TokenVeil in short
TokenVeil is a self-hosted anonymization proxy that sits between your teams and AI models (Claude, Gemini, ChatGPT, Mistral and others). It strips names, IPs, secrets and identifiers before a message reaches an external model, then restores them in the displayed response. The real data never leaves your infrastructure.
How it works
Section titled “How it works”- Your teams write as usual in the TokenVeil chat, in VS Code, or through the file tool.
- The anonymization engine replaces every sensitive value with a neutral token (
<PERSON_1>,<IP_1>) before the message reaches the AI provider. - The response comes back tokenized and TokenVeil restores the real values locally, at display time. The AI provider never saw the real data and cannot recover it: it does not hold the mapping.
Where to start
Section titled “Where to start”- Evaluating the tool: Community vs Enterprise, then the quickstart (ten minutes, Docker).
- Deploying the Community edition: Installation and Configuration.
- CISO or DPO: Security, the GDPR dossier and the technical measures.
- Deploying the licensed Enterprise edition: Enterprise installation.
The product site is tokenveil.eu, with the public benchmark and the roadmap.