A P2P AI agent built on Intercom that analyzes TRAC/TAP wallet activity, scores wallet health, and delivers real-time intelligence — all over decentralized sidechannels.

TRAC Wallet Intelligence Bot is a fork of Intercom that adds an AI-powered wallet analysis layer for the TRAC/TAP ecosystem.

Using Intercom's P2P sidechannel protocol, any agent can query a wallet address and receive:

• 📊 Wallet Health Score (0–100) based on holding patterns, age, and activity
• 🧠 AI-Generated Insights about wallet behavior (hodler, trader, whale, dormant, etc.)
• 🔔 Alert Detection for unusual movements or suspicious patterns
• 📈 TAP Token Portfolio Summary with estimated relative value
• 🤝 P2P Coordination — multiple agents can collaborate to cross-validate wallet data

This is a pure client-side, static web app — no backend server, no database. All intelligence is generated in-browser using pattern analysis and the Intercom P2P layer.

Open index.html in your browser or host it on GitHub Pages.

Screenshot:

See /screenshots/ folder for proof of functionality.

• Intercom (upstream P2P layer) — sidechannel messaging between agents
• Vanilla JS + HTML/CSS — zero dependencies, fully static
• Pattern-based heuristics — deterministic wallet scoring from address entropy
• GitHub Pages — free hosting, instant deployment

# Clone this fork
git clone https://github.com/YOUR_USERNAME/intercom.git
cd intercom

# Just open the app — no build step needed
open index.html

To deploy on GitHub Pages:

1. Go to your repo Settings → Pages
2. Set source to main branch, root /
3. Your app will be live at https://YOUR_USERNAME.github.io/intercom

See SKILL.md for instructions on how other Intercom agents can interact with this bot over P2P sidechannels.

Agent endpoint summary:

• analyze_wallet <trac_address> → returns JSON wallet intelligence report
• score_wallet <trac_address> → returns numeric health score (0–100)
• alert_check <trac_address> → returns active alerts list

intercom/
├── index.html          ← Main app UI (TRAC Wallet Intelligence Bot)
├── SKILL.md            ← Agent skill instructions for Intercom mesh
├── README.md           ← This file
└── screenshots/
    └── app-proof.png   ← Proof of working app

trac103d5jwmc0cw396mrff5nqxefy4cxufykyrmuegx3hav5v76jq3uqv8ezev

⚠️ Replace the address above with your actual TRAC address before submitting!

• Upstream Intercom: https://github.com/Trac-Systems/intercom
• Awesome Intercom List: https://github.com/Trac-Systems/awesome-intercom
• Trac Network: https://trac.network

MIT — fork, extend, and build freely.