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.
npx agentmods add skills/nexus-research-lab/nexus/visualizenpx skills add nexus-research-lab/nexus --skill visualizegit clone --depth 1 https://github.com/nexus-research-lab/nexusWrote 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/nexus-research-lab/nexus/visualize)<a href="https://agentmods.dev/skills/nexus-research-lab/nexus/visualize"><img src="https://agentmods.dev/badge/skills/nexus-research-lab/nexus/visualize.svg" alt="Measured on agentmods" 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 | $0.00076 | $0.01919 |
| Opus 5 | $0.00038 | $0.00959 |
| Sonnet 5 | $0.00015 | $0.00384 |
| Haiku 4.5 | $0.00008 | $0.00192 |
Grade A, and why
visualize 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Nexus Generative UI
Create a custom interactive visual only when it communicates the answer better than ordinary prose or a table.
Workflow
- Call
show_widgetwith a concise title and one self-contained HTML fragment inwidget_code. - Put explanation and conclusions in the normal response around the widget. Do not repeat the visual as Markdown.
- The tool result only confirms delivery to the client. Do not claim that rendering succeeded.
Widget contract
- Return a fragment only. Do not include
html,head, orbodytags. - Inline the widget's CSS and JavaScript. SVG, Canvas, DOM, Web Components, and external CDN libraries are supported.
- Network access and external resources are allowed without a domain allowlist. Prefer established HTTPS CDNs.
- The fragment runs in an isolated iframe and cannot access the Nexus page, cookies, storage, or parent DOM.
- Streaming order is short style, visible content, then scripts last. Scripts run only after the complete tool input arrives. Keep native controls and static content useful before initialization.
- Before calling
show_widget, check every inline script for unmatched quotes, backticks, brackets, and incomplete blocks. Prefer short functions over one monolithic script. - Before calling
show_widget, verify that every id, class, or data attribute referenced by JavaScript exists in the submitted markup and is spelled identically. - Separate source data, derived calculations, rendering, and event binding. Avoid constructing large interfaces through nested template literals or long HTML string concatenation; update existing DOM nodes when practical.
- Use the smallest implementation that fully answers the request. Split independent visuals into separate
show_widgetcalls with short prose between them, but keep one cohesive widget when its views share state. - Make the layout responsive at 320px width and avoid fixed viewport dimensions.
- Use accessible labels, keyboard-operable controls, visible focus, and reduced-motion fallbacks.
- Keep interaction local to the widget. Do not assume a host API or
postMessagebridge.
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.
- 4d ago First seen · 111 lines · 76 tokens per session scan A eb6903098eea
visualize is a skill published in the GitHub repository nexus-research-lab/nexus (143 stars, last pushed today), licensed Apache-2.0. It adds 76 tokens to every session and 1,919 once invoked, about $0.0004 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.
Other skills, from other repositories
html-widget
生成可交互的 HTML 可视化组件 — 算法动画、数据图表、概念演示、交互式教程等。当用户需要交互式演示、动画解释、数据探索、UI 原型等场景时触发，直接输出 html 代码块，前端会在安全沙箱中渲染为可交互组件。.
mermaid-diagram
Use this skill when you need to draw diagrams including: Flowchart, Sequence Diagram, Class Diagram, State Diagram, ER Diagram, User Journey, Gantt Chart, Pie Chart, Quadrant Chart, Requirement Diagram, Gitgraph, Mindmap, Timeline, Sankey, XY Chart (Bar/Line), Block Diagram, Packet Diagram, Kanban, Architecture…
Atrium Desktop — driving apps
Drive the user's Atrium desktop: open files in installed viewer apps (Viv bioimages, Vitessce/Spatial 3D omics, Mol structures, IGV/Gosling genomics, Volume 3D, Cytoscape, MSA, PhyloTree, RDKit), read and steer ANY window — including ones the user opened — and call app backends.
excalidraw
Export Excalidraw-compatible JSON diagrams from natural language descriptions.
Architecture Diagram
Produces clean architecture and system diagrams from actual repo or product truth instead of decorative boxes.
Excalidraw
Shapes sketch-style diagrams and editable visual explanations for flows, product ideas, and architecture discussions.