FAQ
Common questions about AI Spend Monitor coverage, accuracy, and limits.
Updated
Which plugins does it cover?
Any plugin or theme that makes AI calls through the WordPress 7.0 native AI Client — automatically, with zero integration work, including plugins released in the future. Plugins that call providers directly with their own API keys (outside the AI Client) are not visible to the Monitor.
Does it work without WordPress 7.0?
It activates on older versions but cannot record anything, because it relies on the AI Client that ships with 7.0.
Are the cost figures exact?
No — estimates from token counts × list prices. Your provider invoice is the source of truth. Rates are adjustable via the aismon_cost_rates filter.
Does it slow down AI calls?
No. Recording happens after a call completes and is a single local database insert. If recording ever fails, the original AI call is unaffected.
Does it record prompts?
Never. Token counts, provider, model, capability, and the calling plugin only.
Can it block or limit AI usage?
The Monitor watches and notifies. Budgets, multi-threshold alerts, the hard stop, and the kill switch are AI Spend Governance — the paid add-on that uses the Monitor as its data layer.
What’s the “Blocked calls” card then?
When AI Spend Governance is active, calls it blocks are logged into the Monitor’s dashboard so you can see exactly what the hard stop saved you from. Without Governance, the card stays at 0.
Question about this page?
This form tags your question with the product, docs page, and category so support can triage it quickly.