Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    statement analyzer

    1

    Banks swap debit and credit columns, exports skip transactions, and transfers between accounts get counted as expenses.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    statement analyzer

    statement analyzer

    Example session with this skill installed

    I've uploaded three CSVs from my HDFC bank account and my Amex card for the last six months. Can you tell me how much I spent on dining out and what my balance was at the end of each month?

    • Read your context and instructions
    • Compiled the statement analyzer
    • Generated the document

    I have processed the three statements provided (HDFC Account ending in 4492 and Amex Card ending in 1005) covering January 2024 to June 2024. The data reconciled with 100% accuracy against the stated balances.

    Financial Summary: Jan-Jun 2024

    MonthHDFC Ending BalanceDining & DeliveryTop Payee (Dining)
    Jan 2024₹1,42,300₹12,450Zomato
    Feb 2024₹1,28,900₹8,200Swiggy
    Mar 2024₹1,55,200₹15,600Metabolic Fitness
    Apr 2024₹1,10,450₹9,100Starbucks
    May 2024₹1,98,000₹11,300Zomato
    Jun 2024₹1,75,400₹7,850Blinkit

    Analysis Notes

    Internal Transfers: Found ₹50,000 monthly transfer from HDFC to Amex. These have been excluded from "Spending" to prevent double-counting.

    Deduplication: 14 transactions from your expense app were ignored as they matched the HDFC bank statement exactly.

    Categorization: "Metabolic Fitness" was tagged as Dining based on your previous input; let me know if this should be Health.

    Next steps

    • Review the report.html file for a full breakdown of other categories like Rent and Utilities.
    • Check questions.json for 4 large transactions I couldn't identify.
    • Update profile.json if the "Unmatched self-transfers" (totaling ₹12,000) were actually sent to a second savings account.

    report.html

    PDF · document

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Financial statements are messy. Banks swap debit and credit columns, exports skip transactions, and transfers between accounts get counted as expenses. This skill processes raw bank statements, credit card exports, and expense app CSVs to provide a reconciled, categorized view of where money actually went.

    What it does

    • Multi-source reconciliation matches bank balances with transaction lists to ensure zero data loss.
    • Deduplication removes internal transfers between your own accounts and credit card bill payments to avoid double-counting.
    • Automated categorization uses a pattern-matching engine to sort spending into groups like family, investments, and utilities.
    • Interactive profiling asks for context on recurring transfers and unknown payees to build a personalized financial map.
    • Report generation produces a self-contained HTML dashboard and a JSON summary of all month-end balances.

    How it works

    1. Ingest statements by listing accounts, checking date coverage, and converting PDFs to structured CSVs.
    2. Parse the raw data using provided scripts to verify that opening and closing balances match transaction totals.
    3. Refine the analysis by answering specific questions about large unknown transactions and internal account names.
    4. Generate the final HTML report and a summary markdown file containing the month-by-month financial breakdown.

    Frameworks & tools

    This skill uses Python 3 for data processing. It requires openpyxl for Excel files and pdfplumber for PDF extraction.

    Why this beats prompting it yourself

    LLMs often hallucinate arithmetic or lose track of transactions in long statements. This skill uses deterministic Python scripts to handle the math while using the AI only for categorization and user interaction, ensuring the final balance is always accurate.

    Use cases

    • Auditing a full year of spending across multiple bank accounts and cards.
    • Identifying why a bank balance is lower than expected by tracking income versus all outflows.
    • Visualizing the ratio of investments to discretionary spending.
    • Preparing a clean summary of month-end balances for accounting or tax purposes.

    Known limitations

    Category accuracy depends on merchant name quality in bank exports. Users must manually remove passwords from encrypted PDF statements before processing.

    Requires Python 3. Logic focuses on reconciliation and categorization. Does not provide financial advice. Password-protected PDFs must be decrypted by the user first.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    Requires

    Python

    Permissions

    No special permissions declared or detected

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    This skill uses deterministic Python scripts to perform all calculations, ensuring your ending balances are mathematically perfect, whereas generic LLM prompts often hallucinate numbers when processing long transaction lists.

    You will receive the complete Python source code for the analyzer, the necessary configuration scripts for PDF and CSV parsing, and a guide for integrating it with your preferred AI agent framework.

    The skill is designed for any AI agent environment that supports Python execution, such as AutoGPT, LangChain, or custom OpenAI Assistants, provided the environment has `pdfplumber` and `openpyxl` installed.

    Yes, the Deduplication feature is specifically designed to recognize and filter out transfers between your own linked accounts and credit card payments to prevent artificial inflation of your spending reports.

    The tool generates a self-contained, interactive HTML dashboard for visual review, along with a structured JSON summary and a Markdown report for easy record-keeping or further analysis.

    To protect your security, this skill does not support password-protected files; you must remove the encryption from your bank PDFs using your standard PDF viewer before uploading them for analysis.

    New in Finance & Accounting