Initial commit: antigravity-claudekit

skills/ck-docs-seeker/SKILL.md

@@ -0,0 +1,86 @@
+---
+name: ck-docs-seeker
+description: Search library and framework documentation via llms.txt standard. Use for API docs lookup, GitHub repository analysis, technical documentation discovery, latest library features, context7 integration.
+---
+
+# ck-docs-seeker
+
+Script-first documentation discovery using the llms.txt standard via context7.com. Fetch documentation for any library or framework without manual URL construction.
+
+## When to Use
+
+- Looking up API documentation for a specific library
+- Finding the latest features or syntax for a framework
+- Analyzing a GitHub repository's documentation
+- Before implementing with an unfamiliar library
+- When the local codebase docs are insufficient
+
+## Don't Use When
+
+- Searching local project files — use `ck-scout` instead
+- General web research on a broad topic — use `ck-research`
+- The documentation is already loaded in context
+
+## Primary Workflow
+
+Always execute scripts in this order from the `ck-docs-seeker` skill directory:
+
+```bash
+# 1. Detect query type (topic-specific vs general)
+node scripts/detect-topic.js "<user query>"
+
+# 2. Fetch documentation using script output
+node scripts/fetch-docs.js "<user query>"
+
+# 3. Analyze results (if multiple URLs returned)
+cat llms.txt | node scripts/analyze-llms-txt.js -
+```
+
+Scripts handle URL construction, fallback chains, and error handling automatically.
+
+## Scripts
+
+**`detect-topic.js`** — Classify query type
+- Identifies topic-specific vs general queries
+- Extracts library name + topic keyword
+- Returns JSON: `{topic, library, isTopicSpecific}`
+
+**`fetch-docs.js`** — Retrieve documentation
+- Constructs context7.com URLs automatically
+- Handles fallback: topic → general → error
+- Outputs llms.txt content or error message
+
+**`analyze-llms-txt.js`** — Process llms.txt
+- Categorizes URLs (critical/important/supplementary)
+- Recommends agent distribution strategy
+- Returns JSON with strategy
+
+## Examples
+
+**Topic-specific query:** "How do I use date picker in shadcn?"
+```bash
+node scripts/detect-topic.js "<query>"  # → {topic, library, isTopicSpecific: true}
+node scripts/fetch-docs.js "<query>"    # → 2-3 URLs
+# Fetch URLs with web fetch tool
+```
+
+**General query:** "Documentation for Next.js"
+```bash
+node scripts/detect-topic.js "<query>"              # → {isTopicSpecific: false}
+node scripts/fetch-docs.js "<query>"                # → 8+ URLs
+cat llms.txt | node scripts/analyze-llms-txt.js -  # → distribution strategy
+# Deploy parallel agents per recommendation
+```
+
+## Execution Principles
+
+1. **Scripts first** — run scripts instead of manual URL construction
+2. **Zero-token overhead** — scripts run as commands without loading content
+3. **Automatic fallback** — scripts handle topic → general → error chains
+4. **Progressive disclosure** — load only what's needed
+5. **Agent distribution** — scripts recommend parallel agent strategy for large docs
+
+## Don't Use When
+
+- The skill directory or scripts are not available in the environment
+- Simple, single-page documentation is already at a known URL — fetch directly