majiayu000 Research Skill
Conducts comprehensive autonomous research on any topic using Claude Agent SDK with web search capabilities.
Capabilities
- Autonomous Research: Multi-step web search and content extraction
- Insight Extraction: AI-powered analysis of research sources
- Citation Management: Automatic source tracking and formatting
- Multi-Presentation Detection: Identifies when topic should split into multiple presentations
Usage
Via CLI
python -m plugin.cli research "Rochester 2GC Carburetor Rebuild" --output research.json
Via Python
from plugin.skills.research.research_skill import ResearchSkill
skill = ResearchSkill(config)
result = skill.execute({
"topic": "AI in Healthcare",
"depth": "comprehensive",
"max_sources": 20
})
Input Parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
| topic | string | Yes | - | Research topic |
| depth | string | No | "standard" | Research depth: "quick", "standard", "comprehensive" |
| max_sources | int | No | 10 | Maximum sources to gather |
| audience | string | No | "general" | Target audience for content |
Output Format
{
"topic": "string",
"sources": [
{
"url": "string",
"title": "string",
"content": "string",
"relevance_score": 0.95
}
],
"insights": [
{
"key_point": "string",
"supporting_evidence": ["string"],
"citations": ["string"]
}
],
"suggested_presentations": [
{
"title": "string",
"audience": "string",
"focus": "string"
}
]
}
Dependencies
- Claude Agent SDK
- ANTHROPIC_API_KEY environment variable
Examples
Basic Research
python -m plugin.cli research "Machine Learning Basics"
Comprehensive Research with Custom Output
python -m plugin.cli research "Quantum Computing Applications" \
--depth comprehensive \
--max-sources 25 \
--output quantum-research.json
Research for Specific Audience
python -m plugin.cli research "Cloud Architecture" \
--audience "C-level executives" \
--depth quick