Brand Scout Expert

Self-improving lead research expert for autonomous eCommerce brand intelligence gathering.

name: brand-scout-expert description: | Autonomous lead research system with Chrome DevTools MCP, web search, and HubSpot integration. Use when: researching prospects, generating brand profiles, creating email sequences, HubSpot lead creation Triggers on: "brand scout", "scout [brand]", "research [company]", "lead intelligence", "generate brand profile", "prospect research", "email sequence", "shipping intelligence"

Quick Reference

Execution Modes

Decision Rules Summary

BEFORE Research

• IF company already in HubSpot → UPDATE existing record, don't create duplicate
• IF target is B2B/wholesale only → SKIP (not DTC fit)
• IF no website found → SKIP (insufficient data)

DURING Research

• IF shipping page not found → CHECK checkout flow AND FAQ AND search "[BRAND] shipping policy"
• IF no contacts on website → TRY LinkedIn "See all employees" AND press releases
• IF Chrome DevTools timeout → FALLBACK to WebSearch AND Wayback Machine
• IF revenue unknown → ESTIMATE from (funding ÷ 10) OR (employees × $200K)

AFTER Research

• IF confidence < 60% → DO NOT SUBMIT (LOW quality)
• IF confidence 60-79% → FLAG for manual review (MEDIUM)
• IF confidence 80%+ → READY for HubSpot creation (HIGH)
• IF contacts < 2 verified → CONTINUE research until 2+ found

HubSpot Integration

• ALWAYS search by email AND domain BEFORE creating records
• ALWAYS use exact industry values (CONSUMER_GOODS not "consumer goods")
• ALWAYS set hubspot_owner_id = "699257003"
• ALWAYS create Company THEN Contact(s) THEN Associate

Shipping Validation Quality Gate (v3.8)

MANDATORY: Every Brand Scout report MUST validate 5 shipping criteria before completion.

5-Point Checklist

Research Sequence (try in order)

1. /shipping or /pages/shipping
2. /policies/shipping-policy (Shopify standard)
3. /shipping-returns or /delivery
4. /faq (search for shipping section)
5. /help or help center subdomain
6. Return Policy Page - Often reveals 3PL/warehouse address
7. FALLBACK: Coupon sites (Knoji, SimplyCodes) for free threshold

Validation Rules

• Report is INCOMPLETE until all 5 criteria have explicit values
• "Not found" is a valid value (means data is unavailable, not missing research)
• Every report MUST include ## SHIPPING VALIDATION STATUS table
• Score of 5/5 required for report completion

Report Structure (9 Sections)

1. Snapshot - Revenue, AOV, annual volume, growth rate
2. Shipping Intelligence - Carriers, service levels, SLA, complaints, 3PL
3. Company Overview - Legal name, HQ, founding, DTC/wholesale split
4. Stakeholders & Contacts - Decision makers with verified emails/LinkedIn
5. Observations & Competitors - Strengths, complaints %, benchmarks, risks
6. HubSpot Lead Record - Copy/paste fields ready for import
7. CRM Contact Summary - One-line format for quick reference
8. Technical Integration - Platform, APIs, monitoring capabilities
9. Methodology & Versioning - Sources, tools, confidence score

Confidence Levels

Reference Files

Integration Points

With Discovery/Sales Expert

• Brand Scout reports feed 38 Questions discovery prep
• Shipping intelligence informs service level recommendations
• Contact research enables personalized outreach

With HubSpot CRM Expert

• Uses association IDs for lead creation (578, 580)
• Follows timestamp and rate limiting patterns
• Integrates with deal folder naming conventions

With Goal Tracking Expert

• new_leads metric counts Contacts created (not Companies)
• Stage 1 micro-actions reference Brand Scout workflow
• Weekly goals include 15-20 reports target

Common Commands

# Quick research (15 min)
/brand-scout:quick [BRAND] --focus contacts

# Standard research (25-35 min)
/brand-scout:scout [BRAND]

# Full automation with HubSpot
/brand-scout:full [BRAND] --hubspot --folder --email-sequence

# Check if lead exists
/check-lead [BRAND]

# Add to HubSpot after research
/add-lead [BRAND]

# Generate email sequence only
/brand-scout:email-sequence [BRAND]

Learning Mechanism

After each Brand Scout execution:

1. Log outcome (success/failure/partial)
2. Record any new patterns discovered
3. Update expertise.yaml if learning occurred
4. Note Chrome DevTools workarounds that worked

Version History

• v3.8 (Current): 5-Point Shipping Validation Quality Gate - mandatory checklist, research sequence, confidence markers
• v3.7: Full Chrome DevTools integration, 9-section template
• v3.5: Added email sequence generation
• v3.0: HubSpot Lead object support