Method run_debug

src/google/adk/runners.py:1720–1827

Debug helper for quick agent experimentation and testing. This convenience method is designed for developers getting started with ADK who want to quickly test agents without dealing with session management, content formatting, or event streaming. It automatically handles common boil

(
      self,
      user_messages: str | list[str],
      *,
      user_id: str = 'debug_user_id',
      session_id: str = 'debug_session_id',
      run_config: RunConfig | None = None,
      quiet: bool = False,
      verbose: bool = False,
  )

    return True

  async def run_debug(
      self,
      user_messages: str | list[str],
      *,
      user_id: str = 'debug_user_id',
      session_id: str = 'debug_session_id',
      run_config: RunConfig | None = None,
      quiet: bool = False,
      verbose: bool = False,
  ) -> list[Event]:
    """Debug helper for quick agent experimentation and testing.

    This convenience method is designed for developers getting started with ADK
    who want to quickly test agents without dealing with session management,
    content formatting, or event streaming. It automatically handles common
    boilerplate while hiding complexity.

    IMPORTANT: This is for debugging and experimentation only. For production
    use, please use the standard run_async() method which provides full control
    over session management, event streaming, and error handling.

    Args:
        user_messages: Message(s) to send to the agent. Can be: - Single string:
          "What is 2+2?" - List of strings: ["Hello!", "What's my name?"]
        user_id: User identifier. Defaults to "debug_user_id".
        session_id: Session identifier for conversation persistence. Defaults to
          "debug_session_id". Reuse the same ID to continue a conversation.
        run_config: Optional configuration for the agent execution.
        quiet: If True, suppresses console output. Defaults to False (output
          shown).
        verbose: If True, shows detailed tool calls and responses. Defaults to
          False for cleaner output showing only final agent responses.

    Returns:
        list[Event]: All events from all messages.

    Raises:
        ValueError: If session creation/retrieval fails.

    Examples:
        Quick debugging:
        >>> runner = InMemoryRunner(agent=my_agent)
        >>> await runner.run_debug("What is 2+2?")

        Multiple queries in conversation:
        >>> await runner.run_debug(["Hello!", "What's my name?"])

        Continue a debug session:
        >>> await runner.run_debug("What did we discuss?")  # Continues default
        session

        Separate debug sessions:
        >>> await runner.run_debug("Hi", user_id="alice", session_id="debug1")
        >>> await runner.run_debug("Hi", user_id="bob", session_id="debug2")

        Capture events for inspection:
        >>> events = await runner.run_debug("Analyze this")
        >>> for event in events: