Things you'd want to know

A raw LLM is a single conversation that forgets everything when you close the tab. You start from scratch every time. Onnebo agents are persistent. They have a defined role, memory of your preferences, access to the tools you've granted, and they improve based on your feedback. A ChatGPT conversation is calling a taxi. An Onnebo agent is hiring a driver who learns your routes.

Yes. You define per action what an agent can do autonomously versus what needs your approval. You control which data sources it can read, which external tools it can interact with, and what kinds of outputs it can produce. Nothing is open by default. Agents start with minimal permissions and you expand them as trust builds.

You set approval thresholds per action type. If your agent drafts an email, you can require it to show you the draft and wait for your sign-off before sending. If it's doing background research, you can let it run autonomously. The threshold is yours to set per action, not a global on/off switch.

They can't access systems you haven't connected. They can't take actions you haven't permitted. They can't bypass your approval thresholds. And they don't operate outside the capabilities of the underlying LLMs. They're not magic. They're orchestrated intelligence with guardrails you define.

You can bring your own keys from OpenAI, Anthropic, or other supported providers for full cost transparency. Token costs are billed directly by your provider, and Onnebo adds nothing on top. If you'd rather not manage keys, you can use Onnebo-provisioned tokens at the same base rates as the underlying providers.

You pay a monthly subscription based on how many agents you want to build. Starter is $5/month for up to 5 agents, Pro is $49/month for up to 15, and Business is $180/month for unlimited. Token costs for the LLMs your agents use are separate, either through your own API keys or Onnebo-provisioned access. During beta, all plans are free.

Your agents and their configurations are preserved. If you downgrade to a tier with fewer agents than you've built, you choose which ones to keep active. Inactive agents are saved but paused, not deleted. You can reactivate them if you upgrade again.

You need to be able to describe what you want in plain language. That's it. If you can explain a task to a new hire, you can build an agent on Onnebo. There is no code, no flowchart editor, and no configuration files. For users who want deeper control, every setting is available, but nothing requires it.

Onnebo does not train on your data or share it across accounts. Your agent configurations, conversation history, and connected data sources are private to your account. We are pursuing SOC 2 certification and will publish our security practices in detail before general availability.

We haven't set a fixed date. Beta users will get advance notice before any transition to paid plans, and we'll honor early-access pricing for users who join during beta.