Skip to content

The fastest, cheapest, most efficient open-source agent harness. Run more than 500 agents on a $10 VPS.

2.8 KB
Raw Blame

description: "The ownership and lifecycle rules behind the Embed public API."

Concepts

Start with Runtime and agents to understand which settings are shared. Then follow the lifecycle of a request through provider selection, tools, approvals and its durable session. The scope tables distinguish runtime composition from the state a host must isolate per agent or profile.

  • Runtime: A Runtime owns one initialized core and the process-wide resources that its agents share.
  • Runtime defaults: Runtime defaults form the starting point for newly registered agents; explicit agent settings override them.
  • Agents: An Agent combines one identity, derived context and independently configured tool/model behavior on a shared Runtime.
  • Access and approvals: Access configures turn identity and permission policy; the host chooses how parked requests receive decisions.
  • Providers: A provider supplies a route or native ChatModel implementation, with agent overrides and optional ordered fallback.
  • Turns and sessions: A turn is one native model/tool loop; a session is the durable identity that makes subsequent turns a conversation.
  • Tools: Tool catalogs expose only the allowed execution surface; host tools use the same native Tool contract as built-in tools.
  • MCP: MCP servers add remote tool discovery and execution to one agent, with explicit server configuration and tool allowlists.
  • Skills: Skills are copied bundles of instructions and resources discovered inside an agent’s own home.
  • Memory: The typed memory facade binds a namespace before learning, retrieving or forgetting records.
  • Cron: Cron provides durable named jobs, explicit run-now execution and job history addressed to registered agents.
  • Channels: A channel listener binds external messages to one registered agent and keeps transport lifecycle under host control.
  • Profiles and SaaS: ProfileRuntime isolates authenticated users by workspace, credentials, conversations and event streams.
  • Completer: Completer sends stateless model requests without booting a Runtime or entering the agent tool loop.
  • Hooks and seams: Host seams replace infrastructure ports; hooks observe or control selected turn and tool boundaries.
  • Observability: Runtime metadata events track lifecycle and approvals; turn streams carry per-turn progress and the final result.
  • Testing: Deterministic local providers and explicit observations prove behavior without contacting real inference or backend services.
  • Errors: Typed errors distinguish build availability, expected domain states, host setup failures and removed agent instances.