Is my assistant's context encrypted before leaving my machine?

Yes. The Privatemode proxy runs locally and encrypts all data before it leaves your machine. It also performs remote attestation to verify the integrity of the Privatemode service, ensuring end-to-end confidentiality for every request OpenClaw makes.

Does Privatemode store or retain data from my conversations?

No. Privatemode never learns from your data. All requests are processed using confidential computing, and no prompts, memory entries, or responses are retained after processing.

Do all OpenClaw features work with Privatemode?

Any OpenClaw feature that uses the configured agent model, including chat, browser automation, and skills, will route through Privatemode. Features that depend on a specific provider's proprietary API may need that provider configured separately.

How does confidential computing affect performance?

Response times are comparable to standard cloud AI services, while your data stays encrypted in memory during processing.

Integrating Privatemode AI with OpenClaw

OpenClaw is a personal assistant you reach through the messaging apps you already use, with memory, browser automation, and shell access behind it. With Privatemode as its model provider, everything it reads and remembers is processed by open-weight models in an environment whose memory stays encrypted.

Introduction

Use cloud AI in OpenClaw without exposing your assistant's context

The challenge

Personal assistants need powerful models

OpenClaw runs a conversational agent across messaging apps with persistent memory, browser automation, and shell access. Frontier LLMs make the assistant feel competent, but conventional providers process every prompt in plaintext on their servers. For an assistant wired into your personal life and workflows, that is a hard limit.

The solution

Encrypted before it leaves your machine

Privatemode exposes an OpenAI-compatible endpoint that OpenClaw connects to as a custom provider. Every request is encrypted before it leaves your machine and processed inside a hardware-isolated environment whose memory stays encrypted, so by design neither the infrastructure provider nor Privatemode can read your assistant's context. This is verifiable through remote attestation and open-source code.

Benefits

Why use Privatemode AI with OpenClaw?

Built in, nothing to install

End-to-end confidentiality

Learn how

Free tier available

How to get started

How to set up Privatemode in OpenClaw

Using Privatemode as your model provider keeps all OpenClaw AI interactions confidential by design. You still have to take care of reducing the blast radius of your agent's actions. Privatemode does not manage agent permissions or action scope. You should carefully read OpenClaw's security and sandboxing guidelines.

Get your API key

If you don't have a Privatemode API key yet, you can generate one for free here.

Run the Privatemode proxy

The proxy verifies the integrity of the Privatemode service using remote attestation. The proxy also encrypts all data before sending and decrypts data it receives.

Install OpenClaw

Install OpenClaw with npm (Node.js 22 or newer) and run the onboarding, or use the installer script from the official docs.

Configure OpenClaw

This terminal session shows a minimal example configuration to help you get started with OpenClaw. We use Telegram here for its simple setup. Adjust the configuration to fit your needs.

FAQ

Frequently asked questions about using Privatemode with OpenClaw

Minimal. OpenClaw takes the proxy as a custom provider in its config: an entry under models.providers with the proxy URL as baseUrl and api set to openai-completions, then the model is selected as the agents' primary model.