IntentEnforce
Decide what a request is for before your model sees it.
Most filters look at text. IntentEnforce classifies the intent of every prompt against categories you define, then applies your policy before traffic reaches your model.
- Allow, block or clarify rules per intent, with confidence taken into account.
- Proxy layers: put a URL in front of your model endpoint with no code changes, and review a log of every decision.
- Responses are checked by a validator before they go back to the user. Free on your own Claude or Gemini API key.
Know the limit
An intent classifier is a model and can be wrong. Treat it as one layer, and pair it with a deterministic control for high-impact actions.