prospect-research-agent

Type a prospect name → get a 10-section sales-grade psychological profile in ~60 seconds.

🌐 Live demo: https://darshakbhatt-aiproductlabs.github.io/prospect-research-agent/

A two-agent chain built in n8n + LangChain. Agent 1 does background research (Perplexity Sonar Pro + LinkedIn scrape + Google search). Agent 2 turns research into a structured psychological sales profile and emails it.

What's in the box

File	What it does
workflows/prospect-research.json	The full n8n workflow. Chat trigger, two LangChain agents, three tools, Gmail delivery.
docs/	The GitHub Pages landing site.

Architecture

  💬 Chat / 📅 Calendly trigger
              │
              ▼
  Greeter Agent (gpt-4o-mini) ──── asks for prospect name
              │
              ▼
  Extract Name LLM (chatgpt-4o-latest)
              │
              ▼
  Valid name? ── no ──▶ reprompt
              │ yes
              ▼
  ┌─────────── Research Agent (GPT-4o) ───────────┐
  │  tools: Perplexity Sonar Pro                  │
  │         Google Search                         │
  │         LinkedIn scrape                       │
  └────────────────┬───────────────────────────────┘
                   ▼
  Psychological Profiler Agent (chatgpt-4o-latest)
                   │
                   ▼
  📧 Gmail — 10-section profile report

What's in the dossier

The Profiler Agent produces a structured report with these sections:

1. Executive Summary & Core Psychological Profile — archetype, automation readiness, internal narrative
2. Buying Triggers — ranked 1-3 with activation language
3. Operational Pain Points — likely automatable processes
4. Fear Factors — what makes them say no
5. Decision-Making Psychology — proof requirements, success metrics, budget authority
6. Influence Pathways — who they listen to
7. Communication Blueprint — power words, opening lines, value-prop framings
8. Objection Handling — predicted objections + scripted counters
9. Sales Strategy & Execution — audit-to-implementation sequence
10. Action Plan — immediate next steps and 30-day timeline

Setup

1. Import workflows/prospect-research.json into your n8n.
2. Wire credentials:
   • OpenAI API (used by gpt-4o-mini, gpt-4o, and chatgpt-4o-latest)
   • Perplexity API (Sonar Pro model)
   • Gmail OAuth2
   • Calendly OAuth (optional, alternate trigger)
3. Replace your-email@example.com in the Gmail node with your recipient.
4. Activate and point your chat surface (web embed, Slack, Telegram, etc.) at the chat trigger webhook.
5. Send "Hi" — the agent will ask for a name.

Cost

Approximately $0.08–0.15 per profile at gpt-4o pricing (one research pass + one profiler pass + a few small-model calls). Add Perplexity API cost (~$0.005 per query).

⚠️ Ethics note

This tool researches publicly available information about professional contacts. Use it well.

• ✅ For: preparing for sales conversations and partnership discussions with people who've already publicly signaled openness (booked a meeting, opened a thread).
• ❌ Not for: mass cold outreach, scraping non-consenting individuals, building shadow profiles, or any use that violates LinkedIn's terms of service or GDPR.
• The LinkedIn integration is rate-limited and intended for one-prospect-at-a-time use.
• You are responsible for your jurisdiction's rules.

If you use it, anchor on three rules: 1) only profile people who've opted in to commercial contact, 2) destroy profiles after the engagement, 3) never quote the AI-generated profile in conversation — use it as preparation, not script.

License

MIT — fork, modify, ship.

---

Built by Darshak Bhatt.