Mistral AI integration
Mask personal data before it reaches Mistral AI
- Every text field in the request body: user and system messages, tool and function-call arguments, and tool results, across chat completions and any other Mistral endpoint
- Embeddings inputs, masked deterministically
- Streaming responses
- Base64, hex, and percent-encoded text inside the request is decoded and checked too
Teams often choose Mistral for European data handling, but a European model provider is still a third party receiving whatever is in the prompt. Under GDPR, sending less personal data is better than sending it to a better processor.
Maskflare masks it before it leaves. Requests to the Mistral API go through your gateway, sensitive values become tokens, and with response restoration on, the answer comes back with the real values.
Set up Mistral AI with Maskflare
Requests go to /v1/mistral on your Maskflare gateway instead of the provider.
curl https://YOUR_GATEWAY_HOST/v1/mistral/chat/completions \
-H "Authorization: Bearer mf_live_..." \
-H "Content-Type: application/json" \
-d '{"model": "mistral-large-latest",
"messages": [{"role": "user", "content": "Summarize the complaint from luc.martin@example.fr"}]}'What gets masked
- Every text field in the request body: user and system messages, tool and function-call arguments, and tool results, across chat completions and any other Mistral endpoint
- Embeddings inputs, masked deterministically
- Streaming responses
- Base64, hex, and percent-encoded text inside the request is decoded and checked too
Good to know
- Paths after /v1/mistral map to the same paths after https://api.mistral.ai/v1.
- A rule set to block stops the request before it reaches the provider and returns HTTP 403 with the matching rule keys, never the values.
- Store the provider key once in the Maskflare console, or keep sending it per request in pass-through mode.
- No code change option: the Maskflare forward proxy inspects traffic to api.mistral.ai with the same rules.
Frequently asked questions
Why mask data if Mistral is a European provider?
Data minimisation still applies. Masking means personal data isn't processed by the model provider at all, which simplifies your records of processing and your data protection impact assessment.
Does it work with Mistral embeddings?
Yes. Embeddings inputs are masked deterministically, so equal values embed identically without being sent in the clear.
How does the request authenticate?
Send a Maskflare key as the Bearer token. Your Mistral key is stored in the console, or sent separately in pass-through mode.
Other providers
See it on your own data
Book a 30-minute demo.
Bring your hardest prompts.
We'll show detection, masking, and restoration on your providers and data types,
and how it fits your stack.