Works with the AI tools you already use

    CClaude CodeCCursorCCodex CLIGGitHub CopilotGGemini CLIVVS CodeWWindsurf+15 more

    Invoice & Expense Sanity Gate

    by PubsProToolkit

    1

    The Invoice & Expense Sanity Gate is an adversarial reviewer for invoices, vendor bills, receipts, and expense reports.

    Secure checkout via Stripe

    0 installsSecurity scanned

    See it in action

    You say

    Invoice #1042 to Acme Co.\n3 x Consulting hours @ $150 = $450\n1 x Setup fee @ $200 = $200\nSubtotal: $650\nTax (8%): $52\nTotal: $700\nTerms: Net 30. Pay to: (bank details not shown)

    Your agent does

    VERDICT: REVISE\nSUMMARY: Two issues — a grand-total error and missing payment details; do not send yet.\n\nARITHMETIC:\n - Subtotal 650 + tax 52 = 702, but Total shows 700 (off by 2).\n - Tax at 8% of 650 = 52.00 is correct; the error is in the grand total.\nCOMPLETENESS:\n - Payment instructions / bank details are missing.\n - No invoice date is shown.\nDUPLICATES:\n - None detected.\nPOLICY & PLAUSIBILITY:\n - Line items and rates are plausible.\nRED FLAGS:\n - None.\n\nFIX FIRST:\n - Correct the grand total to 702.00.\n - Add invoice date and payment instructions before sending.

    What you get

    Checking an invoice for math and missing details before sending it to a clientVetting a vendor bill for duplicates or red flags before approving paymentReviewing an expense report against policy limits before approval

    About this skill

    The Invoice & Expense Sanity Gate is an adversarial reviewer for invoices, vendor bills, receipts, and expense reports. You hand it a finished document — an invoice you are about to send, a bill you are about to pay, or an expense report you are about to approve — and it stress-tests it for the mistakes that quietly cost money: arithmetic that does not reconcile, tax applied to the wrong base, missing payment or invoice details, duplicate or already-paid line items, entries that break a stated policy, and signals that warrant a second look before payment.\n\nIt runs five passes — arithmetic, completeness, duplicate/already-paid, policy & plausibility, and red-flags — and returns a structured PASS / REVISE / HOLD verdict with the highest-impact fixes listed first. It is built to be the last check before money moves, especially when an AI agent or template generated the document and no one has verified the numbers line by line.\n\nIt judges; it does not generate. It never creates invoices, fills in numbers, categorizes transactions, gives tax or legal advice, or connects to your bank or accounting software.

    How to install

    Drop the file into your AI Agent. Works with Claude, Cursor, ChatGPT, and 20+ more.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    30-day refund

    Not a fit? Get your money back

    Trust & safety

    Security scanned

    Verified clean 1 month ago

    Listed1 month ago

    Creator

    PubsProToolkit builds rigor-first skills for AI agents — they write your docs and content properly, then adversarially review them to catch what's wrong before it ships. The result: cleaner output and a hard quality gate in one toolkit. Built by a CMPP-certified, PhD medical writer who brings regulated-industry standards to developer docs, content, compliance, and research integrity.

    Frequently Asked Questions

    Popular in Finance, Legal & Accounting