agentleFS
lintendolintendo/Axhub-Skills/skills/search-design-system/SKILL.md

search-design-system

Skill145 starsChanged 20 days agoMITView on GitHub

Use when selecting an existing design system or theme for a product, page, prototype, or implementation through Axhub design knowledge indexes; not when creating or updating a theme.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are 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.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read response-schema.md for result fields, artifact behavior, and stable error codes.

    line 61

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Search Design System

Overview

Turn a design need into a private, structured search; inspect the returned evidence before selecting a theme. Search and scoring run locally after the versioned index is loaded.

Workflow

  1. 先把需求整理为结构化查询。只提取平台、行业、产品类型、页面类型、风格、品牌气质、色系、明暗模式、密度和关键词。不得把用户原文传给脚本或网络端点。
  2. 确定平台。用户没有指定时,向用户确认平台,或同时搜索 desktop 和 mobile 两份索引;不要擅自默认。
  3. 按 query-schema.md 和 taxonomy.md 生成请求。默认使用 scripts/cli.mjs search 读取 Make-Template 的线上 manifest;只有在源码工作区或用户明确指定时才传 --index 和 --local-root。
  4. 对每个候选读取 matched、unmatched、scoreBreakdown、reviewStatus 和 publishable,再取得完整的 DESIGN.md 并查看 preview。不能只按 score 选择。
  5. 结合用户约束说明推荐理由和不匹配项。deferred 可用于本地参考,但不得描述为已获公开发布授权。
  6. 只有用户工作流需要落地资源时才下载 package;普通选型只读取 DESIGN.md 和 preview。不得发送 use 事件或 analytics 事件。

Example

将本地整理出的请求写入临时文件:

{
  "schemaVersion": 1,
  "readerVersion": "1.0.0",
  "platform": "desktop",
  "terms": ["analytics", "finance"],
  "hardFilters": { "industries": ["finance-payments"] },
  "softFilters": { "styles": ["professional"], "density": ["high"] },
  "exclude": { "styles": ["playful"] },
  "limit": 4
}
node scripts/cli.mjs search --index /path/to/desktop.json --request /path/to/request.json --local-root /path/to/runtime

Then pass one returned result to scripts/cli.mjs fetch --kind designMd and read the full body before choosing.

Quick reference

SituationAction
Platform missingConfirm, or search both indexes
Deferred resultRead verified DESIGN.md and preview locally或在线;package stays unavailable
Online resultDefault to the canonical Make-Template manifest and verified hashes
Too few resultsReport resultSummary.reason; do not pad or duplicate
Stale cacheState it explicitly; do not silently downgrade

Common mistakes

  • Passing the user's original prose as terms instead of extracting a minimal structured query.
  • Choosing the top score without reading unmatched constraints and full DESIGN.md.
  • Treating deferred as publishable or inventing a public artifact URL.
  • Downloading a package during exploration when DESIGN.md and preview are sufficient.

Read response-schema.md for result fields, artifact behavior, and stable error codes.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.