Researcher Agent

Multi-source technology intelligence that explores docs, videos, GitHub repos, and articles in parallel to deliver production-ready implementation insights.

When to Use

  • New tech evaluation - Before adopting frameworks, libraries, or tools
  • Pre-implementation research - Finding official docs, security concerns, best practices
  • Technical comparison - Evaluating multiple approaches with pros/cons analysis
  • Deep dive validation - Cross-checking implementation patterns across 10+ sources

Key Capabilities

CapabilityDescriptionTools Used
Parallel SearchQuery fan-out across Google, YouTube, websites, GitHub simultaneouslySearchAPI MCP, search_youtube, WebFetch
Video AnalysisExtract technical insights from tutorials with timestampsVidCap MCP, Gemini Vision
Repo AnalysisFind implementation patterns in popular librariesrepomix, GitHub search
Doc SynthesisStructured markdown reports with security/performance sectionsMulti-source validation

Common Use Cases

Who: Product teams choosing payment gateway Prompt: /plan research Stripe vs PayPal integration for SaaS billing Output: 15-page comparison report with security audit, pricing, code examples

Who: DevOps engineer evaluating container orchestration Prompt: /plan investigate Kubernetes vs Docker Swarm for microservices Output: Architecture analysis, scaling patterns, operational complexity breakdown

Who: Frontend dev implementing real-time features Prompt: /plan research WebSocket best practices for chat app Output: Library comparison, connection handling patterns, production configs

Who: Security-focused startup Prompt: /plan research OAuth2 implementation vulnerabilities and mitigations Output: Security audit, common pitfalls, hardening checklist with sources

Who: Tech lead architecting data pipeline Prompt: /plan compare Apache Kafka vs RabbitMQ for event streaming Output: Performance benchmarks, use case fit, operational considerations

How It Works

# Trigger research via planning command
/plan [add Stripe payment integration]

What happens:

  1. Query Fan-Out - Spawns parallel searches:

    • Google: “Stripe Node.js production best practices”
    • YouTube: “Stripe integration tutorial 2025”
    • GitHub: “stripe-samples” repositories
    • Official docs: stripe.com/docs
  2. Multi-Source Analysis - Researcher validates findings across:

    • 5+ documentation sites
    • 4+ video tutorials with transcripts
    • 6+ GitHub repos (popular implementations)
    • 10+ articles/discussions
  3. Synthesis - Generates structured markdown report:

    • Executive summary with recommendation
    • Security considerations (CRITICAL section)
    • Performance implications
    • Implementation guide with code
    • 15+ cited sources for verification

Output: ./plans/research/YYMMDD-stripe-integration.md

Report Structure

# Research Report: [Technology]

## Executive Summary
- Key recommendation with confidence level
- 3-5 critical takeaways

## Integration Approaches
- Option A vs B vs C comparison
- Pros/cons table
- "Best for" recommendations

## Security Best Practices
✅ Critical requirements
⚠ Common vulnerabilities

## Performance Considerations
- Metrics: load time, processing speed
- Optimization strategies

## Implementation Guide
- Minimal working example (TypeScript)
- Step-by-step setup

## Sources (15+)
- Docs, videos, repos, articles

## Next Steps
- 5-step action plan with time estimates

Pro Tips

Be specific in research requests - “Stripe with Next.js App Router” beats “payment integration”

Leverage security sections - Researcher surfaces CVEs and common vulnerabilities automatically

Check source dates - Report includes “Last Updated” for each source; flag outdated info

Use for architecture decisions - Research 2-3 alternatives before committing to stack

Share with planner - Feed research reports directly to planner agent for implementation plans

  • Planner Agent - Converts research findings into step-by-step implementation plans
  • Brainstormer Agent - Debates technical trade-offs discovered in research
  • Scout Agent - Locates existing implementations in your codebase

Key Takeaway

Researcher eliminates “tutorial hell” by validating 15+ sources in parallel, surfacing security risks, and delivering production-ready intelligence before you write a single line of code. Part of ClaudeKit $99 toolkit.