microsoft/autogen · error · InvalidOperationException

Invalid ChatResponseMessage

Error message

Invalid ChatResponseMessage

What it means

Thrown by PostProcessChatResponseMessage when a ChatResponseMessage carries no function call, no tool calls, and null/empty text content. The connector has nothing to map into a TextMessage or ToolCallMessage, so it treats the response as malformed. Typical causes are content filtering (empty completion), reasoning models that return content in a different field, or SDK version drift where content moved.

Source

Thrown at dotnet/src/AutoGen.OpenAI.V1/Middleware/OpenAIChatRequestMessageConnector.cs:167

        {
            var functionToolCalls = chatResponseMessage.ToolCalls
                .Where(tc => tc is ChatCompletionsFunctionToolCall)
                .Select(tc => (ChatCompletionsFunctionToolCall)tc);

            var toolCalls = functionToolCalls.Select(tc => new ToolCall(tc.Name, tc.Arguments) { ToolCallId = tc.Id });

            return new ToolCallMessage(toolCalls, from)
            {
                Content = textContent,
            };
        }

        if (textContent is string content && !string.IsNullOrEmpty(content))
        {
            return new TextMessage(Role.Assistant, content, from);
        }

        throw new InvalidOperationException("Invalid ChatResponseMessage");
    }

    public IEnumerable<IMessage> ProcessIncomingMessages(IAgent agent, IEnumerable<IMessage> messages)
    {
        return messages.SelectMany<IMessage, IMessage>(m =>
        {
            if (m is IMessage<ChatRequestMessage> crm)
            {
                return [crm];
            }
            else
            {
                var chatRequestMessages = m switch
                {
                    TextMessage textMessage => ProcessTextMessage(agent, textMessage),
                    ImageMessage imageMessage when (imageMessage.From is null || imageMessage.From != agent.Name) => ProcessImageMessage(agent, imageMessage),
                    MultiModalMessage multiModalMessage when (multiModalMessage.From is null || multiModalMessage.From != agent.Name) => ProcessMultiModalMessage(agent, multiModalMessage),
                    ToolCallMessage toolCallMessage when (toolCallMessage.From is null || toolCallMessage.From == agent.Name) => ProcessToolCallMessage(agent, toolCallMessage),

View on GitHub (pinned to 027ecf0a37)

Solutions

1. Inspect message.Content.Choices[0].FinishReason to see why content is empty (content_filter, length, etc.) and address that first
2. Increase TokenChoiceCount/max tokens if truncation produced an empty completion
3. Verify the Azure.OpenAI package version matches the one AutoGen.OpenAI.V1 was compiled against
4. Catch InvalidOperationException and retry or substitute a default assistant message

Example fix

// before
var reply = await agent.SendAsync(prompt); // throws 'Invalid ChatResponseMessage'

// after
// check finish reason on the raw completion before the connector maps it
if (completion.Choices[0].FinishReason == CompletionsFinishReason.ContentFiltered)
{
    // handle filtered response explicitly
}

var choice = completion.Choices[0];
bool hasContent = !string.IsNullOrEmpty(choice.Message.Content)
    || choice.Message.FunctionCall is not null
    || choice.Message.ToolCalls.Count > 0;
if (!hasContent) { /* inspect FinishReason before the connector throws */ }

catch (InvalidOperationException ex) when (ex.Message == "Invalid ChatResponseMessage") { /* retry or log finish reason */ }

When it happens

Trigger: A ChatCompletions response whose Message has empty Content and no FunctionCall/ToolCalls — e.g. finish_reason content_filter with empty body, or an assistant reply that only contains tool-call outputs not exposed on this SDK version.

Common situations: Deployments where content filter removes the text; models that return empty content on truncation (max_tokens hit); mismatch between the Azure.OpenAI SDK version the connector was built against and the API's actual response shape.

Related errors

• Invalid return message type {message.GetType().Name}
• Invalid message type: {m.GetType().Name}
• Please set AZURE_OPENAI_API_KEY environment variable.
• Please set AZURE_OPENAI_ENDPOINT environment variable.
• Please set environment variable AZURE_OPENAI_API_KEY

AI-assisted analysis of microsoft/autogen@027ecf0a37 (2026-08-15). Data as JSON: /api/errors/46e70e5cd3e521ba. Report an issue: GitHub.