Omnara

Omnara provides an open-source API for running durable AI agents in production. It supports running agents on any model and any machine, with tools, secrets, permissions, and streaming built in. Users can self-host under Apache 2.0 or opt into Omnara Cloud, enabling flexible deployment across environments while preserving ownership of data and configurations.

The platform supports building agents for internal use cases or external use cases such as Slack bots, code review assistants, and product-dashboard integrations. Access to Omnara is available through a web interface, a CLI, and a REST API, enabling multiple ways to create and manage agents.

The system emphasizes running durable conversations, including state, tool calls, and history, and adapts to various collaboration and deployment workflows.

The documentation outlines how to start with a quickstart guide, explains core concepts like organizations, projects, grants, and agents, and provides guidance to launch agents via dashboard, API, or CLI. It also highlights self-hosting as a free option and the availability of managed cloud services.

Use cases

• Internal Slack bots and dashboards with embedded agents