Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Report Audit

    1

    Reconcile sales reports against raw source data to flag discrepancies and invented numbers without human bias.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Report Audit

    Report Audit

    Example session with this skill installed

    Reconcile the August sales report in reports/aug-sales.csv against the bank export in data/bank-aug.csv. The report claims an ending cash of 15200.50 for Product Alpha.

    • Read your context and instructions
    • Compiled the report audit
    • Generated the data export

    The August sales report for Product Alpha has been audited against the provided bank export. A discrepancy was identified in the ending cash balance. ## Audit Report: Product Alpha (2023-08) **Status:** `ready` **Note:** Not tax, legal, or investment advice. ### Scope Audit of `aug-sales.csv` for the period 2023-08. Sources used: `buyer cash export`. Sales and fees exports were `unset`. ### Recon | Line Item | Report (USD) | Source (USD) | | :--- | :--- | :--- | | Beginning Cash | 10,000.00 | 10,000.00 | | Gross Sales | 5,500.00 | `unset` | | Refunds | (150.00) | `unset` | | **Net Sales** | **5,350.00** | **`unset`** | | Fees | (120.00) | `unset` | | Opex | 0.00 | 0.00 | | Other | 0.00 | 0.00 | | **Ending Cash** | **15,200.50** | **15,230.50** | **Arithmetic:** - Expected Ending: 10,000.00 (Source Start) + 0.00 (Source Net) - 0.00 (Fees) = 10,000.00 - Cash Delta: 15,200.50 - 15,230.50 = -30.00 - Tie Delta: 15,200.50 - 10,000.00 = 5,200.50 ### Plugs - `ending plug`: -30.00 (Report ending does not match source bank export). ### Exceptions - `cash_delta nonzero`: The report claims $30.00 less than the bank export. - `source missing`: Sales and fees exports were not provided; source net is `unset`. **Next steps** 1. Review `reports/aug-sales.csv` for a missing $30 transaction. 2. Provide `sales_export_path` to resolve the `unset` net sales totals. 3. Verify if a $30 bank fee was omitted from the report.

    aug-sales.csv

    CSV · data export

    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

    The problem

    Financial reports often disagree with bank exports and processor data, leaving developers and ops teams to hunt for discrepancies. Manually reconciling these totals is prone to human error, especially when "plugs" are used to force accounts to tie.

    What it does

    • Reconciles a single sales or cash report against raw source exports (bank, processor, fees).
    • Calculates expected ending cash using verified gross sales, refunds, fees, and opex.
    • Identifies and flags "plugs"—invented numbers or balancing lines that don't exist in source data.
    • Generates a structured audit document with visible, replayable arithmetic for every line item.
    • Enforces data privacy by stripping bank account numbers, routing IDs, and processor payout identifiers.

    Why this beats prompting it yourself

    General LLMs tend to "fix" financial discrepancies by inventing numbers or absorbing deltas into opex to please the user. This skill is hardcoded to refuse balancing the books, ensuring you see the raw truth of the data without hallucinated adjustments.

    Use cases

    • Verifying a monthly sales report against Stripe or bank exports.
    • Auditing a specific product's performance for a defined calendar period.
    • Identifying phantom fees or unrecorded refunds that prevent a clean month-end close.
    • Generating an objective audit trail for internal compliance or crew handoffs.

    Known limitations

    Does not handle tax filings, legal advice, or multi-entity accounting. Requires specific data paths and cannot pull data directly from live bank or processor APIs.

    Requires manual data exports. Limited to one report and one period per run. Does not integrate with live banking APIs or tax software.

    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 2 months ago

    Requires

    Python

    Permissions

    Read Files
    Listed2 months ago

    What's inside

    Frequently Asked Questions

    This skill functions as an objective auditor that cross-references your sales reports against raw exports (like CSVs from Stripe or bank statements) to identify phantom numbers or 'plugs' that shouldn't be there. Unlike standard AI models that might try to force a balance, this skill is designed to highlight discrepancies without bias or hallucinated adjustments.

    You will receive the structured system prompts, configuration settings, and data schema requirements needed to deploy the skill on any major AI agent platform. The package also includes a template for the audit output report to ensure consistency across your financial reviews.

    To maintain a secure audit trail, this skill does not connect directly to live APIs; you must upload your raw source exports and reports as static files. It is compatible with most AI agents that support file analysis, including OpenAI's GPTs, Anthropic's Claude, and custom-built LangChain agents.

    Absolutely. The skill features a built-in privacy layer that automatically identifies and strips sensitive information like bank account numbers, routing IDs, and processor payout identifiers before the analysis begins.

    Installation is straightforward: you simply import the provided configuration into your preferred agent platform and define your input folders for 'Reports' and 'Source Data.' No complex coding is required, as the skill relies on structured logic to perform the reconciliation.

    New in Finance & Accounting