AI Agents
AI agents in RealTimeX are workspace-aware assistants that can operate across chat, skills, flows, MCP integrations, browser sessions, and runtime tooling.
They are no longer just a thin wrapper around a few fixed built-in tools. In the current product, agent capability comes from a combination of workspace configuration, plugins, and runtime controls.
Interactive chat agents, Ambient Agent, Agentic CLIs, Browser Tool, and plugin-backed skills are related but separate parts of the system. This section focuses on the interactive @agent experience first.
Start here
Configure workspace skills, flows, MCP servers, and related runtime prerequisites.
Start @agent sessions, continue them correctly, and understand how to end them.
Manage CLIs, working directories, Ambient Agent, diagnostics, and agent feed.
What matters most now
- Agent sessions start from chat with
@agent. - Workspace capability is configured in
Workspace Settings > Agent. - Operational power comes from Agent Runtime.
- Additional behavior often comes from Plugins.