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 skills add alebgl77/claude-inc --skill web-artifactsgit clone --depth 1 https://github.com/alebgl77/claude-incWrote 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/alebgl77/claude-inc/web-artifacts)<a href="https://agentmods.dev/skills/alebgl77/claude-inc/web-artifacts"><img src="https://agentmods.dev/badge/skills/alebgl77/claude-inc/web-artifacts/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/alebgl77/claude-inc/web-artifacts"><img src="https://agentmods.dev/badge/skills/alebgl77/claude-inc/web-artifacts.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Prompt Injection · line 47 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
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 | $0.00098 | $0.01017 |
| Opus 5.5 | $0.00039 | $0.00407 |
| Sonnet 5.5 | $0.00020 | $0.00203 |
| Haiku 4.5 | $0.00010 | $0.00102 |
Grade A, and why
web-artifacts 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 1mo 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 — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Web Artifacts — Prototyper
"Live web prototypes"
When to use
- An idea needs to be felt, not described — "prototype this so I can click through it".
- A stakeholder demo is imminent — "I need something to show in an hour".
- A flow is contested — "would users even get this? build it and let's see".
- De-risking a real build — "prove the risky interaction before we commit the sprint".
Workflow
- Scope-cut first: what is the ONE interaction to prove? Write it as a sentence — "user drags expenses into buckets and the summary recalculates live". Everything not serving that sentence gets faked or cut.
- Choose the thinnest stack that proves it. Default is vanilla HTML/CSS/JS in one file. Add a CDN library (cdnjs only) solely when the core interaction demands one — charts, drag-and-drop, markdown rendering.
- Seed believable fake data. 8–15 records with realistic names, amounts and dates, plus at
least one edge case: the too-long string, the zero value, the overdue date. Declare it as
const DATA = [...]at the top of the script so reviewers can tweak it live. - Build mobile-first. One breakpoint upward if needed; 44px touch targets; the primary action sticky and reachable by thumb.
- Wire the interactive states. The core interaction genuinely works — click, drag, type, submit — with visible state change, an empty state, and a success or error state. Zero dead buttons: stub everything else with a toast ("Not in this prototype").
- Polish for ten minutes, not two hours. One accent color, one font stack, consistent spacing — enough that the idea is not judged on ugliness. This is not the design system pass.
- Ship. One .html file that opens by double-click or pastes straight into a Claude artifact, plus the three-line delivery note: proven / faked / test.
Output format
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Prototype — <the one interaction></title>
<!-- CDN deps only if the core interaction needs them:
https://cdnjs.cloudflare.com/ajax/libs/chart.js/4.4.1/chart.umd.min.js -->
<style>/* mobile-first; one accent; system font stack */</style>
</head>
<body>
<main><!-- the ONE interaction, fully wired --></main>
<script>
const DATA = [/* 8-15 seeded records, edge cases included */];
// state -> render() -> delegated event handlers, all inline
</script>
</body>
</html>
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.
- 1mo ago First seen · 87 lines · 98 tokens per session scan A 3f3e33a889fd
web-artifacts is a skill published in the GitHub repository alebgl77/claude-inc (19 stars, last pushed yesterday), licensed MIT. It adds 98 tokens to every session and 1,017 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-09-04.
Other skills, from other repositories
design-preview
A method for turning an existing or proposed interface into a self-contained HTML preview that can be opened and checked in Chrome.
site-preview
A workflow for turning HTML, Markdown, or design files into phone-friendly review pages and, when authorised hosting is configured, publishing them to a private site.
light-dark-mode
Load this skill whenever the project supports light/dark mode, colour theme switching, high-contrast mode, or responds to prefers-color-scheme. Under no circumstances hard-code colours that break in alternative themes. Absolutely always test colour contrast in both light and dark themes, and respect user OS-level…
building-landing-home
Compose a product or documentation home page from the @lobehub/ui/awesome landing components: LandingHero, AgentSkillCard, LogoMarquee, LandingSection, BentoGrid/BentoCard, CodeShowcase, FeatureGrid, InstallBanner and FluidGradient. Covers page structure, the lobedocs homePage hook, router links, theming the accent…
customer-journey-map
Construye un mapa de recorrido del cliente para un producto, servicio o experiencia. Úsalo cuando se te pida mapear un recorrido del cliente, crear un viaje del usuario, documentar puntos de contacto y puntos de fricción, o diseñar un mapa de experiencia. Produce un mapa de recorrido completo con etapas, puntos de…
design-handoff-brief
Transform feature briefs into structured design briefs that give designers the context they need before opening Figma. Use when asked to write a design brief, create a design handoff, brief a designer on a new feature, or translate a PRD into design requirements. Produces a brief with user goal, emotional context…