For small teams & freelancers

A governor a small team can actually run

You do not have a platform department. You have a handful of people, a few agents, and a duty not to let them wander. Olivares AI is one self-hosted artifact with unlimited users in the open build; it is not a seat product.

View the repository See the product

Small teams hire agents the same way large estates do: a Claude Code session on a laptop, an MCP server on a VPS, a scheduled job that touches the books. The difference is that nobody is spare to watch them.

The kernel is meant to be that operator: discover the agents, map what they reach, keep a stop one click away. Community is the long-term default; the commercial lines are add-on packs on top of it, not a wall in front of the core.

Honest limits

Pre-1.0. There is no contractual SLA and no 24×7 coverage.
Sponsorship is not a support contract and buys no priority; see Support the project.
Hosted models still call their provider. Self-host the model if the work must stay on the box.

Ask Claude

Questions

Is there a seat limit on the open build?

No. The AGPL core has no seat count. Commercial add-ons are additive and described on the pricing page; they do not shrink what is open.

Do I need Kubernetes?

No. A single node with the binary or a container is the intended start. Kubernetes is a topology, not a requirement.

Try it on your own infrastructure

Olivares AI is open-core (AGPL-3.0) and self-hosted. Deploy it and see what your agents can reach.

View the repository See the product