I will verify every new transaction against your customer and invoice records, then update its status with the evidence I found.
Automated payment matches still need proof before anyone can close the books, though confirming it by hand means digging through email and records. This agent runs an escalating verification check on every transaction and flags only unproven matches for manual review.

Verification status on each new transaction, delivered every time a transaction is created or when you mention the agent.
Evidence trail for every verification decision, delivered every time a transaction is created or when you mention the agent.
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur.
Block quote
Ordered list
Unordered list
Bold text
Emphasis
Superscript
Subscript
Manual review flags with clear explanations, delivered every time a transaction is created or when you mention the agent.
After a new transaction lands, the agent pauses briefly to let upstream data finish parsing before it touches anything, so it never works from incomplete records.
It compares the extracted customer name against the recommended match, ignoring generic corporate terms like Ltd or Inc, and closes the loop immediately if the names clearly line up.
If the name match is inconclusive, it searches connected email for messages referencing the customer or invoice and the transaction amount, looking for real proof before moving on.
When no email evidence turns up, it uses Web Search to check trusted sources for a legal or commercial relationship between the two entity names.
If none of the checks pass, it sets the transaction to manual review and logs exactly which checks it ran and why each one fell short, so a human picks up with full context.