Skip to content

cbbkrd-tech/jl-finishes / .claude/skills/ln-001-standards-researcher

ln-001-standards-researcher

Verified

Research standards/patterns via MCP Ref. Generates Standards Research for Story Technical Notes subsection. Reusable worker.

FileKindSize
SKILL.mdskill md13.2 KB
diagram.htmlother5.4 KB
references/research_guidelines.mdmarkdown7.1 KB

diagram.html

View on GitHub
<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>ln-221-library-researcher Workflow</title>
    <link rel="stylesheet" href="../shared/css/diagram.css">
    <script src="https://cdn.jsdelivr.net/npm/mermaid@10/dist/mermaid.min.js"></script>
    <script>
        mermaid.initialize({
            startOnLoad: true,
            theme: 'base',
            themeVariables: {
                primaryColor: '#4A90E2',
                primaryTextColor: '#fff',
                primaryBorderColor: '#2E5C8A',
                lineColor: '#5C6BC0',
                secondaryColor: '#7E57C2',
                tertiaryColor: '#26A69A'
            }
        });
    </script>
</head>
<body>
    <div class="container">
        <h1>ln-221-library-researcher: Library Researcher (Worker)</h1>

        <div class="diagram-section">
            <h2>Workflow Diagram</h2>
            <div class="mermaid">
graph TD
    Start([Input: Epic Description + Story Domain]) --> Phase1[Phase 1: Identify Libraries]

    Phase1 --> Parse[Parse Epic Technical Notes<br/>+ Scope In + Goal]
    Parse --> ExtractLibs[Extract Library List<br/>3-5 libraries max]
    ExtractLibs --> Domain[Determine Story Domain<br/>e.g., rate limiting]

    Domain --> SkipCheck{Libraries<br/>found?}
    SkipCheck -->|No| EmptySummary[Output: Empty Research Summary]
    SkipCheck -->|Yes| Phase2[Phase 2: MCP Context7 Research]

    Phase2 --> LoopLibs[FOR EACH Library]
    LoopLibs --> ResolveID[resolve-library-id<br/>libraryName]
    ResolveID --> GetDocs[get-library-docs<br/>libraryID, topic, tokens=3000]
    GetDocs --> ExtractAPI[Extract:<br/>- Latest version<br/>- Key APIs 2-5<br/>- Constraints<br/>- Deprecations]

    ExtractAPI --> Fallback{Found in<br/>Context7?}
    Fallback -->|No| WebSearch[Fallback: WebSearch<br/>library latest version API]
    Fallback -->|Yes| Phase3[Phase 3: MCP Ref Research]
    WebSearch --> Phase3

    Phase3 --> RefSearch[ref_search_documentation<br/>library + domain best practices]
    RefSearch --> RefStandards[ref_search_documentation<br/>domain standards RFC]
    RefStandards --> ExtractStandards[Extract:<br/>- Industry standards<br/>- Common patterns<br/>- Security considerations<br/>- Integration approaches]

    ExtractStandards --> Phase4[Phase 4: Scan Existing Guides]
    Phase4 --> GlobGuides[Glob docs/guides/*.md]
    GlobGuides --> MatchDomain[Match guides to Story domain<br/>fuzzy match]
    MatchDomain --> CollectPaths[Collect guide paths]

    CollectPaths --> Phase5[Phase 5: Generate Research Summary]
    Phase5 --> CompileTable[Compile:<br/>- Library table version, purpose<br/>- Key APIs list<br/>- Constraints list<br/>- Standards compliance<br/>- Existing guides links]
    CompileTable --> FormatMD[Format as Markdown]
    FormatMD --> End([Output: Research Summary<br/>for Technical Notes])

    EmptySummary --> End

    style Start fill:#4A90E2,stroke:#2E5C8A,color:#fff
    style End fill:#26A69A,stroke:#1B5E20,color:#fff
    style Phase1 fill:#7E57C2,stroke:#4A148C,color:#fff
    style Phase2 fill:#7E57C2,stroke:#4A148C,color:#fff
    style Phase3 fill:#7E57C2,stroke:#4A148C,color:#fff
    style Phase4 fill:#7E57C2,stroke:#4A148C,color:#fff
    style Phase5 fill:#7E57C2,stroke:#4A148C,color:#fff
    style SkipCheck fill:#FF9800,stroke:#E65100,color:#fff
    style Fallback fill:#FF9800,stroke:#E65100,color:#fff
    style WebSearch fill:#FFC107,stroke:#F57C00,color:#000
    style EmptySummary fill:#9E9E9E,stroke:#424242,color:#fff
            </div>
        </div>

        <div class="info-section">
            <h2>Key Information</h2>
            <ul>
                <li><strong>Type:</strong> Worker (reusable)</li>
                <li><strong>Phases:</strong> 5 (sequential workflow)</li>
                <li><strong>Input:</strong> Epic description + Story domain</li>
                <li><strong>Output:</strong> Research Summary (Markdown)</li>
                <li><strong>Time-box:</strong> 15-20 minutes per Epic</li>
                <li><strong>Token usage:</strong> ~10,000 tokens (3-4 libraries)</li>
                <li><strong>Reusability:</strong> Called by ln-220, ln-310</li>
            </ul>
        </div>

        <div class="tools-section">
            <h2>Tools Used</h2>
            <ul>
                <li><code>mcp__context7__resolve-library-id()</code> - Get library ID from name</li>
                <li><code>mcp__context7__get-library-docs()</code> - Get latest API documentation (max 3000 tokens per library)</li>
                <li><code>mcp__Ref__ref_search_documentation()</code> - Search best practices and industry standards</li>
                <li><code>Glob</code> - Scan docs/guides/ directory for existing pattern guides</li>
                <li><code>WebSearch</code> - Fallback for libraries not in Context7</li>
            </ul>
        </div>

        <div class="notes-section">
            <h2>Important Notes</h2>
            <ul>
                <li>Research done ONCE per Epic, results reused for all Stories (5-10 Stories benefit)</li>
                <li>Focus on KEY APIs only (2-5 methods), not exhaustive documentation</li>
                <li>Prefer official docs and RFC standards over blog posts</li>
                <li>Skip research if NO libraries mentioned or trivial CRUD operation</li>
                <li>Parallel MCP calls when possible (Context7 + Ref)</li>
            </ul>
        </div>
    </div>
</body>
</html>
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall