Create Agent
Create a new agent with the specified configuration.
Headers
Header authentication of the form Bearer <token>
Request
The name of the agent.
The blocks to create in the agent’s in-context memory.
The tools used by the agent.
The ids of the tools used by the agent.
The ids of the sources used by the agent.
The ids of the blocks used by the agent.
The tool rules governing the agent.
The system prompt used by the agent.
The type of agent.
The LLM configuration used by the agent.
The embedding configuration used by the agent.
The initial set of messages to put in the agent’s in-context memory.
If true, attaches the Letta core tools (e.g. archival_memory and core_memory related functions).
If true, attaches the Letta multi-agent tools (e.g. sending a message to another agent).
If true, attaches the Letta base tool rules (e.g. deny all tools not explicitly allowed).
The description of the agent.
The metadata of the agent.
The LLM configuration handle used by the agent, specified in the format provider/model-name, as an alternative to specifying llm_config.
The embedding configuration handle used by the agent, specified in the format provider/model-name.
The context window limit used by the agent.
The embedding chunk size used by the agent.
The maximum number of tokens to generate, including reasoning step. If not set, the model will use its default value.
The maximum number of tokens to generate for reasoning step. If not set, the model will use its default value.
Whether to enable internal extended thinking step for a reasoner model.
The template id used to configure the agent
Whether the agent is a template
The environment variables for tool execution specific to this agent.
The variables that should be set for the agent.
The id of the project the agent belongs to.
The id of the template the agent belongs to.
The base template id of the agent.
The ids of the identities associated with this agent.
If set to True, the agent will not remember previous messages (though the agent will still retain state via core memory blocks and archival/recall memory). Not recommended unless you have an advanced use case.
If set to True, memory management will move to a background agent thread.
Deprecated: Project should now be passed via the X-Project header instead of in the request body. If using the sdk, this can be done via the new x_project field below.
Response
Successful Response
The id of the agent. Assigned by the database.
The name of the agent.
The system prompt used by the agent.
The type of agent.
The LLM configuration used by the agent.
The embedding configuration used by the agent.
The in-context memory of the agent.
The tools used by the agent.
The sources used by the agent.
The id of the user that made this object.
The id of the user that made this object.
The timestamp when the object was created.
The timestamp when the object was last updated.
The list of tool rules.
The ids of the messages in the agent’s in-context memory.
The description of the agent.
The metadata of the agent.
The environment variables for tool execution specific to this agent.
The id of the project the agent belongs to.
The id of the template the agent belongs to.
The base template id of the agent.
The ids of the identities associated with this agent.
If set to True, the agent will not remember previous messages (though the agent will still retain state via core memory blocks and archival/recall memory). Not recommended unless you have an advanced use case.
If set to True, memory management will move to a background agent thread.
The multi-agent group that this agent manages