| Name | Type | Description |
|---|---|---|
assistant_id* | str | Agent identifier. |
model_name | str | None | Default: NoneModel spec string. |
model_params | dict[str, Any] | None | Default: None |
auto_approve | bool | Default: False |
interrupt_shell_only | bool | Default: False |
shell_allow_list | list[str] | None | Default: None |
sandbox_type | str | Default: 'none' |
sandbox_id | str | None | Default: None |
sandbox_setup | str | None | Default: None |
enable_shell | bool | Default: True |
enable_ask_user | bool | Default: False |
mcp_config_path | str | None | Default: None |
no_mcp | bool | Default: False |
trust_project_mcp | bool | None | Default: None |
interactive | bool | Default: True |
host | str | Default: '127.0.0.1' |
port | int | Default: 2024 |
Start a LangGraph server and return a connected remote agent client.
Extra model kwargs.
Auto-approve all tools.
Validate shell commands via middleware instead of HITL.
Restrictive shell allow-list for ShellAllowListMiddleware.
Sandbox type.
Existing sandbox ID to reuse.
Path to setup script for the sandbox.
Enable shell execution tools.
Enable ask_user tool.
Path to MCP config.
Disable MCP.
Trust project MCP servers.
Whether the agent is interactive.
Server host.
Server port.