Monize includes an AI assistant that can answer natural language questions about your finances, generate insights about your spending habits, and integrate with external AI clients via the Model Context Protocol (MCP). You bring your own API key -- Monize supports Anthropic Claude, OpenAI, Ollama (local), and any OpenAI-compatible provider.
- Overview
- Setting Up AI
- AI Assistant
- Financial Insights
- MCP Server
- Usage Tracking
- Privacy and Security
Navigate to AI from the top navigation bar to access the AI assistant. The AI features require at least one configured provider -- either a central provider set up by the administrator or a personal provider you configure yourself.
Before using AI features, you need at least one configured AI provider -- either a central provider set up by the administrator, or a personal provider you configure yourself.
An administrator can configure a shared AI provider at the server level so that all users have access to AI features without needing their own API keys. This is done by setting environment variables on the server.
| Variable | Description |
|---|---|
AI_DEFAULT_PROVIDER |
Provider type: anthropic, openai, ollama, or openai-compatible
|
AI_DEFAULT_API_KEY |
API key for the provider |
AI_DEFAULT_MODEL |
Optional -- specific model to use (falls back to the provider's default) |
AI_DEFAULT_BASE_URL |
Optional -- custom base URL for OpenAI-compatible or Ollama providers |
For example, in your .env file:
AI_DEFAULT_PROVIDER=anthropic
AI_DEFAULT_API_KEY=sk-ant-...
When a central provider is configured, all users can use the AI assistant and financial insights immediately. Users who configure their own provider will use their personal provider instead of the central one.
- Navigate to Settings > AI
- Click Add Provider
- Select a provider type
- Enter your API key
- Optionally choose a specific model (or use the default)
- Click Save
You can configure multiple providers and assign priority levels. Monize will fall back to the next provider if the primary one is unavailable.
| Provider | Default Model | Description |
|---|---|---|
| Anthropic | claude-sonnet-4-20250514 | Claude models via the Anthropic API |
| OpenAI | gpt-4o | GPT models via the OpenAI API |
| Ollama | (user-configured) | Local, self-hosted models -- your data never leaves your machine |
| OpenAI-Compatible | (user-configured) | Any API that follows the OpenAI format (Groq, Together, LM Studio, etc.) |
For OpenAI-compatible providers, you can set a custom Base URL to point at your preferred endpoint.
After adding a provider, click Test Connection to verify that your API key and configuration are working. The test sends a minimal request to the provider and reports success or failure.
The AI assistant lets you ask questions about your finances in plain language. It uses tool calling to query your actual financial data and return accurate, sourced answers.
- Navigate to AI
- Type a question in the chat input
- The assistant streams its response in real time
Example questions:
- "How much did I spend on groceries last month?"
- "What are my top 5 expense categories this year?"
- "Compare my spending this month vs last month"
- "What's my net worth trend over the past 6 months?"
- "How much can I still spend this month within my budget?"
- "What's my income breakdown by category?"
The assistant has access to the following financial tools:
| Tool | Description |
|---|---|
| Query Transactions | Search and aggregate transaction data with filters |
| Get Account Balances | Retrieve account balances and net worth |
| Get Spending by Category | Break down spending by category for a date range |
| Get Income Summary | Income breakdown by source, category, or month |
| Get Net Worth History | Historical net worth trends over time |
| Get Budget Status | Current budget vs actual spending |
| Compare Periods | Side-by-side comparison of two time periods |
The AI page displays suggested query buttons to help you get started quickly. These are common financial questions tailored to typical use cases.
When the assistant returns data that lends itself to visualization, an inline chart is rendered directly in the chat. For example, a spending-by-category query may display a bar or pie chart alongside the text response.
Monize's AI automatically analyzes your financial data daily and generates actionable insights. Insights appear on the AI page and highlight patterns, anomalies, and opportunities you might otherwise miss.
| Type | Description | Severity |
|---|---|---|
| Anomaly | Unusual spending -- significantly higher than your historical average for a category | Warning / Alert |
| Trend | A sustained increase or decrease in spending over 3+ months | Info / Warning |
| Subscription | A recurring charge that has changed in price or a consolidation opportunity | Info |
| Budget Pace | You're on track to exceed your budget for the current period | Warning |
| Seasonal | Seasonal spending patterns detected (e.g., holiday months, annual renewals) | Info |
| New Recurring | A newly detected recurring charge that wasn't previously flagged | Info |
Insights are regenerated daily. A maximum of 50 insights are retained per user, and each insight expires after 7 days.
Click the dismiss button on any insight card to remove it. Dismissed insights will not reappear.
The Model Context Protocol (MCP) allows external AI clients -- such as Claude Desktop -- to connect to Monize and access your financial data directly. This means you can ask Claude Desktop questions about your finances without opening Monize in a browser.
- Navigate to Settings > AI
- Scroll to the MCP / Personal Access Tokens section
- Click Generate Token
- Give the token a name and select scopes (read, write, reports)
- Copy the token -- it will only be shown once
- Configure your MCP client (e.g., Claude Desktop) with the Monize MCP server URL and the token
Tokens are stored as SHA-256 hashes in the database. You can revoke a token at any time from the settings page.
External AI clients can use the following tools through MCP:
| Tool | Description |
|---|---|
| Accounts | List, query, and manage accounts |
| Transactions | Search transactions and create new ones |
| Categories | Browse the category tree |
| Payees | Query and manage payees |
| Investments | View portfolio holdings and performance |
| Scheduled Transactions | Manage recurring bills and deposits |
| Net Worth | Calculate current and historical net worth |
| Reports | Generate financial reports |
MCP clients can also read these resources for context:
| Resource | Description |
|---|---|
| Account List | All accounts with current balances |
| Category Tree | Full category hierarchy |
| Recent Transactions | Summarized transactions from the last 30 days |
| Financial Summary | High-level financial snapshot |
Pre-built prompt templates are available for common analysis tasks:
| Prompt | Description |
|---|---|
| Financial Review | Comprehensive financial health review |
| Spending Analysis | Detailed spending pattern analysis |
| Transaction Lookup | Guided transaction search |
| Budget Check | Budget status and recommendations |
Monize tracks AI token usage so you can monitor costs. Navigate to Settings > AI and check the Usage section to see:
- Total tokens used (input and output) per provider
- Request counts by feature type (query, insight)
- Token usage over time
Monize takes several measures to protect your financial data when using AI features:
| Measure | Description |
|---|---|
| Aggregated Data Only | The AI receives aggregated summaries, never individual transaction details |
| Encrypted API Keys | Provider API keys are encrypted at rest using AES-256-GCM with per-key random salts |
| Local Provider Option | Use Ollama to run models locally -- your data never leaves your machine |
| Rate Limiting | AI endpoints are rate-limited (10 queries/min, 5 test connections/min) |
| SSRF Protection | Custom base URLs are validated to prevent server-side request forgery |
| Prompt Injection Prevention | Financial data is delimited to prevent prompt injection attacks |
| Scoped MCP Tokens | Personal access tokens have configurable scopes (read, write, reports) and can be revoked at any time |
| Masked Keys | API keys are masked in all API responses, showing only the last 4 characters |