Give an agent a task, an environment, permissions, and a budget. Its work runs in the cloud, with a conversation you can continue and results your team can inspect.
Choose a one-shot task that closes after its first turn, or an interactive session that accepts follow-up messages. Each session runs with its own environment, permissions, and budget. Closing your browser does not stop the work.
Idle conversations pause their compute. Send another message to resume the same conversation and workspace, including uncommitted changes. Follow-up turns share the session’s budget and history.
Read messages and tool results, inspect diffs, find commits and pull requests, and see what the session cost. Download the transcript or retrieve validated structured output when the task declares an output schema.
Save tasks, run on demand or with triggers
Run tasks and continue conversations
Reusable repositories, toolchains, and compute
Configure responders for mentions and alerts
Store credentials and select where they are used
Signed session notifications and delivery history
Frontier and open models, bring your own key
Transcripts, live streams, search, cost
Scoped credentials, SOC 2, audit trail
Hard caps per session, agent, and developer