agno 1.8.1__py3-none-any.whl → 2.0.0__py3-none-any.whl

agno/utils/print_response/agent.py

@@ -0,0 +1,779 @@
+import json
+from collections.abc import Set
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Sequence, Union, cast, get_args
+
+from pydantic import BaseModel
+from rich.console import Group
+from rich.json import JSON
+from rich.live import Live
+from rich.markdown import Markdown
+from rich.status import Status
+from rich.text import Text
+
+from agno.media import Audio, File, Image, Video
+from agno.models.message import Message
+from agno.reasoning.step import ReasoningStep
+from agno.run.agent import RunEvent, RunOutput, RunOutputEvent, RunPausedEvent
+from agno.utils.log import log_warning
+from agno.utils.message import get_text_from_message
+from agno.utils.response import create_panel, create_paused_run_output_panel, escape_markdown_tags, format_tool_calls
+from agno.utils.timer import Timer
+
+if TYPE_CHECKING:
+    from agno.agent.agent import Agent
+
+
+def print_response_stream(
+    agent: "Agent",
+    input: Union[List, Dict, str, Message, BaseModel, List[Message]],
+    session_id: Optional[str] = None,
+    session_state: Optional[Dict[str, Any]] = None,
+    user_id: Optional[str] = None,
+    audio: Optional[Sequence[Audio]] = None,
+    images: Optional[Sequence[Image]] = None,
+    videos: Optional[Sequence[Video]] = None,
+    files: Optional[Sequence[File]] = None,
+    stream_intermediate_steps: bool = False,
+    knowledge_filters: Optional[Dict[str, Any]] = None,
+    debug_mode: Optional[bool] = None,
+    markdown: bool = False,
+    show_message: bool = True,
+    show_reasoning: bool = True,
+    show_full_reasoning: bool = False,
+    tags_to_include_in_markdown: Set[str] = {"think", "thinking"},
+    console: Optional[Any] = None,
+    add_history_to_context: Optional[bool] = None,
+    dependencies: Optional[Dict[str, Any]] = None,
+    metadata: Optional[Dict[str, Any]] = None,
+    **kwargs: Any,
+):
+    _response_content: str = ""
+    _response_reasoning_content: str = ""
+    response_content_batch: Union[str, JSON, Markdown] = ""
+    reasoning_steps: List[ReasoningStep] = []
+    accumulated_tool_calls: List = []
+
+    with Live(console=console) as live_log:
+        status = Status("Thinking...", spinner="aesthetic", speed=0.4, refresh_per_second=10)
+        live_log.update(status)
+        response_timer = Timer()
+        response_timer.start()
+        # Flag which indicates if the panels should be rendered
+        render = False
+        # Panels to be rendered
+        panels = [status]
+        # First render the message panel if the message is not None
+        if input and show_message:
+            render = True
+            # Convert message to a panel
+            message_content = get_text_from_message(input)
+            message_panel = create_panel(
+                content=Text(message_content, style="green"),
+                title="Message",
+                border_style="cyan",
+            )
+            panels.append(message_panel)
+        if render:
+            live_log.update(Group(*panels))
+
+        for response_event in agent.run(
+            input=input,
+            session_id=session_id,
+            session_state=session_state,
+            user_id=user_id,
+            audio=audio,
+            images=images,
+            videos=videos,
+            files=files,
+            stream=True,
+            stream_intermediate_steps=stream_intermediate_steps,
+            knowledge_filters=knowledge_filters,
+            debug_mode=debug_mode,
+            add_history_to_context=add_history_to_context,
+            dependencies=dependencies,
+            metadata=metadata,
+            **kwargs,
+        ):
+            if isinstance(response_event, tuple(get_args(RunOutputEvent))):
+                if response_event.is_paused:  # type: ignore
+                    response_event = cast(RunPausedEvent, response_event)  # type: ignore
+                    response_panel = create_paused_run_output_panel(response_event)  # type: ignore
+                    panels.append(response_panel)
+                    live_log.update(Group(*panels))
+                    return
+
+                if (
+                    response_event.event == RunEvent.tool_call_started  # type: ignore
+                    and hasattr(response_event, "tool")
+                    and response_event.tool is not None
+                ):
+                    accumulated_tool_calls.append(response_event.tool)
+
+                if response_event.event == RunEvent.run_content:  # type: ignore
+                    if hasattr(response_event, "content"):
+                        if isinstance(response_event.content, str):
+                            _response_content += response_event.content
+                        elif agent.output_schema is not None and isinstance(response_event.content, BaseModel):
+                            try:
+                                response_content_batch = JSON(  # type: ignore
+                                    response_event.content.model_dump_json(exclude_none=True), indent=2
+                                )
+                            except Exception as e:
+                                log_warning(f"Failed to convert response to JSON: {e}")
+                        else:
+                            try:
+                                response_content_batch = JSON(json.dumps(response_event.content), indent=4)
+                            except Exception as e:
+                                log_warning(f"Failed to convert response to JSON: {e}")
+                    if hasattr(response_event, "reasoning_content") and response_event.reasoning_content is not None:  # type: ignore
+                        _response_reasoning_content += response_event.reasoning_content  # type: ignore
+                if hasattr(response_event, "reasoning_steps") and response_event.reasoning_steps is not None:  # type: ignore
+                    reasoning_steps = response_event.reasoning_steps  # type: ignore
+
+            # Escape special tags before markdown conversion
+            if markdown:
+                escaped_content = escape_markdown_tags(_response_content, tags_to_include_in_markdown)  # type: ignore
+                response_content_batch = Markdown(escaped_content)
+
+            response_content_stream: str = _response_content
+
+            # Check if we have any response content to display
+            if response_content_stream and not markdown:
+                response_content = response_content_stream
+            else:
+                response_content = response_content_batch  # type: ignore
+
+            # Sanitize empty Markdown content
+            if isinstance(response_content, Markdown):
+                if not (response_content.markup and response_content.markup.strip()):
+                    response_content = None  # type: ignore
+
+            panels = [status]
+            if show_message:
+                # Convert message to a panel
+                message_content = get_text_from_message(input)
+                message_panel = create_panel(
+                    content=Text(message_content, style="green"),
+                    title="Message",
+                    border_style="cyan",
+                )
+                panels.append(message_panel)
+
+            additional_panels = build_panels_stream(
+                response_content=response_content,
+                response_event=response_event,  # type: ignore
+                response_timer=response_timer,
+                response_reasoning_content_buffer=_response_reasoning_content,
+                reasoning_steps=reasoning_steps,
+                show_reasoning=show_reasoning,
+                show_full_reasoning=show_full_reasoning,
+                accumulated_tool_calls=accumulated_tool_calls,
+            )
+            panels.extend(additional_panels)
+            if panels:
+                live_log.update(Group(*panels))
+
+        if agent.memory_manager is not None and agent.memory_manager.memories_updated:
+            memory_panel = create_panel(
+                content=Text("Memories updated"),
+                title="Memories",
+                border_style="green",
+            )
+            panels.append(memory_panel)
+            live_log.update(Group(*panels))
+            agent.memory_manager.memories_updated = False
+
+        if agent.session_summary_manager is not None and agent.session_summary_manager.summaries_updated:
+            summary_panel = create_panel(
+                content=Text("Session summary updated"),
+                title="Session Summary",
+                border_style="green",
+            )
+            panels.append(summary_panel)
+            live_log.update(Group(*panels))
+            agent.session_summary_manager.summaries_updated = False
+
+        response_timer.stop()
+
+        # Final update to remove the "Thinking..." status
+        panels = [p for p in panels if not isinstance(p, Status)]
+        live_log.update(Group(*panels))
+
+
+async def aprint_response_stream(
+    agent: "Agent",
+    input: Union[List, Dict, str, Message, BaseModel, List[Message]],
+    session_id: Optional[str] = None,
+    session_state: Optional[Dict[str, Any]] = None,
+    user_id: Optional[str] = None,
+    audio: Optional[Sequence[Audio]] = None,
+    images: Optional[Sequence[Image]] = None,
+    videos: Optional[Sequence[Video]] = None,
+    files: Optional[Sequence[File]] = None,
+    stream_intermediate_steps: bool = False,
+    knowledge_filters: Optional[Dict[str, Any]] = None,
+    debug_mode: Optional[bool] = None,
+    markdown: bool = False,
+    show_message: bool = True,
+    show_reasoning: bool = True,
+    show_full_reasoning: bool = False,
+    tags_to_include_in_markdown: Set[str] = {"think", "thinking"},
+    console: Optional[Any] = None,
+    add_history_to_context: Optional[bool] = None,
+    dependencies: Optional[Dict[str, Any]] = None,
+    metadata: Optional[Dict[str, Any]] = None,
+    **kwargs: Any,
+):
+    _response_content: str = ""
+    _response_reasoning_content: str = ""
+    reasoning_steps: List[ReasoningStep] = []
+    response_content_batch: Union[str, JSON, Markdown] = ""
+    accumulated_tool_calls: List = []
+
+    with Live(console=console) as live_log:
+        status = Status("Thinking...", spinner="aesthetic", speed=0.4, refresh_per_second=10)
+        live_log.update(status)
+        response_timer = Timer()
+        response_timer.start()
+        # Flag which indicates if the panels should be rendered
+        render = False
+        # Panels to be rendered
+        panels = [status]
+        # First render the message panel if the message is not None
+        if input and show_message:
+            render = True
+            # Convert message to a panel
+            message_content = get_text_from_message(input)
+            message_panel = create_panel(
+                content=Text(message_content, style="green"),
+                title="Message",
+                border_style="cyan",
+            )
+            panels.append(message_panel)
+        if render:
+            live_log.update(Group(*panels))
+
+        result = agent.arun(
+            input=input,
+            session_id=session_id,
+            session_state=session_state,
+            user_id=user_id,
+            audio=audio,
+            images=images,
+            videos=videos,
+            files=files,
+            stream=True,
+            stream_intermediate_steps=stream_intermediate_steps,
+            knowledge_filters=knowledge_filters,
+            debug_mode=debug_mode,
+            add_history_to_context=add_history_to_context,
+            dependencies=dependencies,
+            metadata=metadata,
+            **kwargs,
+        )
+
+        async for resp in result:  # type: ignore
+            if isinstance(resp, tuple(get_args(RunOutputEvent))):
+                if resp.is_paused:
+                    response_panel = create_paused_run_output_panel(resp)  # type: ignore
+                    panels.append(response_panel)
+                    live_log.update(Group(*panels))
+                    break
+
+                if (
+                    resp.event == RunEvent.tool_call_started  # type: ignore
+                    and hasattr(resp, "tool")
+                    and resp.tool is not None
+                ):
+                    accumulated_tool_calls.append(resp.tool)
+
+                if resp.event == RunEvent.run_content:  # type: ignore
+                    if isinstance(resp.content, str):
+                        _response_content += resp.content
+                    elif agent.output_schema is not None and isinstance(resp.content, BaseModel):
+                        try:
+                            response_content_batch = JSON(resp.content.model_dump_json(exclude_none=True), indent=2)  # type: ignore
+                        except Exception as e:
+                            log_warning(f"Failed to convert response to JSON: {e}")
+                    else:
+                        try:
+                            response_content_batch = JSON(json.dumps(resp.content), indent=4)
+                        except Exception as e:
+                            log_warning(f"Failed to convert response to JSON: {e}")
+                    if resp.reasoning_content is not None:  # type: ignore
+                        _response_reasoning_content += resp.reasoning_content  # type: ignore
+
+                if hasattr(resp, "reasoning_steps") and resp.reasoning_steps is not None:  # type: ignore
+                    reasoning_steps = resp.reasoning_steps  # type: ignore
+
+            response_content_stream: str = _response_content
+
+            # Escape special tags before markdown conversion
+            if markdown:
+                escaped_content = escape_markdown_tags(_response_content, tags_to_include_in_markdown)  # type: ignore
+                response_content_batch = Markdown(escaped_content)
+
+            # Check if we have any response content to display
+            if response_content_stream and not markdown:
+                response_content = response_content_stream
+            else:
+                response_content = response_content_batch  # type: ignore
+
+            # Sanitize empty Markdown content
+            if isinstance(response_content, Markdown):
+                if not (response_content.markup and response_content.markup.strip()):
+                    response_content = None  # type: ignore
+
+            panels = [status]
+
+            if input and show_message:
+                render = True
+                # Convert message to a panel
+                message_content = get_text_from_message(input)
+                message_panel = create_panel(
+                    content=Text(message_content, style="green"),
+                    title="Message",
+                    border_style="cyan",
+                )
+                panels.append(message_panel)
+
+            additional_panels = build_panels_stream(
+                response_content=response_content,
+                response_event=resp,  # type: ignore
+                response_timer=response_timer,
+                response_reasoning_content_buffer=_response_reasoning_content,
+                reasoning_steps=reasoning_steps,
+                show_reasoning=show_reasoning,
+                show_full_reasoning=show_full_reasoning,
+                accumulated_tool_calls=accumulated_tool_calls,
+            )
+            panels.extend(additional_panels)
+            if panels:
+                live_log.update(Group(*panels))
+
+        if agent.memory_manager is not None and agent.memory_manager.memories_updated:
+            memory_panel = create_panel(
+                content=Text("Memories updated"),
+                title="Memories",
+                border_style="green",
+            )
+            panels.append(memory_panel)
+            live_log.update(Group(*panels))
+            agent.memory_manager.memories_updated = False
+
+        if agent.session_summary_manager is not None and agent.session_summary_manager.summaries_updated:
+            summary_panel = create_panel(
+                content=Text("Session summary updated"),
+                title="Session Summary",
+                border_style="green",
+            )
+            panels.append(summary_panel)
+            live_log.update(Group(*panels))
+            agent.session_summary_manager.summaries_updated = False
+
+        response_timer.stop()
+
+        # Final update to remove the "Thinking..." status
+        panels = [p for p in panels if not isinstance(p, Status)]
+        live_log.update(Group(*panels))
+
+
+def build_panels_stream(
+    response_content: Union[str, JSON, Markdown],
+    response_event: RunOutputEvent,
+    response_timer: Timer,
+    response_reasoning_content_buffer: str,
+    reasoning_steps: List[ReasoningStep],
+    show_reasoning: bool = True,
+    show_full_reasoning: bool = False,
+    accumulated_tool_calls: Optional[List] = None,
+):
+    panels = []
+
+    if len(reasoning_steps) > 0 and show_reasoning:
+        # Create panels for reasoning steps
+        for i, step in enumerate(reasoning_steps, 1):
+            # Build step content
+            step_content = Text.assemble()
+            if step.title is not None:
+                step_content.append(f"{step.title}\n", "bold")
+            if step.action is not None:
+                step_content.append(Text.from_markup(f"[bold]Action:[/bold] {step.action}\n", style="dim"))
+            if step.result is not None:
+                step_content.append(Text.from_markup(step.result, style="dim"))
+
+            if show_full_reasoning:
+                # Add detailed reasoning information if available
+                if step.reasoning is not None:
+                    step_content.append(Text.from_markup(f"\n[bold]Reasoning:[/bold] {step.reasoning}", style="dim"))
+                if step.confidence is not None:
+                    step_content.append(Text.from_markup(f"\n[bold]Confidence:[/bold] {step.confidence}", style="dim"))
+            reasoning_panel = create_panel(content=step_content, title=f"Reasoning step {i}", border_style="green")
+            panels.append(reasoning_panel)
+
+    if len(response_reasoning_content_buffer) > 0:
+        # Create panel for thinking
+        thinking_panel = create_panel(
+            content=Text(response_reasoning_content_buffer),
+            title=f"Thinking ({response_timer.elapsed:.1f}s)",
+            border_style="green",
+        )
+        panels.append(thinking_panel)
+
+    if accumulated_tool_calls:  # Use accumulated tool calls instead of just current event
+        # Create bullet points for each tool call
+        tool_calls_content = Text()
+        formatted_tool_calls = format_tool_calls(accumulated_tool_calls)
+        for formatted_tool_call in formatted_tool_calls:
+            tool_calls_content.append(f"• {formatted_tool_call}\n")
+
+        tool_calls_panel = create_panel(
+            content=tool_calls_content.plain.rstrip(),
+            title="Tool Calls",
+            border_style="yellow",
+        )
+        panels.append(tool_calls_panel)
+
+    response_panel = None
+    if response_content:
+        response_panel = create_panel(
+            content=response_content,
+            title=f"Response ({response_timer.elapsed:.1f}s)",
+            border_style="blue",
+        )
+        panels.append(response_panel)
+
+    if (
+        isinstance(response_event, tuple(get_args(RunOutputEvent)))
+        and hasattr(response_event, "citations")
+        and response_event.citations is not None
+        and response_event.citations.urls is not None
+    ):
+        md_content = "\n".join(
+            f"{i + 1}. [{citation.title or citation.url}]({citation.url})"
+            for i, citation in enumerate(response_event.citations.urls)
+            if citation.url  # Only include citations with valid URLs
+        )
+        if md_content:  # Only create panel if there are citations
+            citations_panel = create_panel(
+                content=Markdown(md_content),
+                title="Citations",
+                border_style="green",
+            )
+            panels.append(citations_panel)
+
+    return panels
+
+
+def print_response(
+    agent: "Agent",
+    input: Union[List, Dict, str, Message, BaseModel, List[Message]],
+    session_id: Optional[str] = None,
+    session_state: Optional[Dict[str, Any]] = None,
+    user_id: Optional[str] = None,
+    audio: Optional[Sequence[Audio]] = None,
+    images: Optional[Sequence[Image]] = None,
+    videos: Optional[Sequence[Video]] = None,
+    files: Optional[Sequence[File]] = None,
+    stream_intermediate_steps: bool = False,
+    knowledge_filters: Optional[Dict[str, Any]] = None,
+    debug_mode: Optional[bool] = None,
+    markdown: bool = False,
+    show_message: bool = True,
+    show_reasoning: bool = True,
+    show_full_reasoning: bool = False,
+    tags_to_include_in_markdown: Set[str] = {"think", "thinking"},
+    console: Optional[Any] = None,
+    add_history_to_context: Optional[bool] = None,
+    dependencies: Optional[Dict[str, Any]] = None,
+    metadata: Optional[Dict[str, Any]] = None,
+    **kwargs: Any,
+):
+    with Live(console=console) as live_log:
+        status = Status("Thinking...", spinner="aesthetic", speed=0.4, refresh_per_second=10)
+        live_log.update(status)
+        response_timer = Timer()
+        response_timer.start()
+        # Panels to be rendered
+        panels = [status]
+        # First render the message panel if the message is not None
+        if input and show_message:
+            # Convert message to a panel
+            message_content = get_text_from_message(input)
+            message_panel = create_panel(
+                content=Text(message_content, style="green"),
+                title="Message",
+                border_style="cyan",
+            )
+            panels.append(message_panel)  # type: ignore
+            live_log.update(Group(*panels))
+
+        # Run the agent
+        run_response = agent.run(
+            input=input,
+            session_id=session_id,
+            session_state=session_state,
+            user_id=user_id,
+            audio=audio,
+            images=images,
+            videos=videos,
+            files=files,
+            stream=False,
+            stream_intermediate_steps=stream_intermediate_steps,
+            knowledge_filters=knowledge_filters,
+            debug_mode=debug_mode,
+            add_history_to_context=add_history_to_context,
+            dependencies=dependencies,
+            metadata=metadata,
+            **kwargs,
+        )
+        response_timer.stop()
+
+        additional_panels = build_panels(
+            run_response=run_response,
+            output_schema=agent.output_schema,  # type: ignore
+            response_timer=response_timer,
+            show_reasoning=show_reasoning,
+            show_full_reasoning=show_full_reasoning,
+            tags_to_include_in_markdown=tags_to_include_in_markdown,
+            markdown=markdown,
+        )
+        panels.extend(additional_panels)
+
+        if agent.memory_manager is not None and agent.memory_manager.memories_updated:
+            memory_panel = create_panel(
+                content=Text("Memories updated"),
+                title="Memories",
+                border_style="green",
+            )
+            panels.append(memory_panel)
+            live_log.update(Group(*panels))
+            agent.memory_manager.memories_updated = False
+
+        if agent.session_summary_manager is not None and agent.session_summary_manager.summaries_updated:
+            summary_panel = create_panel(
+                content=Text("Session summary updated"),
+                title="Session Summary",
+                border_style="green",
+            )
+            panels.append(summary_panel)
+            live_log.update(Group(*panels))
+            agent.session_summary_manager.summaries_updated = False
+
+        # Final update to remove the "Thinking..." status
+        panels = [p for p in panels if not isinstance(p, Status)]
+        live_log.update(Group(*panels))
+
+
+async def aprint_response(
+    agent: "Agent",
+    input: Union[List, Dict, str, Message, BaseModel, List[Message]],
+    session_id: Optional[str] = None,
+    session_state: Optional[Dict[str, Any]] = None,
+    user_id: Optional[str] = None,
+    audio: Optional[Sequence[Audio]] = None,
+    images: Optional[Sequence[Image]] = None,
+    videos: Optional[Sequence[Video]] = None,
+    files: Optional[Sequence[File]] = None,
+    stream_intermediate_steps: bool = False,
+    knowledge_filters: Optional[Dict[str, Any]] = None,
+    debug_mode: Optional[bool] = None,
+    markdown: bool = False,
+    show_message: bool = True,
+    show_reasoning: bool = True,
+    show_full_reasoning: bool = False,
+    tags_to_include_in_markdown: Set[str] = {"think", "thinking"},
+    console: Optional[Any] = None,
+    add_history_to_context: Optional[bool] = None,
+    dependencies: Optional[Dict[str, Any]] = None,
+    metadata: Optional[Dict[str, Any]] = None,
+    **kwargs: Any,
+):
+    with Live(console=console) as live_log:
+        status = Status("Thinking...", spinner="aesthetic", speed=0.4, refresh_per_second=10)
+        live_log.update(status)
+        response_timer = Timer()
+        response_timer.start()
+        # Panels to be rendered
+        panels = [status]
+        # First render the message panel if the message is not None
+        if input and show_message:
+            # Convert message to a panel
+            message_content = get_text_from_message(input)
+            message_panel = create_panel(
+                content=Text(message_content, style="green"),
+                title="Message",
+                border_style="cyan",
+            )
+            panels.append(message_panel)
+            live_log.update(Group(*panels))
+
+        # Run the agent
+        run_response = await agent.arun(
+            input=input,
+            session_id=session_id,
+            session_state=session_state,
+            user_id=user_id,
+            audio=audio,
+            images=images,
+            videos=videos,
+            files=files,
+            stream=False,
+            stream_intermediate_steps=stream_intermediate_steps,
+            knowledge_filters=knowledge_filters,
+            debug_mode=debug_mode,
+            add_history_to_context=add_history_to_context,
+            dependencies=dependencies,
+            metadata=metadata,
+            **kwargs,
+        )
+        response_timer.stop()
+
+        additional_panels = build_panels(
+            run_response=run_response,
+            output_schema=agent.output_schema,  # type: ignore
+            response_timer=response_timer,
+            show_reasoning=show_reasoning,
+            show_full_reasoning=show_full_reasoning,
+            tags_to_include_in_markdown=tags_to_include_in_markdown,
+            markdown=markdown,
+        )
+        panels.extend(additional_panels)
+
+        if agent.memory_manager is not None and agent.memory_manager.memories_updated:
+            memory_panel = create_panel(
+                content=Text("Memories updated"),
+                title="Memories",
+                border_style="green",
+            )
+            panels.append(memory_panel)
+            live_log.update(Group(*panels))
+            agent.memory_manager.memories_updated = False
+
+        if agent.session_summary_manager is not None and agent.session_summary_manager.summaries_updated is not None:
+            summary_panel = create_panel(
+                content=Text("Session summary updated"),
+                title="Session Summary",
+                border_style="green",
+            )
+            agent.session_summary_manager.summaries_updated = False
+            panels.append(summary_panel)
+            live_log.update(Group(*panels))
+
+        # Final update to remove the "Thinking..." status
+        panels = [p for p in panels if not isinstance(p, Status)]
+        live_log.update(Group(*panels))
+
+
+def build_panels(
+    run_response: RunOutput,
+    response_timer: Timer,
+    output_schema: Optional[BaseModel] = None,
+    show_reasoning: bool = True,
+    show_full_reasoning: bool = False,
+    tags_to_include_in_markdown: Optional[Set[str]] = None,
+    markdown: bool = False,
+):
+    panels = []
+
+    reasoning_steps = []
+
+    if isinstance(run_response, RunOutput) and run_response.is_paused:
+        response_panel = create_paused_run_output_panel(run_response)
+        panels.append(response_panel)
+        return panels
+
+    if isinstance(run_response, RunOutput) and run_response.reasoning_steps is not None:
+        reasoning_steps = run_response.reasoning_steps
+
+    if len(reasoning_steps) > 0 and show_reasoning:
+        # Create panels for reasoning steps
+        for i, step in enumerate(reasoning_steps, 1):
+            # Build step content
+            step_content = Text.assemble()
+            if step.title is not None:
+                step_content.append(f"{step.title}\n", "bold")
+            if step.action is not None:
+                step_content.append(Text.from_markup(f"[bold]Action:[/bold] {step.action}\n", style="dim"))
+            if step.result is not None:
+                step_content.append(Text.from_markup(step.result, style="dim"))
+
+            if show_full_reasoning:
+                # Add detailed reasoning information if available
+                if step.reasoning is not None:
+                    step_content.append(Text.from_markup(f"\n[bold]Reasoning:[/bold] {step.reasoning}", style="dim"))
+                if step.confidence is not None:
+                    step_content.append(Text.from_markup(f"\n[bold]Confidence:[/bold] {step.confidence}", style="dim"))
+            reasoning_panel = create_panel(content=step_content, title=f"Reasoning step {i}", border_style="green")
+            panels.append(reasoning_panel)
+
+    if isinstance(run_response, RunOutput) and run_response.reasoning_content is not None:
+        # Create panel for thinking
+        thinking_panel = create_panel(
+            content=Text(run_response.reasoning_content),
+            title=f"Thinking ({response_timer.elapsed:.1f}s)",
+            border_style="green",
+        )
+        panels.append(thinking_panel)
+
+    # Add tool calls panel if available
+    if isinstance(run_response, RunOutput) and run_response.tools:
+        # Create bullet points for each tool call
+        tool_calls_content = Text()
+        formatted_tool_calls = format_tool_calls(run_response.tools)
+        for formatted_tool_call in formatted_tool_calls:
+            tool_calls_content.append(f"• {formatted_tool_call}\n")
+
+        tool_calls_panel = create_panel(
+            content=tool_calls_content.plain.rstrip(),
+            title="Tool Calls",
+            border_style="yellow",
+        )
+        panels.append(tool_calls_panel)
+
+    response_content_batch: Union[str, JSON, Markdown] = ""  # type: ignore
+    if isinstance(run_response, RunOutput):
+        if isinstance(run_response.content, str):
+            if markdown:
+                escaped_content = escape_markdown_tags(run_response.content, tags_to_include_in_markdown)  # type: ignore
+                response_content_batch = Markdown(escaped_content)
+            else:
+                response_content_batch = run_response.get_content_as_string(indent=4)
+        elif output_schema is not None and isinstance(run_response.content, BaseModel):
+            try:
+                response_content_batch = JSON(run_response.content.model_dump_json(exclude_none=True), indent=2)
+            except Exception as e:
+                log_warning(f"Failed to convert response to JSON: {e}")
+        else:
+            try:
+                response_content_batch = JSON(json.dumps(run_response.content), indent=4)
+            except Exception as e:
+                log_warning(f"Failed to convert response to JSON: {e}")
+
+    # Create panel for response
+    response_panel = create_panel(
+        content=response_content_batch,
+        title=f"Response ({response_timer.elapsed:.1f}s)",
+        border_style="blue",
+    )
+    panels.append(response_panel)
+
+    if (
+        isinstance(run_response, RunOutput)
+        and run_response.citations is not None
+        and run_response.citations.urls is not None
+    ):
+        md_content = "\n".join(
+            f"{i + 1}. [{citation.title or citation.url}]({citation.url})"
+            for i, citation in enumerate(run_response.citations.urls)
+            if citation.url  # Only include citations with valid URLs
+        )
+        if md_content:  # Only create panel if there are citations
+            citations_panel = create_panel(
+                content=Markdown(md_content),
+                title="Citations",
+                border_style="green",
+            )
+            panels.append(citations_panel)
+
+    return panels