How to Handle a Flagged Transaction
A flagged transaction is the firewall asking a question. Here's how to answer it well.
Step 1 — Read the context
Amount, merchant, category, and the rule that fired — all in the queue.
Step 2 — Decide
Approve if legitimate; deny if not. Both are logged.
Step 3 — Tune
Patterns of approval → raise the threshold. Patterns of denial → the agent is trying the wrong things.
The loop
FLAG -> review -> approve/deny -> log -> tune threshold
Review target: minutes, not days.
The queue is designed to be fast.
FAQ
Who should review flags?
The budget owner — the same person who sets the thresholds.
What if I'm unsure?
Deny by default — the agent can retry with a different approach.
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