a-teams/docs/agent-index.md

a-teams by Worksona - worksona agents and agentic teams in claude.ai. Enterprise-grade multi-agent workflow system with 60+ specialized agents, comprehensive template system, and advanced orchestration capabilities for business, technical, and research ta

# Elite Research Agents Index

## Overview

The Elite Research Team consists of 12 specialized AI agents designed for comprehensive research analysis, multivariant search, POV analysis, synthesis, and long-form research generation. This system enables assignment handoff and produces research reports from indexes to summaries to synthesis.

## Research Workflow Modes

### Standard Research Mode (2-4 hours)
- Multi-source analysis with basic expert consultation
- Executive summary with key findings and recommendations
- Cross-source validation and fact-checking
- Standard deliverable package

### Deep Research Mode (8-24 hours)
- Comprehensive academic and expert network mining
- Advanced predictive modeling and scenario analysis
- Multi-tiered executive communication suite
- Comprehensive deliverable package with appendices

## Elite Research Agents

### Core Coordination
- **research-coordinator**: Strategic research orchestration, workflow management, and quality assurance
- **source-indexer**: Advanced cataloging, citation network analysis, and research intelligence

### Discovery & Mining
- **search-specialist**: Multivariant search across academic, industry, news, and government sources
- **deep-miner**: Premium database access, expert network consultation, and archival research

### Analysis & Validation
- **content-analyzer**: Multi-layered content analysis with thematic extraction and bias detection
- **pov-analyst**: Multi-perspective analysis, stakeholder mapping, and bias neutralization
- **fact-checker**: Multi-source triangulation, accuracy validation, and credibility assessment
- **trend-analyst**: Historical pattern analysis, predictive modeling, and scenario forecasting

### Synthesis & Intelligence
- **research-synthesizer**: Cross-source synthesis with predictive insights and strategic recommendations
- **expert-synthesizer**: Expert opinion compilation with consensus analysis and authority weighting

### Communication & Deliverables
- **report-generator**: Comprehensive research report creation with professional formatting
- **executive-summarizer**: Multi-tiered executive summaries optimized for C-suite consumption

## Usage Patterns

### Direct Agent Invocation
```bash
claude-agents run research-coordinator --task "Execute deep research on AI market trends"
claude-agents run search-specialist --task "Multivariant search on quantum computing"
claude-agents run pov-analyst --task "Stakeholder analysis for renewable energy policy"
```

### Research Workflow Execution
```bash
# Standard research assignment (2-4 hours)
npm run research-standard

# Deep research assignment (8-24 hours) 
npm run research-deep

# Custom research workflow
node -e "import('./src/utils/research-workflow.js').then(m => m.executeResearchExample())"
```

### Agent Coordination Examples
```javascript
// Multi-agent research coordination
const agents = [
  'search-specialist', 
  'deep-miner', 
  'content-analyzer', 
  'pov-analyst'
];

// Execute agents in parallel for maximum performance
agents.forEach(agent => {
  claude-agents run ${agent} --task "${researchTask}"
});
```

## Memory Integration

All agents share knowledge through the memory system:

```javascript
// Research coordination memory
memory.set("research:assignment", {
  topic: "AI Market Analysis",
  mode: "deep",
  timeline: "16 hours",
  deliverables: ["comprehensive-report", "executive-summary", "presentations"]
});

// Cross-agent knowledge sharing
memory.get("search:sources"); // Source discovery results
memory.get("analysis:insights"); // Content analysis insights  
memory.get("synthesis:recommendations"); // Strategic recommendations
```

## Context-Forge Integration

All agents are context-forge aware and integrate with existing PRPs:

```bash
# Agents automatically detect and integrate with context-forge projects
claude-agents init --respect-context-forge

# Commands are placed in .claude/commands/agents/ to avoid conflicts
ls .claude/commands/agents/
```

## Agent Capabilities Matrix

| Agent | Search | Analysis | Synthesis | Validation | Communication |
|-------|--------|----------|-----------|------------|---------------|
| research-coordinator | ★ | ★★★ | ★★★ | ★★★ | ★★★ |
| search-specialist | ★★★ | ★ | ★ | ★★ | ★ |
| deep-miner | ★★★ | ★★ | ★ | ★★★ | ★ |
| content-analyzer | ★ | ★★★ | ★★ | ★★ | ★ |
| pov-analyst | ★ | ★★★ | ★★★ | ★★ | ★★ |
| fact-checker | ★ | ★★ | ★ | ★★★ | ★ |
| trend-analyst | ★ | ★★★ | ★★ | ★★ | ★★ |
| research-synthesizer | ★ | ★★ | ★★★ | ★★ | ★★ |
| expert-synthesizer | ★ | ★★ | ★★★ | ★★★ | ★★ |
| report-generator | ★ | ★ | ★★ | ★★ | ★★★ |
| executive-summarizer | ★ | ★ | ★★ | ★ | ★★★ |
| source-indexer | ★★ | ★★ | ★ | ★★★ | ★ |

## Quality Standards

All agents maintain enterprise-grade quality standards:

- **Source Coverage**: 95%+ coverage of relevant sources within domain
- **Accuracy Validation**: Multi-source triangulation with >90% accuracy
- **Expert Authority**: Consultation with recognized domain authorities
- **Bias Mitigation**: Systematic bias detection and neutralization
- **Executive Focus**: Strategic insights optimized for decision-making

## Performance Optimization

### Concurrent Execution
Always execute related agents concurrently for maximum performance:

```javascript
// Parallel agent execution for 3x performance improvement
const parallelExecution = [
  'search-specialist',
  'deep-miner', 
  'content-analyzer'
].map(agent => executeAgent(agent, task));

await Promise.all(parallelExecution);
```

### Memory Coordination
Agents share insights through the memory system to avoid redundant work:

```javascript
// Check for existing analysis before starting
const existingAnalysis = memory.get("analysis:completed");
if (!existingAnalysis) {
  // Proceed with analysis
}
```

## Integration Examples

### Academic Research Pipeline
```bash
# 1. Deep source mining
claude-agents run deep-miner --task "Academic database mining for machine learning"

# 2. Content analysis
claude-agents run content-analyzer --task "Thematic analysis of ML research"

# 3. Trend analysis  
claude-agents run trend-analyst --task "ML research trend forecasting"

# 4. Synthesis
claude-agents run research-synthesizer --task "ML research synthesis"
```

### Market Intelligence Workflow
```bash
# 1. Multivariant search
claude-agents run search-specialist --task "Market intelligence on fintech trends"

# 2. POV analysis
claude-agents run pov-analyst --task "Stakeholder analysis for fintech regulation"

# 3. Expert synthesis
claude-agents run expert-synthesizer --task "Expert opinion on fintech future"

# 4. Executive communication
claude-agents run executive-summarizer --task "C-suite fintech briefing"
```

### Comprehensive Research Assignment
```javascript
// Full research workflow with assignment handoff
import { EliteResearchWorkflow } from './src/utils/research-workflow.js';

const researchWorkflow = new EliteResearchWorkflow();

const assignment = {
  title: "Future of Autonomous Vehicles",
  mode: "deep", // or "standard"
  objective: "Strategic analysis of AV market and regulatory landscape",
  timeline: "16 hours",
  deliverables: ["comprehensive-report", "executive-summary", "presentation"]
};

const results = await researchWorkflow.executeResearchAssignment(assignment);
```

## Command Reference

### Agent Installation
```bash
# Install all elite research agents
claude-agents install elite-research

# Install specific agent
claude-agents install research-coordinator
claude-agents install deep-miner
```

### Agent Execution
```bash
# Standard agent execution
claude-agents run <agent-name> --task "<task-description>"

# With specific parameters
claude-agents run research-coordinator --task "Deep research on blockchain" --mode deep --timeline "12 hours"
```

### Workflow Management
```bash
# List all agents
claude-agents list --category elite-research

# Check agent status
claude-agents status research-coordinator

# View agent capabilities
claude-agents info deep-miner
```

## Dashboard Access

Access the research dashboard for real-time monitoring:

```bash
# Launch web dashboard
claude-agents dashboard

# Access at http://localhost:7842
# - Agent overview and status
# - Memory inspection and coordination
# - Research progress tracking
# - Performance metrics
```

## Support and Documentation

- **Agent Documentation**: Each agent includes comprehensive methodology documentation
- **Research Quality**: Multi-source validation with confidence metrics
- **Performance Tracking**: Built-in metrics for research quality and execution time
- **Expert Networks**: Integration with professional expert consultation networks

The Elite Research Team transforms complex research assignments into executive-ready strategic intelligence through systematic multi-agent coordination, comprehensive source analysis, and professional deliverable generation.