Borrowing it
Nothing to install: this file belongs to mituan-ai/PaperBanana-CN. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/mituan-ai/PaperBanana-CN/main/.claude/skills/evaluate-diagram/SKILL.mdgit clone --depth 1 https://github.com/mituan-ai/PaperBanana-CNWrote 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.
[](https://agentmods.dev/skills/mituan-ai/paperbanana-cn/evaluate-diagram)<a href="https://agentmods.dev/skills/mituan-ai/paperbanana-cn/evaluate-diagram"><img src="https://agentmods.dev/badge/skills/mituan-ai/paperbanana-cn/evaluate-diagram/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.
<a href="https://agentmods.dev/skills/mituan-ai/paperbanana-cn/evaluate-diagram"><img src="https://agentmods.dev/badge/skills/mituan-ai/paperbanana-cn/evaluate-diagram.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00313 |
| Opus 5 | $0.00000 | $0.00156 |
| Sonnet 5 | $0.00000 | $0.00063 |
| Haiku 4.5 | $0.00000 | $0.00031 |
Grade A, and why
evaluate-diagram 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 12d 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.
This is a copy
92% identical to evaluate-diagram — 4 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
Evaluate Diagram
Evaluate a generated diagram against a human reference using PaperBanana's VLM-as-Judge scoring.
Instructions
$ARGUMENTS[0]is the path to the generated image.$ARGUMENTS[1]is the path to the human reference image.- Ask the user for:
- Source context: the methodology text (or a file path to read it from). If the user provides a file path, read that file to get the text.
- Figure caption: a description of what the diagram communicates.
- Call the MCP tool
evaluate_diagramwith:generated_path: the generated image pathreference_path: the reference image pathcontext: the methodology text contentcaption: the figure caption
- Present the evaluation scores to the user. Scores cover 4 dimensions: Faithfulness, Conciseness, Readability, and Aesthetics.
CLI Fallback
If the MCP tool is not available, fall back to the CLI:
paperbanana-cn evaluate --generated <generated-img> --reference <reference-img> --context <context-file> --caption "<caption>"
Example
/evaluate-diagram output.png reference.png
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.
- 12d ago First seen · 40 lines · 0 tokens per session scan A 8c87c47aa32d
evaluate-diagram is a skill published in the GitHub repository mituan-ai/PaperBanana-CN (79 stars, last pushed 8d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 313 tokens. A static security scan graded it A with 0 findings. It is 92% identical to evaluate-diagram, differing in 4 lines, and is treated as a copy.
Other skills, from other repositories
ip-as-logo
Generate extremely simple, cute, personified square character images with rounded heavy forms, two purposeful character colors, one solid background color, and a dominant lower-corner composition. Use when creating an animal, creature, robot, ghost, plant, object, or other character image, including when the agent…
ima2-uiux
Design direction discovery and UX judgment for ima2 users. Use for UI/UX direction and design judgment — vague visual briefs, onboarding, empty/error/loading states, layout vocabulary, typography breaks, favicons, logos, and brand identity choices. Pairs with ima2-front: this skill decides the design direction, then…
visio-image-rebuilder
Rebuild or restyle editable Microsoft Visio diagrams from reference images and existing .vsdx files, then export deliverables as .vsdx, PNG, SVG, PDF, or PPTX. Use when the user asks Codex to open Visio, recreate a diagram from a PNG/JPG/screenshot/reference image, match a scientific model figure, update…
xiaobei-skill-rebuild-image-in-powerpoint
Use when a user wants Codex to directly operate PowerPoint and reconstruct an entire reference image as editable PowerPoint objects while preserving its full layout, labels, connectors, and visual relationships. Do not use for paper-to-deck generation or VBA-only output.
oil-visual
A visual guide for creating bitmap images in a consistent oil-painting and manga-ink style. It supports either complete explanatory images with labels or transparent character illustrations for use in another layout.
signal-geometry
Create Signal Geometry abstract illustrations and posters from concepts or briefs. Use when the user asks for sparse geometric visuals, editorial posters, spatial systems, or prompt-only image recipes.