Categorize Transactions

io.github.theluckystrike由 GitHub 帳號 theluckystrike 發布v0.21.1更新於 Oct 25, 2026

Import a bank CSV, categorise it, summarise it per currency and reconcile it against your expenses.

已驗證Streamable HTTP可網頁執行檔案與儲存金融資料與分析

概覽

AI 產生的概覽

將銀行 CSV 匯出檔匯入本機帳本,依規則分類交易,依幣別彙總支出,並把扣款與費用紀錄對帳。

可以做什麼
匯入銀行 CSV 匯出檔,自動辨識表頭列、各欄位、日期順序與數字格式,重複匯入時略過既有資料列。列出、搜尋並手動分類已儲存的交易,管理自動套用的分類規則。依類別、月份、帳戶或交易對象彙總收入、支出與淨額,一律依幣別分別統計。找出週期性扣款,提供週期、典型金額、下次到期日與年度成本。將銀行扣款與費用紀錄對帳,並把某段日期範圍匯出為 CSV 或 JSON。
適用情境
適合手邊有銀行匯出檔、想在對話中看懂某個月份的情況:總額、未分類項目、訂閱扣款,以及哪些扣款沒有對應單據。面向不需要會計 SaaS 的個人或小型企業記帳。
執行需求
可作為本機程序執行(.mcpb 一鍵包,或以 Node.js 複製原始碼建置;npm 套件據說明尚未發布),也可連線託管端點。託管端點需要 Authorization bearer 標頭,值為 Pro 授權金鑰或匿名權杖。Pro 功能也可透過 MCP_LICENSE_KEY 環境變數或 license_activate 啟用。
安裝前請注意
涉及敏感的銀行流水資料;README 稱本機執行以純 JSON 檔案儲存且不發出網路請求,但託管端點會接收上傳的流水檔。Pro 為一次性付費購買。工具會寫入本機帳本、整份取代分類規則清單,並可刪除已上傳的流水檔。匯出可能覆寫既有檔案。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Categorize Transactions,將其新增到工作區。
  2. 連接 Authorization token 憑證。它們會被私密儲存,絕不會在公開頁面上展示。
  3. 為需要使用其工具的對話啟用該服務。

Web executable,透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式,或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南,先向你展示發布者、端點和登入方式,經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.theluckystrike%2Fcategorize-transactions/install.md,並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前,先告訴我它的發布者、端點和登入方式,等我確認。不要把 API 金鑰寫進命令或我會提交的檔案;問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "categorize-transactions": {
      "type": "http",
      "url": "https://mcp.zovo.one/mcp/categorize-transactions",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      }
    }
  }
}

README

mcp-bank-statement

In the official MCP Registry (io.github.theluckystrike/bank-statement-csv-categorize-reconcile-ledger).

[bank-statement demo]

Export the CSV from your bank, say "import this", and the month is readable. This MCP server turns a bank export into a local ledger: it finds the header row under whatever preamble the bank prints above it, works out which column is the date, which is the money and which way the money went, and reads amounts in the file's own locale, so 1 234,56 from mBank and 1,234.56 from a US bank both become the same number. A debit is stored negative and a credit positive, once, at import, so nothing downstream has to guess. Re-importing the same export adds nothing: every line carries a hash of its date, amount, currency and description, with an occurrence index, so two identical coffees on one day stay two transactions while a second import of the same file stays zero. Your rules categorise transactions as they arrive, summaries report money in, money out and the net per currency and never mix currencies, recurring_detect finds the subscriptions and what each costs per year, and reconcile_expenses matches bank debits against the receipts in mcp-expense-tracker so you can see which debits have no receipt and which receipts never reached the bank. Everything is a plain JSON file on your own machine; nothing is uploaded anywhere.

Turn a bank CSV into a categorised, reconciled month in chat, no accounting SaaS required.

Bank profiles: Revolut, Wise, mBank, PKO BP, ING, N26, and a generic reader that works from the headers alone.

File shapes it reads without being asked: UTF-8 and UTF-16 (either byte order, as Excel on Windows saves it), CRLF, LF and CR-only line endings, quoted fields holding commas, quotes and newlines, 1.234,56 and 1,234.56, accounting parentheses (12.50) and a trailing minus 12.50- (both negative), and a mixture of currencies inside one file. A cell holding - or nothing is skipped and named in skipped_lines rather than read as zero. When the date column is ambiguous (03/04/2026 with no value above 12 anywhere), the assumed order is stated in the response instead of silently picking one.

60-second install

npm publish for @theluckystrike/mcp-bank-statement is pending. Until then, the .mcpb one-click bundle or a clone+build is the working path, both are verified below.

One-click (.mcpb): download bank-statement.mcpb from the latest release and double-click it in Claude Desktop: https://github.com/theluckystrike/mcp-servers/releases/latest

(claude_desktop_config.json):

json
{  "mcpServers": {    "bank-statement": {      "command": "npx",      "args": ["-y", "@theluckystrike/mcp-bank-statement"]    }  }}

Claude Code:

sh
claude mcp add bank-statement -- npx -y @theluckystrike/mcp-bank-statement

(.cursor/mcp.json):

json
{  "mcpServers": {    "bank-statement": {      "command": "npx",      "args": ["-y", "@theluckystrike/mcp-bank-statement"]    }  }}

The npx form above starts working the moment the package is published. Until then, use the .mcpb bundle above, or build from source with exactly these three commands:

sh
git clone https://github.com/theluckystrike/mcp-servers.git && cd mcp-serversnpm installnpm run build -w packages/mcp-license -w servers/spreadsheet -w servers/bank-statement

Then point your client's command at node with one arg: the absolute path to servers/bank-statement/dist/index.js.

To run in Pro mode set MCP_LICENSE_KEY in the same config block, or call license_activate once with your key.

Tools

ToolWhat it does
statement_importRead a bank CSV into the ledger. Detects the header row under any preamble, the date, description, amount (or debit and credit) and currency columns, the date order and the number locale. Reports what it detected, what it stored, how many lines were duplicates of what is already there, and every line it skipped with the reason
transactions_listList transactions in a date range, filtered by account, category or uncategorized, with totals per currency
transactions_searchFind transactions whose description, counterparty, category or account contains the query. Substring only, never a regex
category_rulesSet or list the rules that categorise transactions. An empty match is refused, because it would stamp one category on the whole ledger. A plain match is a case-insensitive substring; regex: true compiles the pattern only if it cannot backtrack exponentially, and (a+)+ is refused and used as a substring. Setting rules re-applies them to stored transactions
transaction_categorizeSet the category on transactions by id. Nothing changes unless every id exists
statement_summaryMoney in, money out and the net for a range, grouped by category, month, account or counterparty, always per currency
reconcile_expensesMatch bank debits against the receipts in mcp-expense-tracker: same currency, same amount, within a date window. Reports matches, bank lines with no receipt, and receipts that never reached the bank. The expense ledger is only ever read
recurring_detectFind subscriptions and recurring charges: the same counterparty, an amount that barely moves, a steady cadence. Reports the cadence, the typical amount, the next expected date and the annualised cost. Two charges are one interval, so the yearly cost is withheld (cadence_confirmed: false) until a third charge confirms it
statement_exportWrite a range to csv or json and return the path. Written atomically, so a failure never leaves a half file. Replacing an existing file is allowed and reported (overwrote_existing_file)
accounts_listThe imported accounts, with bank, currencies, transaction count, date range and last balance
license_statusShow free or Pro mode
license_activateActivate a Pro key (verified offline)

Resource bank://month returns this month's money in, money out and net, grouped by category, per currency. Prompt monthly_review drives the whole month-end pass: totals, uncategorised lines, subscriptions, reconciliation.

Free vs Pro

FreePro
Accounts2unlimited
History read backlast 12 monthsall of it
Category rules5unlimited
Import, list, search, summariseyesyes
reconcile_expensesyes, 31 days at a timeany date range
recurring_detectyes, last 3 months, 5 chargesfull history, every charge
statement_exportnoyes

Import always stores every row, whatever the tier: silently dropping lines on the way in would make the ledger disagree with the bank. The free limit is on what is read back.

Get Pro: https://mcp.zovo.one/buy/bank-statement, $19 one-time for this server, $39 for every server, lifetime. Keys are verified offline; nothing is sent anywhere.

Pairs with

  • mcp-expense-tracker, log the receipts and the mileage; reconcile_expenses reads that ledger and tells you which bank debits have no receipt behind them.
  • mcp-spreadsheet, this server parses your bank CSV with the spreadsheet server's own RFC 4180 reader and locale number parser, so 1 234,56 means the same thing in both. Use it to slice an exported range.
  • mcp-invoice, the income side: raise the invoice, then watch the payment land in statement_summary grouped by counterparty.

Privacy

Everything stays on your machine. Statements are parsed locally and stored in a plain JSON file under ${XDG_DATA_HOME:-~/.local/share}/mcp-servers/bank-statement/. The server makes no network calls at all, not for licensing, not for parsing, not for anything. Deleting that directory resets it.

Built by theluckystrike. Support: [email protected]

Use these docs as an MCP server

Any MCP client (Claude, Cursor, Windsurf, VS Code) can read this repository's documentation directly via GitMCP - no install:

來源:servers/bank-statement/README.md,提交 6535b33

工具

15

statement_import

Import a bank CSV

Call this tool to read a bank CSV into the local ledger. Columns detected from headers (date, description, amount, currency, balance); amounts use the file's locale; stored lines skipped. Returns detected/stored/skipped.

輸入參數bankpath必填accountcurrencyoverwrite

transactions_list

List transactions

List stored transactions in a date range, optionally for one account or category, or only the ones no rule has categorised yet. Totals are reported per currency and never added across currencies.

輸入參數tofromlimitaccountcategoryuncategorized

category_rules

Set or list category rules

Read the category rules, or pass rules to REPLACE the whole list. A plain match is a substring; a regex compiles only if it cannot backtrack exponentially. An empty match is refused. Free: 5 rules.

輸入參數rulesreapply_all

transaction_categorize

Categorise transactions by id

Set or clear the category on transactions by id, for a one-off no rule is worth writing for. Every id is checked first, so one unknown id refuses the whole call. category_rules matches by text instead.

輸入參數ids必填category必填

statement_summary

What I spent, from the bank account

What was spent/received in a date range per the BANK ACCOUNT, grouped by category, month, account or counterparty. For "what did I spend in August" once imported. Totals are per currency, never summed or converted.

輸入參數totopfromaccountgroup_by

reconcile_expenses

Reconcile against the expense ledger

Match bank debits against mcp-expense-tracker entries: same currency and amount, date within a few days. Reports matches, unmatched bank lines, expenses never hitting the bank. Read-only. Free: 31 days; Pro: any range.

輸入參數to必填from必填accountwindow_days

recurring_detect

Subscriptions and recurring charges in the bank data

Find charges that come back: debits grouped by counterparty and currency with a steady amount and interval, with cadence, typical amount, next due date and annual cost. Free: 3 months, 5 charges.

輸入參數monthsaccountmin_occurrences

statement_export

Export bank transactions to a file

Export the BANK transactions of a date range (a month, a quarter, a year) as .csv or .json and return a download link valid for one hour. This is the tool for "export September" once a statement has been imported. Nothing partial is ever written.

輸入參數to必填from必填path必填format必填accountcategory

accounts_list

List accounts

List the imported accounts: bank, currencies, transaction count, first and last date, and the closing balance when the file carried one. Start here for the account names transactions_list and statement_export take.

無需輸入參數。

license_status

License status

Report this endpoint's licence state for your token as JSON: the product, the tier free or pro, why it is not Pro, and the checkout URL. Call it to explain a free-tier refusal. No arguments, nothing changes.

無需輸入參數。

license_activate

Activate license

Turn Pro on for this connection with key, an MCPL1.<payload>.<signature> issued at checkout for this server or the bundle. Data under your token stays; a wrong or expired key changes nothing. license_status confirms it.

輸入參數key必填

bank_upload

Upload a bank statement file

Send a bank export to this hosted endpoint. There is no filesystem here, so instead of a path you upload the file once with bank_upload and then pass its name as `path` to statement_import. Give exactly one of content (the export as text, header row included - this is the normal case for a CSV), content_base64 (the file's bytes, which keeps a UTF-16 export from Excel readable) or url. url: fetch a public file instead of pasting base64 (recommended above about 10 KB): the url is fetched here with a 10 second timeout, at most 3 redirects, public http(s) hosts only, and a 1 MB cap; the fetched file has to be delimited text, so a PDF statement is refused rather than stored. Uploads are kept for your token between calls; bank_files lists them and bank_delete_upload removes one. The request body cap is 256 KB, so a large export pasted as text has to be split by month.

輸入參數urlname必填contentcontent_base64

bank_files

List uploaded statements

List the statement files stored for your token, with their sizes.

無需輸入參數。

bank_delete_upload

Delete an uploaded statement

Delete one statement file stored for your token. The transactions already imported from it are kept; transactions_list and the summary still see them.

輸入參數name必填

版本歷史

2
  1. v0.21.1最新Sep 23, 2026
  2. v0.21.0Sep 16, 2026

評分

已滿足適用檢查項的 100%

該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分