Automations
Save a task with its prompt, model, environment, permissions, and budget. Run it on demand, on a schedule, or when a matching event occurs.
An automation saves the task and the settings each session needs: its prompt, model, environment, permissions, and budget. Store its YAML in your repository so the team can review changes together.
The version of the file on your default branch is the live agent. Merge a change and the next session uses it. There is no deploy step and nothing to restart.
Add your own MCP servers, scope what the agent can do on GitHub, and set spend limits per session, day, week, and month. It all sits next to the prompt, so a reviewer sees what an agent can reach and spend before it runs.
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