Research
Research topics using web, docs, and code search. Scale depth to complexity.
Triggers
- •"research X", "look into X", "investigate X"
- •"what's the best way to", "how should I"
- •"what library for", "which approach"
- •Before implementing unfamiliar feature
- •Evaluating libraries/tools
Depth Scaling
| Complexity | Example | Sources | Output |
|---|---|---|---|
| Quick | "best JSON library for rust" | 1-2 sources | 2-3 options, pick one |
| Medium | "auth patterns for API" | Multiple sources | Options table, recommendation |
| Deep | "distributed system architecture" | All sources, parallel | Full analysis, tradeoffs, diagrams |
Assess first: How consequential is this decision? Scale depth accordingly.
Sources (Use Best Tool for Task)
| Query Type | Best Tool | Notes |
|---|---|---|
| Library/framework docs | Context7 | Resolve ID first, LLM-optimized |
| Code examples, API patterns | Exa | get_code_context_exa |
| General web, news, articles | Web search | Built-in or Exa |
| Local patterns | Read/Grep | Check codebase first |
Process
- •Parallel search: Hit all sources simultaneously
- •Filter noise: Prioritize official docs, recent content, high-quality sources
- •Synthesize: Combine findings into actionable summary
- •Cite sources: Include links for reference
Output Format
markdown
# Research: [Topic] ## Summary [2-3 sentence TL;DR] ## Key Findings ### Best Practices - [Finding 1] - [Finding 2] ### Options Considered | Option | Pros | Cons | | ------ | ---- | ---- | | A | ... | ... | | B | ... | ... | ### Recommended Approach [What to do and why] ## Code Examples [Relevant snippets if applicable] ## Sources - [Title](url)