Agent runtime
Internal agents use Auto, Fast, or Deep modes. Deep runs can execute for up to ~2 hours per task with recovery context if interrupted.
Configure agents in the Console, connect integrations and knowledge, then deploy to the web, messaging channels, or your own app via API.
Persistent sandboxes · Fast & Deep execution modes · 20+ native integrations · Real-time voice · Event-driven flows
Agent runtime
Internal agents use Auto, Fast, or Deep modes. Deep runs can execute for up to ~2 hours per task with recovery context if interrupted.
Persistent sandboxes
Isolated containers with browser, terminal, and filesystem — one session per agent, reused across runs.
Multi-modal content
Generate images and videos (plan-dependent), plus presentations, diagrams, and documents.
Event-driven flows
Trigger agents from GitHub, Slack, inbound email, Telegram, WhatsApp, webhooks, and cron schedules.
Skills
Reusable workflow templates you attach to agents — invoked by the agent when relevant.
Built-in capabilities
Sandbox tools, browser automation, web search, knowledge retrieval, and OAuth integrations from the catalog.
Define your agent
Describe what your agent should do and how it should behave.
Connect tools and data
Link GitHub, Slack, knowledge collections, and custom APIs.
Deploy anywhere
Web widget, Telegram, WhatsApp, Slack (internal agents), or REST API.
Monitor and improve
Track performance, gather feedback, and continuously optimize.
Quickstart (10 min)
Create and run your first AI agent in under 10 minutes. Start now →
Explore use cases
See real-world examples and templates for common scenarios. Browse →
Read the docs
Comprehensive guides for agents, tools, and deployment. Learn →
Join the community
Get help, share ideas, and connect with other builders. Join Discord →
Read these three pages in order: