Scenario: Unknown Vendor Payment
An agent finds a service it needs, from a vendor nobody has vetted, and initiates payment. This is the second-most-common runaway shape in the incident database.
What happens without a firewall
The payment settles. The vendor is real — or a typo-squat. Either way, procurement finds out on the invoice.
What happens with sipi.bot
Merchant not on the allowlist → BLOCKED. The agent can't pay an unvetted vendor, full stop.
The legit-new-vendor path
Set the rule to FLAG new merchants instead: the purchase waits in the approval queue while procurement vets the vendor.
The decision
{
"decision": "BLOCKED",
"reason": "Merchant not on allowlist",
"rule_id": "rul_allow_07",
"transaction_id": "txn_5f31e0"
}Unknown vendors are the default-deny.
FAQ
Does this block legitimate new vendors?
Only if the rule hard-blocks. FLAG instead: new vendors wait for a human, then proceed.
Is unknown-vendor spend common?
It's one of the documented patterns in the incident database.
Related
Stop the next $12,400 night.
One API call (or MCP tool) in front of every agent transaction — APPROVED, BLOCKED, or FLAGGED, deterministic, ~5 ms, fully logged.
See plans — from $99/mo Try a live check