praisonaiagents 0.0.12__tar.gz → 0.0.13__tar.gz

{praisonaiagents-0.0.12 → praisonaiagents-0.0.13}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: praisonaiagents
-Version: 0.0.12
+Version: 0.0.13
 Summary: Praison AI agents for completing complex tasks with Self Reflection Agents
 Author: Mervin Praison
 Requires-Dist: pydantic

{praisonaiagents-0.0.12 → praisonaiagents-0.0.13}/praisonaiagents/agent/agent.py

@@ -140,7 +140,7 @@ class Agent:
         max_rpm: Optional[int] = None,
         max_execution_time: Optional[int] = None,
         memory: bool = True,
-        verbose: bool = False,
+        verbose: bool = True,
         allow_delegation: bool = False,
         step_callback: Optional[Any] = None,
         cache: bool = True,
@@ -191,6 +191,8 @@ class Agent:
         self.max_reflect = max_reflect
         self.min_reflect = min_reflect
         self.reflect_llm = reflect_llm
+        self.console = Console()  # Create a single console instance for the agent
+
     def execute_tool(self, function_name, arguments):
         """
         Execute a tool dynamically based on the function name and arguments.
@@ -235,7 +237,6 @@ class Agent:
         return f"Agent(name='{self.name}', role='{self.role}', goal='{self.goal}')"
 
     def _chat_completion(self, messages, temperature=0.2, tools=None, stream=True):
-        console = Console()
         start_time = time.time()
         logging.debug(f"{self.name} sending messages to LLM: {messages}")
 
@@ -305,12 +306,24 @@ class Agent:
                     stream=True
                 )
                 full_response_text = ""
-                with Live(display_generating("", start_time), refresh_per_second=4) as live:
+                
+                # Create Live display with proper configuration
+                with Live(
+                    display_generating("", start_time),
+                    console=self.console,
+                    refresh_per_second=4,
+                    transient=False,  # Changed to False to preserve output
+                    vertical_overflow="ellipsis",
+                    auto_refresh=True
+                ) as live:
                     for chunk in response_stream:
                         if chunk.choices[0].delta.content:
                             full_response_text += chunk.choices[0].delta.content
                             live.update(display_generating(full_response_text, start_time))
-
+                
+                # Clear the last generating display with a blank line
+                self.console.print()
+                
                 final_response = client.chat.completions.create(
                     model=self.llm,
                     messages=messages,
@@ -347,7 +360,11 @@ Your Goal: {self.goal}
         if system_prompt:
             messages.append({"role": "system", "content": system_prompt})
         messages.extend(self.chat_history)
-        messages.append({"role": "user", "content": prompt})
+        if isinstance(prompt, list):
+            # If we receive a multimodal prompt list, place it directly in the user message
+            messages.append({"role": "user", "content": prompt})
+        else:
+            messages.append({"role": "user", "content": prompt})
 
         final_response_text = None
         reflection_count = 0
@@ -356,7 +373,14 @@ Your Goal: {self.goal}
         while True:
             try:
                 if self.verbose:
-                    display_instruction(f"Agent {self.name} is processing prompt: {prompt}")
+                    # Handle both string and list prompts for instruction display
+                    display_text = prompt
+                    if isinstance(prompt, list):
+                        # Extract text content from multimodal prompt
+                        display_text = next((item["text"] for item in prompt if item["type"] == "text"), "")
+                    
+                    if display_text and str(display_text).strip():
+                        display_instruction(f"Agent {self.name} is processing prompt: {display_text}", console=self.console)
 
                 response = self._chat_completion(messages, temperature=temperature, tools=tools if tools else None)
                 if not response:
@@ -376,13 +400,13 @@ Your Goal: {self.goal}
                         arguments = json.loads(tool_call.function.arguments)
 
                         if self.verbose:
-                            display_tool_call(f"Agent {self.name} is calling function '{function_name}' with arguments: {arguments}")
+                            display_tool_call(f"Agent {self.name} is calling function '{function_name}' with arguments: {arguments}", console=self.console)
 
                         tool_result = self.execute_tool(function_name, arguments)
 
                         if tool_result:
                             if self.verbose:
-                                display_tool_call(f"Function '{function_name}' returned: {tool_result}")
+                                display_tool_call(f"Function '{function_name}' returned: {tool_result}", console=self.console)
                             messages.append({
                                 "role": "tool",
                                 "tool_call_id": tool_call.id,
@@ -407,7 +431,7 @@ Your Goal: {self.goal}
                     self.chat_history.append({"role": "assistant", "content": response_text})
                     if self.verbose:
                         logging.info(f"Agent {self.name} final response: {response_text}")
-                    display_interaction(prompt, response_text, markdown=self.markdown, generation_time=time.time() - start_time)
+                    display_interaction(prompt, response_text, markdown=self.markdown, generation_time=time.time() - start_time, console=self.console)
                     return response_text
 
                 reflection_prompt = f"""
@@ -430,26 +454,26 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
                     reflection_output = reflection_response.choices[0].message.parsed
 
                     if self.verbose:
-                        display_self_reflection(f"Agent {self.name} self reflection (using {self.reflect_llm if self.reflect_llm else self.llm}): reflection='{reflection_output.reflection}' satisfactory='{reflection_output.satisfactory}'")
+                        display_self_reflection(f"Agent {self.name} self reflection (using {self.reflect_llm if self.reflect_llm else self.llm}): reflection='{reflection_output.reflection}' satisfactory='{reflection_output.satisfactory}'", console=self.console)
 
                     messages.append({"role": "assistant", "content": f"Self Reflection: {reflection_output.reflection} Satisfactory?: {reflection_output.satisfactory}"})
 
                     # Only consider satisfactory after minimum reflections
                     if reflection_output.satisfactory == "yes" and reflection_count >= self.min_reflect - 1:
                         if self.verbose:
-                            display_self_reflection("Agent marked the response as satisfactory after meeting minimum reflections")
+                            display_self_reflection("Agent marked the response as satisfactory after meeting minimum reflections", console=self.console)
                         self.chat_history.append({"role": "user", "content": prompt})
                         self.chat_history.append({"role": "assistant", "content": response_text})
-                        display_interaction(prompt, response_text, markdown=self.markdown, generation_time=time.time() - start_time)
+                        display_interaction(prompt, response_text, markdown=self.markdown, generation_time=time.time() - start_time, console=self.console)
                         return response_text
 
                     # Check if we've hit max reflections
                     if reflection_count >= self.max_reflect - 1:
                         if self.verbose:
-                            display_self_reflection("Maximum reflection count reached, returning current response")
+                            display_self_reflection("Maximum reflection count reached, returning current response", console=self.console)
                         self.chat_history.append({"role": "user", "content": prompt})
                         self.chat_history.append({"role": "assistant", "content": response_text})
-                        display_interaction(prompt, response_text, markdown=self.markdown, generation_time=time.time() - start_time)
+                        display_interaction(prompt, response_text, markdown=self.markdown, generation_time=time.time() - start_time, console=self.console)
                         return response_text
 
                     logging.debug(f"{self.name} reflection count {reflection_count + 1}, continuing reflection process")
@@ -460,12 +484,12 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
                     continue  # Continue the loop for more reflections
 
                 except Exception as e:
-                    display_error(f"Error in parsing self-reflection json {e}. Retrying")
+                    display_error(f"Error in parsing self-reflection json {e}. Retrying", console=self.console)
                     logging.error("Reflection parsing failed.", exc_info=True)
                     messages.append({"role": "assistant", "content": f"Self Reflection failed."})
                     reflection_count += 1
                     continue  # Continue even after error to try again
                 
             except Exception as e:
-                display_error(f"Error in chat: {e}")
+                display_error(f"Error in chat: {e}", console=self.console)
                 return None 

{praisonaiagents-0.0.12 → praisonaiagents-0.0.13}/praisonaiagents/agents/agents.py

@@ -11,6 +11,33 @@ from ..main import display_error, TaskOutput, error_logs, client
 from ..agent.agent import Agent
 from ..task.task import Task
 
+def encode_file_to_base64(file_path: str) -> str:
+    """Base64-encode a file."""
+    import base64
+    with open(file_path, "rb") as f:
+        return base64.b64encode(f.read()).decode("utf-8")
+
+def process_video(video_path: str, seconds_per_frame=2):
+    """Split video into frames (base64-encoded)."""
+    import cv2
+    import base64
+    base64_frames = []
+    video = cv2.VideoCapture(video_path)
+    total_frames = int(video.get(cv2.CAP_PROP_FRAME_COUNT))
+    fps = video.get(cv2.CAP_PROP_FPS)
+    frames_to_skip = int(fps * seconds_per_frame)
+    curr_frame = 0
+    while curr_frame < total_frames:
+        video.set(cv2.CAP_PROP_POS_FRAMES, curr_frame)
+        success, frame = video.read()
+        if not success:
+            break
+        _, buffer = cv2.imencode(".jpg", frame)
+        base64_frames.append(base64.b64encode(buffer).decode("utf-8"))
+        curr_frame += frames_to_skip
+    video.release()
+    return base64_frames
+
 class PraisonAIAgents:
     def __init__(self, agents, tasks, verbose=0, completion_checker=None, max_retries=5, process="sequential", manager_llm=None):
         self.agents = agents
@@ -58,6 +85,19 @@ class PraisonAIAgents:
             display_error(f"Error: Task with ID {task_id} does not exist")
             return
         task = self.tasks[task_id]
+        
+        # Only import multimodal dependencies if task has images
+        if task.images and task.status == "not started":
+            try:
+                import cv2
+                import base64
+                from moviepy import VideoFileClip
+            except ImportError as e:
+                display_error(f"Error: Missing required dependencies for image/video processing: {e}")
+                display_error("Please install with: pip install opencv-python moviepy")
+                task.status = "failed"
+                return None
+
         if task.status == "not started":
             task.status = "in progress"
 
@@ -83,7 +123,47 @@ Expected Output: {task.expected_output}.
         if self.verbose >= 2:
             logging.info(f"Executing task {task_id}: {task.description} using {executor_agent.name}")
         logging.debug(f"Starting execution of task {task_id} with prompt:\n{task_prompt}")
-        agent_output = executor_agent.chat(task_prompt, tools=task.tools)
+
+        if task.images:
+            def _get_multimodal_message(text_prompt, images):
+                content = [{"type": "text", "text": text_prompt}]
+
+                for img in images:
+                    # If local file path for a valid image
+                    if os.path.exists(img):
+                        ext = os.path.splitext(img)[1].lower()
+                        # If it's a .mp4, convert to frames
+                        if ext == ".mp4":
+                            frames = process_video(img, seconds_per_frame=1)
+                            content.append({"type": "text", "text": "These are frames from the video."})
+                            for f in frames:
+                                content.append({
+                                    "type": "image_url",
+                                    "image_url": {"url": f"data:image/jpg;base64,{f}"}
+                                })
+                        else:
+                            encoded = encode_file_to_base64(img)
+                            content.append({
+                                "type": "image_url",
+                                "image_url": {
+                                    "url": f"data:image/{ext.lstrip('.')};base64,{encoded}"
+                                }
+                            })
+                    else:
+                        # Treat as a remote URL
+                        content.append({
+                            "type": "image_url",
+                            "image_url": {"url": img}
+                        })
+                return content
+
+            agent_output = executor_agent.chat(
+                _get_multimodal_message(task_prompt, task.images),
+                tools=task.tools
+            )
+        else:
+            agent_output = executor_agent.chat(task_prompt, tools=task.tools)
+
         if agent_output:
             task_output = TaskOutput(
                 description=task.description,

{praisonaiagents-0.0.12 → praisonaiagents-0.0.13}/praisonaiagents/main.py

@@ -25,43 +25,92 @@ logging.basicConfig(
 # Global list to store error logs
 error_logs = []
 
-def display_interaction(message: str, response: str, markdown: bool = True, generation_time: Optional[float] = None):
-    console = Console()
-    if generation_time is not None:
+def _clean_display_content(content: str, max_length: int = 20000) -> str:
+    """Helper function to clean and truncate content for display."""
+    if not content or not str(content).strip():
+        return ""
+        
+    content = str(content)
+    # Handle base64 content
+    if "base64" in content:
+        content_parts = []
+        for line in content.split('\n'):
+            if "base64" not in line:
+                content_parts.append(line)
+        content = '\n'.join(content_parts)
+    
+    # Truncate if too long
+    if len(content) > max_length:
+        content = content[:max_length] + "..."
+    
+    return content.strip()
+
+def display_interaction(message, response, markdown=True, generation_time=None, console=None):
+    """Display the interaction between user and assistant."""
+    if console is None:
+        console = Console()
+    if generation_time:
         console.print(Text(f"Response generated in {generation_time:.1f}s", style="dim"))
-    else:
-        console.print(Text("Response Generation Complete", style="dim"))
+
+    # Handle multimodal content (list)
+    if isinstance(message, list):
+        # Extract just the text content from the multimodal message
+        text_content = next((item["text"] for item in message if item["type"] == "text"), "")
+        message = text_content
+
+    message = _clean_display_content(str(message))
+    response = _clean_display_content(str(response))
 
     if markdown:
         console.print(Panel.fit(Markdown(message), title="Message", border_style="cyan"))
         console.print(Panel.fit(Markdown(response), title="Response", border_style="cyan"))
     else:
         console.print(Panel.fit(Text(message, style="bold green"), title="Message", border_style="cyan"))
-        console.print(Panel.fit(Text(response, style="bold white"), title="Response", border_style="cyan"))
-
-def display_self_reflection(message: str):
-    console = Console()
+        console.print(Panel.fit(Text(response, style="bold blue"), title="Response", border_style="cyan"))
+
+def display_self_reflection(message: str, console=None):
+    if not message or not message.strip():
+        return
+    if console is None:
+        console = Console()
+    message = _clean_display_content(str(message))
     console.print(Panel.fit(Text(message, style="bold yellow"), title="Self Reflection", border_style="magenta"))
 
-def display_instruction(message: str):
-    console = Console()
+def display_instruction(message: str, console=None):
+    if not message or not message.strip():
+        return
+    if console is None:
+        console = Console()
+    message = _clean_display_content(str(message))
     console.print(Panel.fit(Text(message, style="bold blue"), title="Instruction", border_style="cyan"))
 
-def display_tool_call(message: str):
-    console = Console()
+def display_tool_call(message: str, console=None):
+    if not message or not message.strip():
+        return
+    if console is None:
+        console = Console()
+    message = _clean_display_content(str(message))
     console.print(Panel.fit(Text(message, style="bold cyan"), title="Tool Call", border_style="green"))
 
-def display_error(message: str):
-    console = Console()
+def display_error(message: str, console=None):
+    if not message or not message.strip():
+        return
+    if console is None:
+        console = Console()
+    message = _clean_display_content(str(message))
     console.print(Panel.fit(Text(message, style="bold red"), title="Error", border_style="red"))
     # Store errors
     error_logs.append(message)
 
 def display_generating(content: str = "", start_time: Optional[float] = None):
+    if not content or not str(content).strip():
+        return Panel("", title="", border_style="green")  # Return empty panel when no content
     elapsed_str = ""
     if start_time is not None:
         elapsed = time.time() - start_time
         elapsed_str = f" {elapsed:.1f}s"
+    
+    content = _clean_display_content(str(content))
     return Panel(Markdown(content), title=f"Generating...{elapsed_str}", border_style="green")
 
 def clean_triple_backticks(text: str) -> str:

{praisonaiagents-0.0.12 → praisonaiagents-0.0.13}/praisonaiagents/task/task.py

@@ -22,7 +22,8 @@ class Task:
         status: str = "not started",
         result: Optional[TaskOutput] = None,
         create_directory: Optional[bool] = False,
-        id: Optional[int] = None
+        id: Optional[int] = None,
+        images: Optional[List[str]] = None
     ):
         self.description = description
         self.expected_output = expected_output
@@ -40,6 +41,7 @@ class Task:
         self.result = result
         self.create_directory = create_directory
         self.id = id
+        self.images = images if images else []
 
         if self.output_json and self.output_pydantic:
             raise ValueError("Only one output type can be defined")

{praisonaiagents-0.0.12 → praisonaiagents-0.0.13}/praisonaiagents.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: praisonaiagents
-Version: 0.0.12
+Version: 0.0.13
 Summary: Praison AI agents for completing complex tasks with Self Reflection Agents
 Author: Mervin Praison
 Requires-Dist: pydantic

{praisonaiagents-0.0.12 → praisonaiagents-0.0.13}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "praisonaiagents"
-version = "0.0.12"
+version = "0.0.13"
 description = "Praison AI agents for completing complex tasks with Self Reflection Agents"
 authors = [
     { name="Mervin Praison" }