SharedTreeChatModel Interface

A plugin interface that handles tree-agent model invocations.

This API is provided as an alpha preview and may change without notice.

To use, import via @fluidframework/tree-agent/alpha.

For more information about our API support guarantees, see here.

Signature

export interface SharedTreeChatModel

Remarks

This wraps the underlying communication with an LLM. See @fluidframework/tree-agent-langchain for a drop-in implementation based on the LangChain library.

Properties

Property	Alerts	Modifiers	Type	Description
editToolName	Alpha	optional	string	The name of the tool that the model should use to edit the tree.
name	Alpha	optional	string	A optional name of this chat model.

Methods

Method	Alerts	Return Type	Description
invoke(history)	Alpha	Promise<TreeAgentChatResponse>	Invoke the model with the given message history and return a single response.

Property Details

editToolName

The name of the tool that the model should use to edit the tree.

This API is provided as an alpha preview and may change without notice.

For more information about our API support guarantees, see here.

Signature

editToolName?: string;

Type: string

Remarks

If supplied, this is included in the generated system prompt so that the underlying LLM can request edits. The model should "implement" the tool by registering it with the underlying LLM API. The tool should accept an LLM-generated JavaScript function as input. If not supplied, the model cannot be used for semantic editing.

name

A optional name of this chat model.

This API is provided as an alpha preview and may change without notice.

For more information about our API support guarantees, see here.

Signature

name?: string;

Type: string

Remarks

If supplied, this may be used in logging or debugging information.

Example

"gpt-5"

Method Details

invoke

Invoke the model with the given message history and return a single response.

This API is provided as an alpha preview and may change without notice.

For more information about our API support guarantees, see here.

Signature

invoke(history: readonly TreeAgentChatMessage[]): Promise<TreeAgentChatResponse>;

Remarks

This is the stateless API. The model should not maintain any internal message history; all context is provided via the history parameter. Implementations should convert the TreeAgentChatMessage array to the underlying LLM's message format, invoke the LLM, and return either a TreeAgentToolCallMessage (if the model wants to call a tool) or a TreeAgentAssistantMessage (if the model is done).

Parameters

Parameter	Type	Description
history	readonly TreeAgentChatMessage[]

Returns

Return type: Promise<TreeAgentChatResponse>