Operator is the agent on the other side of every conversation at operator.io/chat. You describe what you want in plain language. It spins up instances, writes skills, manages secrets, schedules automations, and reports results, all without leaving the chat window.Documentation Index
Fetch the complete documentation index at: https://docs.operator.io/llms.txt
Use this file to discover all available pages before exploring further.
How it works
You type a goal. Operator decides the steps, picks the tools, and executes them. It has access to every dashboard section: creating instances, browsing files, granting secrets, reading activity logs, writing skills, setting up automations, and sending messages into the agents running on those instances.Conversations persist
Each chat keeps its full history. Operator remembers which instance it created for which project, what it’s already produced, what’s in flight, and what failed last time. Come back a week later and pick up where you left off:What Operator can do
| Category | Actions |
|---|---|
| Instances | Create, delete, restart, read logs, browse files |
| Skills | Write, install, update, preview before deploying |
| Secrets | Create, grant, revoke, scope per instance |
| Config | Read and update any instance setting |
| Automations | Schedule cron jobs on an instance |
| Channels | Wire Telegram or Discord to a specific agent |
| Fleet | Operate across multiple instances in a single turn |
Mentioning files and agents
Type@ in the chat to mention a specific file or agent. The popover searches across every active instance and inserts a visible chip into the message. Operator routes the work to whichever instance owns the mention:
Scheduling vs triggering
Two ways to run Operator without being at the keyboard:Automations
Cron-based jobs that fire on a schedule. The agent inside the target instance runs the prompt.
Channel messages
Send a message to a connected Telegram or Discord bot and the agent responds.
Related
Quickstart
Deploy your first agent from chat.
Agents
Build, test, evolve lifecycle for a single agent.
Skills
How Operator captures and reuses approaches across chats.