microsoft/semantic-kernel · warning · RuntimeError

No chart generated

Error message

No chart generated

What it means

Identical logic to error 625 but in the streaming variant of the Bedrock code-interpreter sample. After consuming the streamed response and deleting the agent/thread, if no BinaryContent was captured into binary_item during the stream, the sample raises RuntimeError to indicate the model produced no downloadable chart.

Source

Thrown at python/samples/concepts/agents/bedrock_agent/bedrock_agent_with_code_interpreter_streaming.py:60

        # Invoke the agent
        print("Response: ")
        async for response in bedrock_agent.invoke_stream(
            messages=ASK,
            thread=thread,
        ):
            print(response, end="")
            thread = response.thread
            if not binary_item:
                binary_item = next((item for item in response.items if isinstance(item, BinaryContent)), None)
        print()
    finally:
        # Delete the agent
        await bedrock_agent.delete_agent()
        await thread.delete() if thread else None

    # Save the chart to a file
    if not binary_item:
        raise RuntimeError("No chart generated")

    # Securely assemble the file path and validate it's within the expected directory
    # This is a defense-in-depth measure against directory traversal attacks
    output_dir = Path(__file__).parent.resolve()
    file_path = (output_dir / binary_item.metadata["name"]).resolve()

    # Verify the resolved path is within the expected directory
    if not file_path.is_relative_to(output_dir):
        raise RuntimeError("Invalid filename: would write outside the expected directory")

    binary_item.write_to_file(file_path)
    print(f"Chart saved to {file_path}")

    # Sample output (using anthropic.claude-3-haiku-20240307-v1:0):
    # Response:
    # Here is the bar chart for the given data:
    # [A bar chart showing the following data:
    # Panda   5

View on GitHub (pinned to c028a0c7dc)

Solutions

1. Make the prompt explicitly request a chart/file so the code interpreter runs.
2. Verify the Bedrock agent's action group enables the code interpreter tool.
3. Log response.items inside the stream loop (before the finally delete) to see what the model actually returned.
4. Catch RuntimeError and degrade gracefully to printing the streamed text.

Example fix

# before
if not binary_item:
    raise RuntimeError("No chart generated")
# after
if not binary_item:
    print("No chart produced by the code interpreter.")
    return

from semantic_kernel.contents import BinaryContent

def find_binary(response) -> BinaryContent | None:
    return next((i for i in getattr(response, "items", []) if isinstance(i, BinaryContent)), None)

binary_item = find_binary(response)
if binary_item is None:
    print("No chart produced during stream; item types:",
          [type(i).__name__ for i in response.items])
    return

from semantic_kernel.contents import BinaryContent
def has_binary_content(response) -> bool:
    return any(isinstance(i, BinaryContent) for i in getattr(response, "items", []))

try:
    # ... consume stream, capture binary_item ...
    if not binary_item:
        raise RuntimeError("No chart generated")
except RuntimeError:
    # print whatever was streamed and exit gracefully
    pass

When it happens

Trigger: Streaming the Bedrock agent turn and never observing an item of type BinaryContent in response.items (model produced text only, did not run code, or the code interpreter returned no file).

Common situations: Prompt did not clearly request a chart; code-interpreter action group not enabled on the agent; the model finished the turn before emitting a file; or an upstream Bedrock streaming error produced no binary frame.

Related errors

• No chart generated
• Invalid filename: would write outside the expected directory
• Failed to handle Bedrock Agent stream event.
• Failed to handle Bedrock Agent stream event: {responseEvent}
• Invalid operator

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/341b8cc7f2deb893. Report an issue: GitHub.