Spenzo
Omni-chain financial MCP server that tracks expenses, manages Web3 portfolios, simulates DEX swaps, and generates payment links via natural language in WhatsApp and Claude Desktop.
README
Spenzo : The Omni-Chain Financial Superpower

Spenzo is an advanced, Omni-Chain Financial Superpower. It doesn't just track your fiat expenses: it operates as a consolidated NLP layer to manage your wealth. Monitor live Web3 portfolios, simulate advanced DEX routing quotes, and extract invoice metadata entirely through programmatic chat integrations natively inside WhatsApp and Claude Desktop.
Website: spenzo.xyz
Features
- WhatsApp Native: Text your expenses naturally ("Spent ₹150 on chai"). Spenzo auto-categorizes and logs them instantly.
- Receipt Vision: Snap a photo of a bill or receipt on WhatsApp. Spenzo extracts the merchant, total, and date automatically using GPT-4o.
- Smart Bill Splitting: "Dinner was 3000 but we split it 3 ways." Spenzo handles the math and logs your exact share.
- Live Crypto Prices: Built-in real-time CoinGecko integration. Ask "What's the price of SOL?" to log crypto purchases accurately.
- Claude Desktop Integration: Use Spenzo as a native MCP (Model Context Protocol) tool inside Claude for hands-free financial management.
- Instant Summaries: Ask "How much did I spend on food this week?" and Spenzo computes category breakdowns on the fly — right inside Claude or WhatsApp.
- Cross-Platform Sync: Securely link your WhatsApp number to your desktop session via OTP so all your financial data is unified.
Architecture
Spenzo uses a modern, serverless-friendly microservice architecture:
flowchart TD
%% ── Client Layer ──
WA["📱 WhatsApp User"] <-->|"Text · Receipts · Images"| TW["Twilio WhatsApp API"]
CD["🖥️ Claude Desktop"] <-->|"FastMCP Protocol<br/>(stdio / SSE)"| MCP
%% ── WhatsApp Pipeline ──
TW -->|"POST /webhook"| BOT["bot.py<br/>FastAPI Webhook Server"]
BOT -->|"Reply Message"| TW
BOT <-->|"Chat Completions<br/>+ Tool Use"| LLM(("GPT-4o-mini<br/>+ Vision"))
BOT <-->|"stdio subprocess<br/>(MCP Client → Server)"| MCP["main.py<br/>FastMCP Server · 16 Tools"]
%% ── Database ──
MCP <-->|"CRUD · IOUs · OTP Auth"| DB[("Supabase<br/>PostgreSQL")]
%% ── External APIs ──
MCP -->|"Live Prices"| CB["Coinbase · CoinGecko"]
MCP -->|"Wallet · Gas · Staking"| W3["Alchemy · Helius · Blockscout"]
MCP -->|"Swap Routing"| JUP["Jupiter v6 DEX"]
MCP -->|"Payment Links"| UPI["UPI Intent + QR Generator"]
MCP -.->|"OTP Delivery"| TW
%% ── Styles ──
classDef client fill:#a7dd5d,stroke:#000,stroke-width:2px,color:#000
classDef infra fill:#1c1c1c,stroke:#fff,stroke-width:1px,color:#fff
classDef ai fill:#6c47ff,stroke:#fff,stroke-width:2px,color:#fff
classDef db fill:#3ECF8E,stroke:#000,stroke-width:2px,color:#000
classDef api fill:#2d2d2d,stroke:#555,stroke-width:1px,color:#ccc
class WA,CD client
class TW,BOT,MCP infra
class LLM ai
class DB db
class CB,W3,JUP,UPI api
QuickStart Guide
1. Try the WhatsApp Bot
- Message
+1 (415) 523-8886on WhatsApp. - Send the exact phrase
join conversation-headingto enter the secure Twilio sandbox. - Say "Hi" to see what Spenzo can do!
- Ask "How much did I spend this week?" for an instant summary.
Note: The backend sleeps on Render's free tier. Your very first message might take ~45 seconds while the server wakes up. Consecutive messages are instant.
2. Connect to Claude Desktop (3 Options)
Since Spenzo is an MCP Server, it must attach to a client like Claude Desktop. Choose the right method for you:
1. Download the App for Mac/Windows (No Coding Required)
You can run Spenzo as a single, zero-dependency executable. No source code or environment setup required.
- Download the compiled binary for your platform from the latest GitHub Release:
- Edit your
claude_desktop_config.json:"mcpServers": { "spenzo": { "command": "/Users/YOUR_NAME/Desktop/spenzo-mac", "env": { "SUPABASE_URL": "...", "SUPABASE_KEY": "..." } } }
2. Cloud API (Zero Local Compute)
Spenzo's main.py detects cloud environments automatically and starts a Server-Sent Events (SSE) listener.
- Run Spenzo on Render/AWS.
- Download the tiny
sse_relay.pyconnection script to your Desktop. - Edit your
claude_desktop_config.json:"mcpServers": { "spenzo-cloud": { "command": "python", "args": ["/Users/YOUR_NAME/Desktop/sse_relay.py", "https://YOUR-RENDER-URL.onrender.com/sse"] } }
3. From Source (For Developers)
Prerequisites: Python 3.11+ and the uv package manager.
- Clone & Install:
git clone https://github.com/Arav-Arun/Spenzo.git cd Spenzo uv sync - Environment Variables:
Create a
.envfile in the root directory:SUPABASE_URL="https://your-project.supabase.co" SUPABASE_KEY="your-service-role-key" OPENAI_API_KEY="sk-..." TWILIO_ACCOUNT_SID="ACxxx..." TWILIO_AUTH_TOKEN="your-auth-token" TWILIO_WHATSAPP_NUMBER="whatsapp:+14155238886" ALCHEMY_API_KEY="your-alchemy-key" HELIUS_API_KEY="your-helius-key" - Link to Claude Desktop:
Open your Claude Desktop config file and add:
Restart Claude, and you can now ask Claude to manage your expenses!{ "mcpServers": { "spenzo": { "command": "uv", "args": ["run", "fastmcp", "run", "/ABSOLUTE/PATH/TO/Spenzo/server/main.py"] } } }
Available MCP Tools
Developers can utilize the following database tools via server/main.py:
| Tool | Description |
|---|---|
add_expense |
Log a new expense (date, amount, category, subcategory, note). |
list_expenses |
Fetch chronological expense entries within a date range. |
summarize |
Generate category-based spending totals and breakdowns. |
search_expenses |
Keyword search across all categories, subcategories, or notes. |
edit_expense / delete_expense |
Modify or remove specific transaction rows securely. |
get_crypto_price |
Fetch live USD/INR prices for seamless crypto logging. |
register_phone / verify_phone |
Issue and verify Twilio OTPs to securely bridge platforms. |
The Web3 Edge (Advanced MCP Tools)
Spenzo seamlessly acts as a Bloomberg Terminal for personal crypto wallets, pulling live data from Alchemy (Ethereum) and Helius (Solana) RPCs.
Whenever you provide a public Web3 wallet (0x... or base58), Spenzo exposes 4 powerful analytical tools:
| Web3 Tool | Capabilities |
|---|---|
analyze_web3_wallet |
Scans an ETH/SOL wallet to pull real-time token balances and fiat Net Worth. |
simulate_dex_swap |
Hits DEX aggregators (Jupiter/Uniswap) to quote trades and slippage. |
analyze_gas_burn |
Sweeps strictly on-chain transactions via Helius/Etherscan to compute real gas fees. |
check_staking_yield |
Scans for LST balances (stETH, JitoSOL) and native validator staking rewards. |
Spenzo securely transitions from a read-only intelligence layer to an active Execution tool. By dynamically constructing deep-link payloads, Spenzo handles complex routing and computation in the background, then hands you a 1-click confirmation prompt without ever requesting access to your private keys or banking credentials.
| Execution Tool | Action & Routing Mechanism |
|---|---|
simulate_dex_swap |
Hits Jupiter v6 Quote API to format exact 1-Click Transaction Execution links. |
generate_upi_payment_link |
Constructs upi://pay intent payloads and dynamic QR codes for 1-click fiat drops. |
log_debt / list_debts |
Serves as an isolated IOU ledger, natively returning auto-generated UPI collection links. |
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.