Disabling state streaming - Copilotkit

What is this?

By default, CopilotKit will stream both your state and tool calls to the frontend. You can disable this by using CopilotKit's custom RunnableConfig.

When should I use this?

Occasionally, you'll want to disable streaming temporarily — for example, the LLM may be doing something the current user should not see, like emitting tool calls or questions pertaining to other employees in an HR system.

Implementation

Disable all streaming

You can disable all message streaming and tool call streaming by passing emit_messages=False and emit_tool_calls=False to the CopilotKit config.

<Tabs groupId="language_langgraph_agent" items={['Python', 'TypeScript']} default="Python" persist> <Tab value="Python"> ```python from copilotkit.langgraph import copilotkit_customize_config

    async def frontend_actions_node(state: AgentState, config: RunnableConfig):

        # 1) Configure CopilotKit not to emit messages
        modifiedConfig = copilotkit_customize_config(
            config,
            emit_messages=False, # if you want to disable message streaming # [!code highlight]
            emit_tool_calls=False # if you want to disable tool call streaming # [!code highlight]
        )

        # 2) Provide the actions to the LLM
        model = ChatOpenAI(model="gpt-5.4").bind_tools([
          *state["copilotkit"]["actions"],
          # ... any tools you want to make available to the model
        ])

        # 3) Call the model with CopilotKit's modified config  # [!code highlight]
        response = await model.ainvoke(state["messages"], modifiedConfig) # [!code highlight]

        # don't return the new response to hide it from the user
        return state
    ```

<Callout type="warn" title="BEWARE!">
    In LangGraph Python, the `config` variable in the surrounding namespace is **implicitly** passed into LangChain LLM calls, even when not explicitly provided.

    This is why we create a new variable `modifiedConfig` rather than modifying `config` directly. If we modified `config` itself, it would change the default configuration for all subsequent LLM calls in that namespace.

    ```python
    # if we override the config variable name with a new value
    config = copilotkit_customize_config(config, ...)

    # it will affect every subsequent LangChain LLM call in the same namespace, even when `config` is not explicitly provided
    response = await model2.ainvoke(*state["messages"]) # implicitly uses the modified config!
    ```
</Callout>
</Tab>
<Tab value="TypeScript">
    ```typescript

    async function frontendActionsNode(state: AgentState, config: RunnableConfig): Promise<AgentState> {
        // 1) Configure CopilotKit not to emit messages
        const modifiedConfig = copilotkitCustomizeConfig(config, {
            emitMessages: false, // if you want to disable message streaming
            emitToolCalls: false, // if you want to disable tool call streaming
        });

        // 2) Provide the actions to the LLM
        const model = new ChatOpenAI({ temperature: 0, model: "gpt-5.4" });
        const modelWithTools = model.bindTools!([
            ...convertActionsToDynamicStructuredTools(state.copilotkit?.actions || []),
            ...tools,
        ]);

        // 3) Call the model with CopilotKit's modified config
        const response = await modelWithTools.invoke(state.messages, modifiedConfig);

        // don't return the new response to hide it from the user
        return state;
    }
    ```
</Tab>

</Tabs>