Wallet BudgetBakers MCP
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Wallet BudgetBakers MCP show my accounts"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Wallet by BudgetBakers — MCP Server
Read-only MCP server for accessing Wallet by BudgetBakers financial data from Claude.
Prerequisites
Wallet Premium subscription (API access requires Premium)
API token from https://web.budgetbakers.com/settings/apiTokens
Python 3.11+
uv package manager
Related MCP server: BudgetBakers Wallet MCP Server
Setup
1. Set your API token
export WALLET_API_TOKEN=your_token_here2. Claude Code (recommended)
The project includes .mcp.json — Claude Code auto-discovers the server. Just open the project directory:
cd wallet-bb-mcp
claude
# /mcp should show "wallet-bb" connected3. Claude Desktop / Cowork
Open Claude Desktop → Settings (gear icon) → Developer → Edit Config
This opens
claude_desktop_config.jsonin your editor. Add themcpServersblock:
{
"mcpServers": {
"wallet-bb": {
"command": "/Users/YOUR_USERNAME/.local/bin/uv",
"args": ["--directory", "/path/to/wallet-bb-mcp", "run", "server.py"],
"env": {
"WALLET_API_TOKEN": "your_token_here"
}
}
}
}Replace the values:
command— full path touv(runwhich uvin terminal to get it)--directoryargument — full path to this projectWALLET_API_TOKEN— your actual token (env var interpolation doesn't work here)
Save the file and restart Claude Desktop
You should see an MCP tools indicator (hammer icon) in the chat input area
4. Claude.ai Connector (remote)
The server can be deployed to Google Cloud Run and used as a remote MCP connector:
Go to claude.ai → Profile → Settings → Integrations
Click Add custom integration
Enter the server URL:
https://<your-cloud-run-service-url>/mcpNo authentication required — the Wallet API token is stored server-side
Cloud Run Deployment
The server supports both local (stdio) and remote (HTTP) transport. Cloud Run deployment is automated via GitHub Actions.
How it works
Push to
maintriggers.github/workflows/deploy.ymlCloud Build builds the Docker image from
DockerfileCloud Run deploys the new revision to
europe-central2Auth is handled via Workload Identity Federation (keyless)
Manual deployment
gcloud run deploy wallet-bb-mcp \
--source . \
--region europe-central2 \
--project wallet-bb-mcp \
--allow-unauthenticated \
--set-env-vars "WALLET_API_TOKEN=your_token" \
--port 8080Local HTTP mode
MCP_TRANSPORT=streamable-http WALLET_API_TOKEN=your_token uv run server.py
# Health check: curl http://localhost:8080/health
# MCP endpoint: http://localhost:8080/mcpAvailable Tools
Tool | Description |
| List financial accounts (bank, cash, cards). Start here. |
| Get transactions for an account (requires account ID) |
| Get specific records by IDs (max 30) |
| List spending/income categories |
| List budgets with limits and spent amounts |
| List savings goals with progress |
| List custom labels/tags |
| List recurring payments |
| List auto-categorization rules |
| Check API usage quota |
Filter Syntax
All text fields support filter prefixes:
Prefix | Example | Meaning |
|
| Exact match |
|
| Contains (case-sensitive) |
|
| Contains (case-insensitive) |
Range filters for dates, amounts, and timestamps:
Prefix | Example | Meaning |
|
| Greater than or equal |
|
| Less than or equal |
Combined |
| Between 100 and 500 |
Rate Limits
500 requests per hour
Use
get_api_usage()to check remaining quotaServer returns structured error with
Retry-Afteron HTTP 429
Available Tools
10 toolsget_accountsA
Get all financial accounts (bank accounts, cash, cards, etc.).
Use this first to discover account IDs — you'll need them for get_records().
Args: ctx: MCP context limit: Items per page (1-100, default 30) offset: Items to skip for pagination account_id: Filter by exact account ID name: Filter by account name (use prefix: eq., contains., contains-i.) account_type: Filter by type (e.g. general, loan, insurance) currency_code: Filter by ISO currency code (e.g. UAH, USD, EUR) bank_account_number: Filter by bank account number created_at: Filter by creation date (e.g. "gte.2025-01-01") updated_at: Filter by last update date (e.g. "gte.2025-01-01")
Inputs: name, limit, offset, account_id, +5 more. Returns: result
get_api_usageB
Get API usage statistics.
Args: ctx: MCP context period: Time period — format is Xdays, Xweeks, or Xmonths (e.g. "30days", "4weeks", "3months")
Inputs: period. Returns: result
get_budgetsA
Get all budgets with their limits, spent amounts, and periods.
Args: ctx: MCP context limit: Items per page (1-100, default 30) offset: Items to skip budget_id: Filter by exact budget ID name: Filter by budget name (use prefix: eq., contains., contains-i.) currency_code: Filter by currency (e.g. UAH, USD) created_at: Filter by creation date (e.g. "gte.2025-01-01") updated_at: Filter by last update date (e.g. "gte.2025-01-01")
Inputs: name, limit, offset, budget_id, +3 more. Returns: result
get_categoriesA
Get spending/income categories.
Categories are organized hierarchically (parent -> children). Use category IDs to filter records by category.
Args: ctx: MCP context limit: Items per page (1-100, default 50) offset: Items to skip category_id: Filter by exact category ID name: Filter by name (use prefix: contains-i.food) created_at: Filter by creation date (e.g. "gte.2025-01-01") updated_at: Filter by last update date (e.g. "gte.2025-01-01")
Inputs: name, limit, offset, created_at, +2 more. Returns: result
get_goalsA
Get all savings goals with current/target amounts and deadlines.
Args: ctx: MCP context limit: Items per page (1-100, default 30) offset: Items to skip goal_id: Filter by exact goal ID name: Filter by goal name (use prefix: eq., contains., contains-i.) note: Filter by note text (use prefix: contains-i.vacation) created_at: Filter by creation date (e.g. "gte.2025-01-01") updated_at: Filter by last update date (e.g. "gte.2025-01-01")
Inputs: name, note, limit, offset, +3 more. Returns: result
get_labelsA
Get custom labels (tags) used to organize records.
Args: ctx: MCP context limit: Items per page (1-100, default 50) offset: Items to skip label_id: Filter by exact label ID name: Filter by label name (use prefix: eq., contains., contains-i.) created_at: Filter by creation date (e.g. "gte.2025-01-01") updated_at: Filter by last update date (e.g. "gte.2025-01-01")
Inputs: name, limit, offset, label_id, +2 more. Returns: result
get_record_rulesA
Get auto-categorization rules for records.
These rules automatically assign categories/labels to matching transactions.
Args: ctx: MCP context limit: Items per page (1-100, default 30) offset: Items to skip rule_id: Filter by exact rule ID name: Filter by rule name (use prefix: eq., contains., contains-i.) created_at: Filter by creation date (e.g. "gte.2025-01-01") updated_at: Filter by last update date (e.g. "gte.2025-01-01")
Inputs: name, limit, offset, rule_id, +2 more. Returns: result
get_recordsA
Get financial records (transactions) for a specific account.
IMPORTANT: account_id is required. Use get_accounts() first to find it. Max date range is 370 days. If no date filter, returns last 3 months.
Date filter examples:
"gte.2025-01-01" — from Jan 1, 2025
"gte.2025-01-01,lte.2025-01-31" — January 2025
Amount filter examples:
"gte.100" — amount >= 100
"gte.100,lte.500" — between 100 and 500
Text filter prefixes: eq., contains., contains-i.
Args: ctx: MCP context account_id: Required. Account ID (use get_accounts to find IDs) record_date: Date range filter with prefix (e.g. "gte.2025-01-01") limit: Items per page (1-100, default 30) offset: Items to skip for pagination category_id: Filter by category ID label_id: Filter by label ID note: Filter by note text (use prefix: contains-i.grocery) payee: Filter expenses by payee (use prefix: contains-i.amazon) payer: Filter income by payer amount: Filter by amount (use prefix: gte.100,lte.500) record_type: Filter by type: income or expense. Transfers are not a separate type — they are identified by transferId/transferAccountId fields in the response sort_by: Sort field and direction (e.g. "recordDate,asc" or "amount,desc") created_at: Filter by creation date (e.g. "gte.2025-01-01") updated_at: Filter by last update date (e.g. "gte.2025-01-01")
Inputs: account_id (required), note, limit, payee, +10 more. Returns: result
get_records_by_idA
Get specific records by their IDs (comma-separated, max 30).
Args: ctx: MCP context record_ids: Comma-separated record IDs (e.g. "id1,id2,id3")
Inputs: record_ids (required). Returns: result
get_standing_ordersA
Get recurring/scheduled payments (standing orders).
Shows planned future transactions: bills, subscriptions, salary, etc.
Args: ctx: MCP context limit: Items per page (1-100, default 30) offset: Items to skip standing_order_id: Filter by exact standing order ID name: Filter by name (use prefix: eq., contains., contains-i.) currency_code: Filter by ISO currency code (e.g. UAH, USD, EUR) created_at: Filter by creation date (e.g. "gte.2025-01-01") updated_at: Filter by last update date (e.g. "gte.2025-01-01")
Inputs: name, limit, offset, created_at, +3 more. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 10 tools
Each tool targets a distinct resource (accounts, records, categories, budgets, goals, labels, standing orders, rules, API usage). The only potential overlap is get_records vs get_records_by_id, but the latter explicitly retrieves by IDs, making the distinction clear.
All tools follow the get_<resource> pattern with consistent snake_case. The sole variation, get_records_by_id, still adheres to the overall convention and is predictable.
10 tools is well-scoped for a read-only financial data API, covering all major entities without unnecessary bloat or missing core resources.
The tool set provides comprehensive read-only coverage of the budgeting domain: accounts, transactions, categories, budgets, goals, labels, standing orders, rules, and API usage. No obvious gaps exist for querying financial data, and filters allow for detailed retrieval.
Maintenance
Related MCP Connectors
Query your real net worth, spending, transactions, budgets and portfolio from any MCP client.
Read-only finance MCP server. Connects your banks, cards, payment apps, books and payroll into one ledger for your AI. Can't move money. Open source (AGPL-3.0), free during beta.
An MCP server that provides read access to your cloud storage providers, bank accounts and more.
Ask your AI about your own accounts, debts, budgets and goals. OAuth sign-in, read-only by default.
Related MCP Servers
- FlicenseAqualityDmaintenanceRead-only MCP server for accessing personal finance data from Ogarni.AI, including receipts, categories, summaries, notifications, and more.20-
- AlicenseCqualityDmaintenanceEnables querying of BudgetBakers Wallet data including accounts, transactions, categories, and labels through natural language commands.834 npmMIT
- AlicenseAqualityDmaintenanceRead-only MCP and HTTP proxy server for accessing Monarch Money financial data, enabling transaction analysis, budget tracking, and cashflow insights through natural language.6MIT

RiseUp MCP Serverofficial
AlicenseAqualityBmaintenanceMCP server for programmatic read-only access to RiseUp cashflow data, allowing AI assistants to retrieve budget information via natural language.71,180 npm40MIT