web-component-inspiration
Skill1 starsChanged 16 days agoMITView on GitHub
Find and assess web UI component references for a feature.
In this repository57 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-loginSkill
- animateSkill
- app-intentsSkill
- app-store-connectSkill
- architectSkill
- arenaSkill
- blast-radiusSkill
- codebase-designSkill
- code-qualitySkill
- correctSkill
- diagnosing-bugsSkill
- domain-modelingSkill
- eli5Skill
- engineering-principlesSkill
- gauntlet-loopSkill
- gh-stackSkill
- goal-swarmSkill
- grillingSkill
- grill-with-docsSkill
- handoffSkill
- howSkill
- improve-codebase-architectureSkill
- ios-hapticsSkill
- koubouSkill
- mergeSkill
- orchestrationSkill
- prototypeSkill
- prSkill
- react-best-practicesSkill
- react-doctorSkill
- reframeSkill
- repo-cleanupSkill
- researchSkill
- retroSkill
- review-fix-loopSkill
- reviewSkill
- review-sweepSkill
- seoSkill
- setup-sreestackSkill
- show-dont-tellSkill
- showroomSkill
- snipsnap-do-choresSkill
- snip-snapSkill
- swift-testing-modernizationSkill
- swiftuiSkill
- tddSkill
- to-specSkill
- to-ticketsSkill
- triageSkill
- ui-designSkill
- uikitSkill
- verifySkill
- wayfinderSkill
- whySkill
- writing-for-agentsSkill
- xcode-security-auditSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 28
Use ui-design to judge visual fit and browser behavior or accessibility.
- Distinctive: line 33
Read sources.md. Honor a source the user names.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Web Component Inspiration
Find web patterns that could make a real feature clearer, more useful, or more distinct. Treat “the current design is stronger” as a valid result.
Frame the opportunity
Inspect the request, current interface, and nearby code. Write a short fit brief covering the feature’s job, location, web stack, existing visual and interaction rules, and material constraints such as accessibility, motion, performance, package policy, or paid code.
This catalog is web-only. For a SwiftUI or UIKit target, state that boundary
and use ui-design with swiftui or uikit instead. Consult these web sources
in sources.md only when the user explicitly wants
mechanic-level inspiration across platforms, and label the fidelity limits.
Identify only the parts where a different pattern could improve a user outcome, such as clearer feedback, faster scanning, spatial context, or a deliberate moment of delight. Drop opportunities that add novelty without value.
Use ui-design to judge visual fit and browser behavior or accessibility. Use
animate when motion is part of the opportunity.
Scout live sources
Read sources.md. Honor a source the user names; otherwise choose the source groups that fit the opportunity and stack. Follow the linked guide when selecting Kinetics or Forever Components. Search other catalogs live using the interface role, user action, visual character, and mechanic rather than relying on a component’s marketing name.
Build a candidate pool broad enough to reveal distinct tradeoffs. Open the exact demo or documentation page for every serious candidate. Then read source-code.md, inspect the public source without changing the project, and record:
- what the live example does;
- the exact demo, documentation, and source URLs, plus a revision for repository source;
- its framework, packages, assets, and setup cost;
- whether access is free, paid, or unclear, and the current reuse terms before suggesting a direct port;
- keyboard, touch, focus, reduced-motion, and narrow-screen gaps;
- likely runtime, rendering, or bundle costs.
Group candidates by the user-facing pattern, not the source. Keep alternatives only when they expose a real tradeoff, such as lower implementation cost, better accessibility, or a quieter visual treatment. Rank product fit first, then clarity, implementation cost, accessibility, and performance. Use novelty only to break a close tie.
Present the shortlist
Return only enough checked picks to expose the useful choices. Honor a count the user specifies when that many genuinely distinct, checked candidates exist; otherwise return the supported set and explain the shortfall. For each, include:
- the opportunity it serves;
- the component and source with a direct demo link and a public source link when available;
- how its code can be inspected or integrated, or what access is missing for an inspiration-only pick;
- why it fits this product;
- the mechanic or behavior worth borrowing;
- implementation cost, packages, and assets;
- cost, terms, access, motion, or speed concerns.
End with a first choice and a plain “skip it” option when the gain looks weak. For an ideas-only request, leave the project unchanged and let the user choose.
Adapt a chosen pattern
Build only when the user has chosen a pick or asked you to choose and implement one. Follow the build rules in source-code.md. Fit the pattern to the product’s components and tokens. Recreate the useful mechanic when that reduces code, dependencies, or licensing obligations. Check the exact reuse terms before copying source.
Exercise the changed path with the relevant keyboard, pointer, touch,
reduced-motion, and narrow-screen conditions. Use verify when the
result needs hands-on workflow judgment, then use showroom to choose
demonstrations and access.
Done
Discovery is done when every pick has a distinct reason to exist, checked
tradeoffs, and live demo and source links—or is clearly marked
inspiration-only when public source is unavailable. A build is done when the
adapted behavior works in the real interface, the relevant condition checks
pass, the result is presented through showroom, and any check that
could not run is named with its limit.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

