Skip to content
SkillModel-led1,082 tokens in use

html-prototype

Builds a polished HTML mockup or a working interactive prototype of one credible flow, with realistic states, keyboard support and responsive layout, in a single file.

4,275 stars on the repository MIT Licenseplannotator/effective-html · skills/html-prototype
npx skills add plannotator/effective-html --skill html-prototype
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing html-prototype costs

87tokens idle
1,082tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Fair
  • Frontmatter12/15
  • Structure16/25
  • Examples0/15
  • Output15/15
  • Size15/15
  • Resources0/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Fair
  • Development activity60
  • Maintenance24
  • Community48
  • Maturity24
  • Momentum40

html-prototype is a specialist skill for testing a product decision in the browser. The agent chooses between a mockup, when the open question is visual hierarchy or fit, and a prototype, when the question is navigation, input, state change or recovery.

It derives the visual direction from your conversation, references and project, scopes the smallest flow that answers the review question, lists the states that flow can reach, and verifies them at desktop and mobile widths. Delivery is one self-contained HTML file.

What You Get

  • One self-contained HTML file with inline CSS and JavaScript and no build step
  • A mockup or prototype mode chosen from the open question
  • Realistic content and the reachable states: loading, empty, error, success, disabled and mobile
  • Keyboard support, visible focus and dialog behavior with focus restoration
  • A handoff naming the scenario, the states built and the production behavior left out

Common Use Cases

  • Testing a checkout or onboarding flow before engineering starts
  • Showing stakeholders a polished layout of a new screen
  • Exploring how a form should validate and recover from errors
  • Comparing a static mockup and a working prototype that share the same content

What Makes It Unique

It keeps fidelity matched to the question. A mockup stays mostly static and a prototype implements one flow end to end, with dead buttons removed and out-of-scope actions explained. It also checks computed colors on every surface and says which checks remain unverified when browser tooling is missing.

Under The Hood

The skill is a single instruction file with no scripts. The agent reads project design documents and writes one HTML file to your workspace. Verification asks it to open the page in a browser, resize it, press keys, read the console and test reduced motion. The delivered prototype needs no authentication, live API or external service.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers