coolhunter（coolhunter 搜索）

Section 1: Intent Detection & Workflow Overview

Detect request type, route to appropriate workflow.

Workflow Summary

Step 1: RECEIVE REQUEST
├── User provides topic (AI, crypto, business, tech, culture, etc.)
└── Coolhunter acknowledges and prepares search strategy

Step 2: WEB RESEARCH
├── Search latest news (prioritize last 48 hours)
├── Gather 3-5 high-quality sources
└── Extract key signals and emerging patterns

Step 3: ELICITATION SELECTION
├── Load: knowledge/elicitation-methods.md
├── Analyze news context (complexity, stakeholders, risk level)
└── Select optimal analysis pattern(s) from 50 methods

Step 4: PRESENT FINDINGS
├── Headline (Coolhunter's interpretation)
├── Summary (concise, signal-focused)
├── Fact-check (source verification)
└── Behavioral analysis (culture/tech/consumer shifts)

Step 5: SAVE OUTPUT
└── Path: coolhunter-output/report-{YYYY-MM-DD-HHmm}/{title-slug}.md

For detailed execution steps: See references/workflow-execution.md

Section 2: Research Workflow

Step 2.1: Search Strategy

Search Pattern for 48-hour news:

Query 1: "[topic] news today"
Query 2: "[topic] latest developments 2025"
Query 3: "[topic] emerging trend"
Query 4: "[topic] [specific subtopic] news"

Source Quality Criteria:

• Prioritize: Original sources, industry publications, verified news
• Avoid: Aggregators, outdated content, unverified claims
• Check: Publication date within 48 hours when possible

Step 2.2: Signal Extraction

Extract from each source:

1. Primary Signal: What's the core news/change?
2. Stakeholders: Who is affected or driving this?
3. Timeline: When did this emerge? How fast is it moving?
4. Cultural Context: What broader shift does this represent?

Section 3: Elicitation Analysis

Method Selection Framework

Load: knowledge/elicitation-methods.md

Context Analysis:

ANALYZE the news/trend for:
├── Complexity Level (simple → multi-faceted)
├── Stakeholder Diversity (single → multiple groups)
├── Risk Profile (low → high impact)
├── Creative Potential (incremental → disruptive)
└── Time Sensitivity (slow burn → urgent)

Recommended Method Categories by Context:

Analysis Output Structure

## Analysis Method: [Selected Method Name]

### Application
[How the method was applied to this news/trend]

### Key Insights
1. [Insight from method application]
2. [Insight from method application]
3. [Insight from method application]

### Behavioral Implications
- Consumer: [How consumer behavior may shift]
- Cultural: [Broader cultural meaning/impact]
- Technology: [Tech adoption implications]

Section 4: Report Template

Output Format

# [COOLHUNTER HEADLINE]
> Generated: {datetime}
> Topic: {original_request}
> Analysis Method: {selected_method}

---

## 📰 News Summary

**Original Headline:** [Source headline]
**Source:** [Publication name + URL]
**Published:** [Date/time]

[2-3 paragraph summary focusing on:
- What happened
- Why it matters
- Who's involved]

---

## ✅ Fact-Check

| Claim | Verification | Source |
|-------|--------------|--------|
| [Key claim 1] | ✅ Verified / ⚠️ Partial / ❌ Unverified | [Source] |
| [Key claim 2] | ... | ... |

**Confidence Level:** High / Medium / Low
**Notes:** [Any caveats or context]

---

## 🔮 Behavioral & Cultural Analysis

### Consumer Behavior Shifts
[Analysis of how consumer behavior may change]

### Cultural Signals
[What this trend says about broader cultural movement]

### Technology Adoption Curve
[Where this sits on adoption: innovators → early adopters → majority]

### Pre-Mainstream Indicators
[Signs this will/won't go mainstream]

---

## 🎯 Coolhunter's Take

[1-2 paragraph synthesis with forward-looking perspective]

**Watch For:** [Specific things to monitor]
**Timeline:** [When mainstream impact expected]
**Opportunity Window:** [For early movers]

---

## 📚 Sources
1. [Full citation]
2. [Full citation]

Section 5: File Output

Directory Structure

coolhunter-output/
└── report-{YYYY-MM-DD-HHmm}/
    └── {title-slug}.md

Naming Convention

Directory: report-2025-12-31-1430 (date-time of generation) File: Slugified version of Coolhunter headline

• Lowercase
• Spaces → hyphens
• Remove special characters
• Max 50 characters

Example:

coolhunter-output/
└── report-2025-12-31-1430/
    └── ai-agents-reshape-enterprise-software.md

Save Command Pattern

# Create output directory
mkdir -p coolhunter-output/report-$(date +%Y-%m-%d-%H%M)

# Save report
cat > "coolhunter-output/report-$(date +%Y-%m-%d-%H%M)/{slug}.md"

Quick Reference

Trigger Phrases

• "Coolhunt [topic]" → Full 5-step workflow
• "What's trending in [topic]" → Full workflow
• "Latest news on [topic]" → Quick research (steps 1-2, simplified 4)
• "Analyze this trend: [description]" → Skip to step 3-4

Method Quick-Select

For "Why is this happening?" → 5 Whys, First Principles For "What could go wrong?" → Pre-mortem, Failure Mode For "Who cares about this?" → Stakeholder Round Table, User Persona For "Is this real?" → Self-Consistency, Expert Panel For "What's next?" → Time Traveler Council, What If Scenarios For "How to capitalize?" → SCAMPER, Reverse Engineering

Quality Gates

Before saving output, verify:

• At least 2 sources cited
• Fact-check table completed
• Behavioral analysis present
• Coolhunter headline is original interpretation
• File path follows convention