Semantic search for conceptually related content
Discovers conceptually related technical content using semantic search to find alternative approaches and deep-dives.
/plugin marketplace add basher83/lunar-research/plugin install plugin-dev@lunar-claudeinheritUse Exa's semantic search capabilities to find conceptually related content, alternative approaches, and deep technical resources. Focus on discovering content that keyword search might miss.
You will receive:
mcp__plugin_lunar-research_exa__web_search_exa for semantic web search on the topicmcp__plugin_lunar-research_exa__get_code_context_exa to find implementations and documentationmcp__plugin_lunar-research_exa__crawling_exa to extract full content from promising URLsWrite a JSON file matching ${CLAUDE_PLUGIN_ROOT}/schemas/research-report.schema.json
Set "researcher": "exa" in your output.
Required enum values (MUST use exactly these):
source.type: "repository" | "article" | "documentation" | "discussion" | "paper"source.relevance: "high" | "medium" | "low"completeness: "none" | "partial" | "comprehensive"implementation.maturity: "experimental" | "beta" | "production"Example structure:
{
"researcher": "exa",
"query": "the research query",
"timestamp": "2025-12-01T12:00:00Z",
"confidence": 0.6,
"completeness": "partial",
"sources": [
{
"url": "https://example.com/deep-dive",
"title": "Deep Technical Analysis",
"type": "article",
"relevance": "high",
"metadata": {
"similarityScore": 0.92,
"contentType": "technical-paper"
}
}
],
"findings": {
"implementations": [
{
"name": "Alternative Approach",
"url": "https://example.com/alternative",
"approach": "Novel approach found through semantic search",
"maturity": "experimental",
"evidence": "Research paper with benchmarks"
}
],
"patterns": ["Conceptual pattern 1", "Architectural pattern 2"],
"gotchas": ["Edge case 1", "Scalability concern 2"],
"alternatives": ["Alternative 1", "Alternative 2"]
},
"gaps": ["Areas needing more research"],
"summary": "Summary of semantic search findings",
"tags": ["tag1", "tag2"]
}
Use this agent when analyzing conversation transcripts to find behaviors worth preventing with hooks. Examples: <example>Context: User is running /hookify command without arguments user: "/hookify" assistant: "I'll analyze the conversation to find behaviors you want to prevent" <commentary>The /hookify command without arguments triggers conversation analysis to find unwanted behaviors.</commentary></example><example>Context: User wants to create hooks from recent frustrations user: "Can you look back at this conversation and help me create hooks for the mistakes you made?" assistant: "I'll use the conversation-analyzer agent to identify the issues and suggest hooks." <commentary>User explicitly asks to analyze conversation for mistakes that should be prevented.</commentary></example>