Agents
Agents are headless execution aggregates owned by one Organization. Configure an Agent’s Runtime and versioned behavior, manage its lifecycle and access, and add zero or more Communication Connections independently.
- For
- Agent operators and administrators
Recommended
Agent Lifecycle and Configuration
Create a headless Agent, choose its Runtime and configuration, manage lifecycle transitions and overrides safely, and add independent Communication Connections.
GuideHeadless Agents and Communication Connections
For an initial conversation without external provider setup, use the experimental Chat tab once the Agent is Running and Working. See Chat with an Agent in the dashboard for permissions and thread behavior. External Connections can be configured separately.
Every new Agent begins STOPPED and without a Communication Platform. Communication Connections are optional Agent-subordinate resources with their own settings, encrypted credentials, access policies, health, diagnostics, and lifecycle.
Runtime and Communication Platform are independent. An Agent can remain headless, own multiple Connections, or own several Connections for the same platform. Adding, editing, enabling, disabling, or retiring a Connection does not require an Agent restart.
Available documentation
Agent Lifecycle and Configuration
Create a headless Agent, choose its Runtime and configuration, manage lifecycle transitions and overrides safely, and add independent Communication Connections.
Communication Connections
Connect an Agent to Slack, Microsoft Teams, Telegram, or Discord, then manage each connection separately.
Choose an Agent runtime
Compare Hermes and OpenClaw execution, approval, and runtime transport independently of Communication Connections.
Configure an Agent
Configure an Agent's model, Hermes approval and progress settings, and pinned behavior.
Manage the Agent lifecycle
Start, pause, restart, and retire an Agent. Learn which changes need a restart and how to handle runtime failures separately from chat connection problems.
Control Communication Connection Access
Configure Connection-scoped channel, group, server, direct-message, user, role, and mention policies through each Platform Plugin’s supported settings.
Review Agent health and logs
Interpret Agent lifecycle and Runtime health, inspect live and retained Runtime logs, and determine when a failure belongs to a Communication Connection instead.
Share access to an Agent
Grant Agent Access for an Agent and its subordinate Connections, Conversations, Tool Calls, Secrets, private Skills, configuration, logs, and costs.
Chat with an Agent in the dashboard
Use the experimental dashboard Chat tab, manage user-scoped threads, and understand Agent access and stop-generation behavior.
Agent Webhooks
Trigger Agent runs over authenticated HTTP with signed request verification, idempotency controls, and native platform delivery.
Agent Restore Points
Capture, inspect, and restore point-in-time snapshots of an Agent workspace, configuration pins, and state using durable archive jobs.