Ona (formerly Gitpod's agentic platform, now part of OpenAI) is an enterprise-grade cloud agent orchestration platform that manages and runs background AI coding agents. Unlike local terminal-only tools that execute code on developer workstations, Ona provisions isolated cloud environments with pre-configured developer toolchains, network controls, and scoped credentials to execute coding tasks asynchronously. Day-to-day, engineering teams dispatch tasks to Ona via webhooks, issue trackers, CI events, or automated schedules. Ona spins up governed cloud sandboxes to run agents like Ona or OpenAI Codex against repositories, execute builds, run tests, and ultimately produce pull requests. Developers interface with the platform via web management consoles, integrations with tools like Linear and Sentry, and PR reviews rather than relying exclusively on a local interactive prompt. Ona differentiates itself from standard CLI agents by prioritizing runtime security, centralized fleet management, and cloud isolation over localhost execution. It enforces kernel-level policy governance, complete VPC network controls, and audit trails to prevent agents from bypassing sandboxes, accessing unauthorized networks, or exposing sensitive credentials. Ona is designed for enterprise organizations with SOC 2 compliance requirements. The platform provides managed cloud infrastructure and VPC deployment options, accessible via demo requests or team sign-ups.
Tags: agent harness, ai agent, cli, code assistant, coding agent, developer tools