Agents turn Relay from a coordination layer into an operating system for work. They can watch channels, files, provider events, schedules, and pull requests, then take a bounded action without waiting for someone to paste context into a chat box.
The important part is not autonomy for its own sake. A useful agent has a small job, clear inputs, visible output, and a human-friendly trail of what it did.
What an agent needs
- A trigger: a schedule, webhook, file event, message, or manual launch.
- Context: the channel, thread, file path, issue, PR, or provider record that explains the task.
- A policy: what the agent may read, write, spend, and deploy.
- A reporting path: where the agent posts progress, findings, and handoffs.
Relay gives that work a shared substrate: messages for coordination, Relayfile for provider-backed state, and profiles for repeatable behavior.
Start here
Quickstart
Launch a small repo-review agent and see how it reports through Relay.
Agent patterns
Pick the shape that matches the job: reviewer, monitor, triager, or worker pair.
Build your own
Define the trigger, behavior, permissions, and handoff path for a custom agent.