific-agent-skills · 39 tokens

geniml

Use Geniml for audited local genomic-interval workflows: validate BED and universe contracts, plan Region2Vec or scEmbed runs, inspect model/tokenizer compatibility, and assess consensus universes.

K-Dense-AI/scientific-agent-skills · 43 tokens