Skip to main content

Proxy

Proxy Base URL: https://api.promptguard.co/api/v1Point your LLM SDK’s base_url / baseURL to this endpoint. Your existing provider API key goes in the Authorization header as usual; add your PromptGuard key via X-API-Key.
Model ids starting mistral-, ministral-3, ministral-8b, ministral-14b, magistral-, codestral, devstral, voxtral, pixtral- or ocr- go to Mistral. Hugging Face’s mistralai/ hub ids are not Mistral’s API: they go to Hugging Face. Mistral speaks OpenAI’s wire format, so PromptGuard forwards your request to api.mistral.ai/v1 as you sent it and returns Mistral’s reply unchanged. Any OpenAI client works.
To keep the Mistral key out of your application, store it as a provider key under mistral. PromptGuard then sends the stored key and ignores the Authorization header you send.

Auto-instrumentation

The SDKs do not patch mistralai. Python’s init() names it in its startup warning when it is installed, so you know its calls are not scanned. Use the proxy.