Your data, under your control
Export everything. Delete everything. Inspect everything.
AI safety you can trust
Setting up personal agents is hard. We make it easy.
Personal agents such as OpenClaw, Hermes, etc can't safely run on your laptop without time-consuming restrictions. They need their own server with full system access to do what makes them valuable. We provision it, harden the setup, and hand you a working agent. You bring your API keys.
Single-tenant · hardened by default · your keys stay yours
Custody summary
- Instance
- Dedicated, no shared tenancy
- Access
- SSH key only · root disabled
- Credentials
- Stored on your instance alone
- Audit
- Outside the agent's reach · exportable
- Training
- Never used for models
What you get
Everything you need to run agents safely. Nothing you don't.
One opinionated platform for orchestrating AI agents, with safety and observability built into the core — not bolted on.
Multi-agent orchestration
Run teams of agents. Hot-swap models. Full logs for debugging if something goes sideways.
Your choice of AI
Anthropic Claude, OpenAI GPT, plus Bedrock, Azure OpenAI and self-hosted endpoints. Switch anytime. No vendor lock-in.
Production observability
Token, latency, and cost telemetry per agent, per run, per tenant. Slice by anything. Alert on anything.
Audit logs the agent can't reach
Every prompt, tool call and decision is recorded outside the agent's reach, so it can't rewrite its own history. Export the lot for your auditors.
Security updates, handled
OS patches, dependency updates, security fixes — we keep your server current so you don't think about it.
Pro tier
Small local models, on your own machine
Pro includes local inference on your instance — no API call required. Heavy lifting still flows through your Claude and GPT keys. Pro gives your agent capabilities for the tasks that shouldn't leave the building.
- Local embeddings for semantic memory and document search.
- Speech-to-text — voice notes transcribed locally via whisper.cpp.
- Small vision models for image understanding, no roundtrip.
- Faster retrieval. Memory and search run on your hardware.
Full local LLM support on GPU instances later this year.
Privacy, by default
Your data is yours. Full stop.
Built for people who care about AI safety and privacy. Most AI platforms treat your data as their training set. Dorne treats it as a liability — minimized, scoped, encrypted, and deleted on your command.
- i
You own every artifact.
Prompts, completions, tool outputs, embeddings — exportable anytime as portable JSON.
- ii
Your keys, your machine.
Entered during setup, stored only on your instance, never transmitted to us.
- iii
Nobody reads your messages.
Conversations never transit our systems. Our control plane handles billing and provisioning — zero access to agent data.
- iv
Hardened by default.
SSH key-only, root disabled, locked-down firewall, API keys at 600 permissions.
- v
Walk away with everything.
Cancel anytime and export your data. Your configuration is fully transparent — you can maintain the server independently.
- vi
Architecture, not a promise.
"Trust us" isn't good enough for the data an AI agent handles. We built it so we couldn't peek even if we wanted to.
- 0
- prompts used for training
- 100%
- data exportable, anytime
- 0
- messages routed through our servers
- You
- own everything. Always.
Pricing
Simple plans. No surprises.
One plan, single-tenant, with no shared infrastructure and no data limits. You bring your own Anthropic and OpenAI keys — we never touch them. Larger Enterprise offerings coming later.
Pro
$100/mo
or $95/mo billed annually — $1,140/yr
A dedicated instance plus local inference for the work that shouldn't leave your machine.
- 2 GB RAM, 50 GB disk
- Single-tenant VM, dedicated filesystem
- SSH-hardened, firewalled, auto-restart
- BYO model
- Priority email support
GPU instances for full local LLMs later this year.
Get started with ProQuestions
Answers to what you're already wondering.
Do you train on my prompts or data?
No. We do not train any models on your prompts, completions, tool calls, embeddings, or any other tenant data — ever. This is a contractual commitment in our terms, not a setting you have to find and toggle.
Do I need to know anything about servers?
No. We handle everything: provisioning, configuration, hardening, updates, and monitoring. You interact with your agent through Signal and Telegram.
Where is my data stored?
On your own server that you control, provisioned by Dorne.
Can I bring my own model API keys?
Yes. We support BYO keys for OpenAI, Anthropic, AWS Bedrock, Azure OpenAI, and self-hosted endpoints. Your credentials are stored on your server.
What if I want to leave?
Cancel and keep everything. Your data and your agent configuration are yours. Our deployment is fully transparent, so you have what you need to maintain the server independently.
How does billing work?
Plans are billed monthly per instance via Stripe. You can cancel anytime from the billing page.
Five minutes, start to finish
Set up your own personal agents today.
From "I want one" to "I have one running." Cancel anytime.