invoices-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| INVOICES_DIR | Yes | Path to the folder containing your invoice files (Nubank CSVs and Itaú PDFs). Can also be supplied with the --dir command-line option, e.g. --dir /path/to/invoices. | |
| INVOICES_GMAIL_QUERY | No | Gmail search query used to find invoice attachments, e.g. label:invoices. Must contain from:, label: or subject:. Required only if you enable Gmail fetching. Can also be supplied with --gmail-query. | |
| INVOICES_DRIVE_FOLDER | No | Google Drive folder id or folder link whose files are copied into the invoice folder. Required only if you enable Drive fetching. Can also be supplied with --drive-folder. | |
| INVOICES_PDF_PASSWORD | No | Password used to open password-protected PDF statements. It is used only to open the file and is never stored or logged. | |
| INVOICES_GOOGLE_CREDENTIALS | No | Path to the Google OAuth client credentials JSON file (Desktop app type). Required only if you enable the optional Gmail and/or Google Drive invoice fetching. Can also be supplied with --google-credentials. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_banksA | Which banks and file formats this server can read. Files from other banks are not supported yet and are reported as skipped by list_invoices. |
| list_invoicesA | List the invoice files that were found: bank, month, type (card or account), due date, transaction count and spend. Where a statement prints its own total, |
| spending_summaryA | Totals, spend by bank and category, top merchants and largest purchases. Optionally narrow by bank, invoice, date range or category. Amounts are in BRL. A positive amount is money leaving the account or card (a purchase or fee); refunds, card payments and deposits are reported separately as credits. |
| search_transactionsA | Find individual transactions by text, bank, date range, category or amount, newest first. min_amount and max_amount compare the absolute amount in BRL. Amounts are in BRL. A positive amount is money leaving the account or card (a purchase or fee); refunds, card payments and deposits are reported separately as credits. |
| monthly_totalsA | Spend and credits per calendar month, optionally for one bank or category. Months come from transaction dates. Amounts are in BRL. A positive amount is money leaving the account or card (a purchase or fee); refunds, card payments and deposits are reported separately as credits. |
| compare_monthsA | Compare spending by category between two calendar months and show what changed most. Amounts are in BRL. A positive amount is money leaving the account or card (a purchase or fee); refunds, card payments and deposits are reported separately as credits. |
| find_recurring_chargesA | Detect likely subscriptions: the same merchant charged a similar amount (within 5%), about once a month, in several different months. Installment purchases are ignored. Amounts are in BRL. A positive amount is money leaving the account or card (a purchase or fee); refunds, card payments and deposits are reported separately as credits. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 7 tools
Most tools have clearly distinct purposes: list_banks vs list_invoices separate source banks from statement files, search_transactions is item-level, compare_months and find_recurring_charges are specific analyses. spending_summary and monthly_totals overlap somewhat since both produce aggregated totals, but the monthly and category/bank breakdowns help differentiate them.
All names use consistent snake_case, which is good. However, the verb-first convention (list_, search_, compare_, find_) is broken by two noun-phrase names (spending_summary, monthly_totals), a minor deviation that is still readable.
Seven tools sit squarely in the well-scoped range for an invoice-analysis server. Each tool covers a distinct analytic need (inventory, totals, trends, search, comparison, recurring detection) with no redundant entries.
The read-only analysis surface is well covered: discovery, summaries, monthly trends, search, comparison, and recurring-charge detection. Minor gaps remain, such as per-transaction drill-down detail or category management, but these are workable given the apparent read-only scope.