praisonaiagents 0.0.156__tar.gz → 0.0.158__tar.gz

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: praisonaiagents
-Version: 0.0.156
+Version: 0.0.158
 Summary: Praison AI agents for completing complex tasks with Self Reflection Agents
 Author: Mervin Praison
 Requires-Python: >=3.10

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/agent/agent.py

@@ -716,7 +716,7 @@ Your Goal: {self.goal}
                 error=f"Agent guardrail validation error: {str(e)}"
             )
 
-    def _apply_guardrail_with_retry(self, response_text, prompt, temperature=0.2, tools=None, task_name=None, task_description=None, task_id=None):
+    def _apply_guardrail_with_retry(self, response_text, prompt, temperature=1.0, tools=None, task_name=None, task_description=None, task_id=None):
         """Apply guardrail validation with retry logic.
         
         Args:
@@ -859,7 +859,7 @@ Your Goal: {self.goal}"""
             self._system_prompt_cache[cache_key] = system_prompt
         return system_prompt
 
-    def _build_messages(self, prompt, temperature=0.2, output_json=None, output_pydantic=None, tools=None):
+    def _build_messages(self, prompt, temperature=1.0, output_json=None, output_pydantic=None, tools=None):
         """Build messages list for chat completion.
         
         Args:
@@ -1172,7 +1172,7 @@ Your Goal: {self.goal}"""
             reasoning_steps=reasoning_steps
         )
 
-    def _chat_completion(self, messages, temperature=0.2, tools=None, stream=True, reasoning_steps=False, task_name=None, task_description=None, task_id=None):
+    def _chat_completion(self, messages, temperature=1.0, tools=None, stream=True, reasoning_steps=False, task_name=None, task_description=None, task_id=None):
         start_time = time.time()
         logging.debug(f"{self.name} sending messages to LLM: {messages}")
 
@@ -1336,7 +1336,7 @@ Your Goal: {self.goal}"""
         #         expand=False
         #     )
 
-    def chat(self, prompt, temperature=0.2, tools=None, output_json=None, output_pydantic=None, reasoning_steps=False, stream=None, task_name=None, task_description=None, task_id=None):
+    def chat(self, prompt, temperature=1.0, tools=None, output_json=None, output_pydantic=None, reasoning_steps=False, stream=None, task_name=None, task_description=None, task_id=None):
         # Reset the final display flag for each new conversation
         self._final_display_shown = False
         
@@ -1694,7 +1694,7 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
             cleaned = cleaned[:-3].strip()
         return cleaned  
 
-    async def achat(self, prompt: str, temperature=0.2, tools=None, output_json=None, output_pydantic=None, reasoning_steps=False, task_name=None, task_description=None, task_id=None):
+    async def achat(self, prompt: str, temperature=1.0, tools=None, output_json=None, output_pydantic=None, reasoning_steps=False, task_name=None, task_description=None, task_id=None):
         """Async version of chat method with self-reflection support.""" 
         # Reset the final display flag for each new conversation
         self._final_display_shown = False
@@ -2046,7 +2046,7 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
                         final_response = await self._openai_client.async_client.chat.completions.create(
                             model=self.llm,
                             messages=messages,
-                            temperature=0.2,
+                            temperature=1.0,
                             stream=True
                         )
                         full_response_text = ""
@@ -2169,7 +2169,7 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
                         prompt=actual_prompt,
                         system_prompt=self._build_system_prompt(tool_param),
                         chat_history=self.chat_history,
-                        temperature=kwargs.get('temperature', 0.2),
+                        temperature=kwargs.get('temperature', 1.0),
                         tools=tool_param,
                         output_json=kwargs.get('output_json'),
                         output_pydantic=kwargs.get('output_pydantic'),
@@ -2220,7 +2220,7 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
                     tool_param = tools
                 
                 # Build messages using the helper method
-                messages, original_prompt = self._build_messages(actual_prompt, kwargs.get('temperature', 0.2), 
+                messages, original_prompt = self._build_messages(actual_prompt, kwargs.get('temperature', 1.0), 
                                                                kwargs.get('output_json'), kwargs.get('output_pydantic'))
                 
                 # Store chat history length for potential rollback
@@ -2249,7 +2249,7 @@ Output MUST be JSON with 'reflection' and 'satisfactory'.
                     completion_args = {
                         "model": self.llm,
                         "messages": messages,
-                        "temperature": kwargs.get('temperature', 0.2),
+                        "temperature": kwargs.get('temperature', 1.0),
                         "stream": True
                     }
                     if formatted_tools:

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/agent/context_agent.py

@@ -434,7 +434,7 @@ This report contains all agent interactions and outputs from a complete ContextA
                     
                     Provide comprehensive analysis that follows the PRD template principles and enables 
                     AI assistants to implement features that perfectly match existing codebase patterns.""",
-                    llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+                    llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
                     verbose=getattr(self, 'verbose', True)
                 )
                 
@@ -515,7 +515,7 @@ codebase style and architecture following PRD template principles."""
             role="Expert Manual Codebase Analysis Specialist",
             goal="Perform comprehensive manual codebase analysis following PRD methodology",
             instructions="""Analyze the codebase samples following PRD template methodology for complete understanding.""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -566,7 +566,7 @@ Analyze following PRD principles to extract patterns, conventions, and architect
             6. Design pattern implementations
             7. Code complexity metrics
             8. API and interface patterns""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -624,7 +624,7 @@ Extract comprehensive patterns that follow PRD template principles for implement
             
             For each pattern, provide the pattern name, where it's used, and how to replicate it 
             following PRD template principles.""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -674,7 +674,7 @@ patterns and best practices for first-try success."""
             goal="Analyze testing patterns for comprehensive validation framework design",
             instructions="""Analyze testing patterns to understand validation approaches and create 
             comprehensive test frameworks following PRD methodology.""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -801,7 +801,7 @@ Extract testing patterns for validation framework creation following PRD princip
             Confidence level for one-pass implementation
             
             Generate PRPs following this EXACT structure for first-try implementation success.""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -850,7 +850,7 @@ on the first try following PRD template principles."""
             6. CODE QUALITY: Complexity analysis, maintainability
             7. DOCUMENTATION VALIDATION: Documentation completeness
             8. DEPENDENCY VALIDATION: Dependency analysis and security""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -897,7 +897,7 @@ following PRD template principles."""
             instructions="""Compile all available documentation following PRD methodology including:
             README files, API documentation, setup guides, architecture docs, and any other 
             relevant documentation that provides context for implementation.""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -943,7 +943,7 @@ following PRD template principles."""
             instructions="""Analyze integration points following PRD methodology including:
             APIs, databases, external services, configuration points, and any other 
             integration requirements that affect implementation.""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -1004,7 +1004,7 @@ following PRD template principles."""
             8. DOCUMENTATION UPDATES: Documentation to create/update
             9. INTEGRATION STEPS: How to integrate with existing systems
             10. VALIDATION CHECKPOINTS: Validation steps at each phase""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -1302,7 +1302,7 @@ Every agent interaction has been saved for full audit trail and reproducibility.
             GOAL: [extracted implementation goal]
             
             Be precise and extract only what is explicitly mentioned or clearly implied.""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -1536,7 +1536,7 @@ Note: Detailed function/class metadata not available due to content access limit
             5. Documentation topics
             
             Make the output easy for a file selection agent to understand which files contain what functionality.""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -1770,7 +1770,7 @@ Focus on creating clear, structured metadata that will help with intelligent fil
             ["README.md", "src/auth/login.py", "config/settings.py", ...]
             
             Maximum 50 files for efficient analysis.""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -1923,7 +1923,7 @@ Maximum 50 files.""".format(goal=goal)
             8. EXAMPLES: Similar features that can guide {goal} implementation
             
             Since these files were pre-selected for relevance, provide deep analysis of how each contributes to implementing: {goal}""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -2024,7 +2024,7 @@ Since these files were pre-selected for relevance, explain how each contributes
             - Success criteria for {goal}
             
             Focus everything on successfully implementing: {goal}""",
-            llm=self.llm if hasattr(self, 'llm') else "gpt-4o-mini",
+            llm=self.llm if hasattr(self, 'llm') else "gpt-5-nano",
             verbose=getattr(self, 'verbose', True)
         )
         
@@ -2303,13 +2303,13 @@ def create_context_agent(llm: Optional[Union[str, Any]] = None, **kwargs) -> Con
     Factory function to create a ContextAgent following Context Engineering and PRD methodology.
     
     Args:
-        llm: Language model to use (e.g., "gpt-4o-mini", "claude-3-haiku")
+        llm: Language model to use (e.g., "gpt-5-nano", "claude-3-haiku")
         **kwargs: Additional arguments to pass to ContextAgent constructor
         
     Returns:
         ContextAgent: Configured ContextAgent for comprehensive context generation following PRD principles
     """
     if llm is None:
-        llm = "gpt-4o-mini"
+        llm = "gpt-5-nano"
     
     return ContextAgent(llm=llm, **kwargs)

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/agent/router_agent.py

@@ -44,7 +44,7 @@ class RouterAgent(Agent):
         # Initialize model router
         self.model_router = model_router or ModelRouter()
         self.routing_strategy = routing_strategy
-        self.fallback_model = fallback_model or os.getenv('OPENAI_MODEL_NAME', 'gpt-4o-mini')
+        self.fallback_model = fallback_model or os.getenv('OPENAI_MODEL_NAME', 'gpt-5-nano')
         
         # Process models configuration
         self.available_models = self._process_models_config(models)

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/llm/llm.py

@@ -7,6 +7,7 @@ from typing import Any, Dict, List, Optional, Union, Literal, Callable
 from pydantic import BaseModel
 import time
 import json
+import xml.etree.ElementTree as ET
 from ..main import (
     display_error,
     display_tool_call,
@@ -61,7 +62,7 @@ class LLM:
         # OpenAI
         "gpt-4": 6144,                    # 8,192 actual
         "gpt-4o": 96000,                  # 128,000 actual
-        "gpt-4o-mini": 96000,            # 128,000 actual
+        "gpt-5-nano": 96000,            # 128,000 actual
         "gpt-4-turbo": 96000,            # 128,000 actual
         "o1-preview": 96000,             # 128,000 actual
         "o1-mini": 96000,                # 128,000 actual
@@ -281,6 +282,8 @@ class LLM:
         self.min_reflect = extra_settings.get('min_reflect', 1)
         self.reasoning_steps = extra_settings.get('reasoning_steps', False)
         self.metrics = extra_settings.get('metrics', False)
+        # Auto-detect XML tool format for known models, or allow manual override
+        self.xml_tool_format = extra_settings.get('xml_tool_format', 'auto')
         
         # Token tracking
         self.last_token_metrics: Optional[TokenMetrics] = None
@@ -359,6 +362,25 @@ class LLM:
         
         return False
 
+    def _is_qwen_provider(self) -> bool:
+        """Detect if this is a Qwen provider"""
+        if not self.model:
+            return False
+        
+        # Check for Qwen patterns in model name
+        model_lower = self.model.lower()
+        return any(pattern in model_lower for pattern in ["qwen", "qwen2", "qwen2.5"])
+
+    def _supports_xml_tool_format(self) -> bool:
+        """Check if the model should use XML tool format"""
+        if self.xml_tool_format == 'auto':
+            # Auto-detect based on known models that use XML format
+            return self._is_qwen_provider()
+        elif self.xml_tool_format in [True, 'true', 'True']:
+            return True
+        else:
+            return False
+
     def _generate_ollama_tool_summary(self, tool_results: List[Any], response_text: str) -> Optional[str]:
         """
         Generate a summary from tool results for Ollama to prevent infinite loops.
@@ -658,6 +680,10 @@ class LLM:
         if any(self.model.startswith(prefix) for prefix in ["gemini-", "gemini/"]):
             return True
         
+        # Models with XML tool format support streaming with tools
+        if self._supports_xml_tool_format():
+            return True
+        
         # For other providers, default to False to be safe
         # This ensures we make a single non-streaming call rather than risk
         # missing tool calls or making duplicate calls
@@ -1427,6 +1453,64 @@ class LLM:
                         except (json.JSONDecodeError, KeyError) as e:
                             logging.debug(f"Could not parse Ollama tool call from response: {e}")
                     
+                    # Parse tool calls from XML format in response text 
+                    # Try for known XML models first, or fallback for any model that might output XML
+                    if not tool_calls and response_text and formatted_tools:
+                        # Check if this model is known to use XML format, or try as fallback
+                        should_try_xml = (self._supports_xml_tool_format() or 
+                                        # Fallback: try XML if response contains XML-like tool call tags
+                                        '<tool_call>' in response_text)
+                        
+                        if should_try_xml:
+                            tool_calls = []
+                            
+                            # Try proper XML parsing first
+                            try:
+                                # Wrap in root element if multiple tool_call tags exist
+                                xml_content = f"<root>{response_text}</root>"
+                                root = ET.fromstring(xml_content)
+                                tool_call_elements = root.findall('.//tool_call')
+                                
+                                for idx, element in enumerate(tool_call_elements):
+                                    if element.text:
+                                        try:
+                                            tool_json = json.loads(element.text.strip())
+                                            if isinstance(tool_json, dict) and "name" in tool_json:
+                                                tool_calls.append({
+                                                    "id": f"tool_{iteration_count}_{idx}",
+                                                    "type": "function",
+                                                    "function": {
+                                                        "name": tool_json["name"],
+                                                        "arguments": json.dumps(tool_json.get("arguments", {}))
+                                                    }
+                                                })
+                                        except (json.JSONDecodeError, KeyError) as e:
+                                            logging.debug(f"Could not parse tool call JSON: {e}")
+                                            continue
+                            except ET.ParseError:
+                                # Fallback to regex if XML parsing fails
+                                tool_call_pattern = r'<tool_call>\s*(\{(?:[^{}]|{[^{}]*})*\})\s*</tool_call>'
+                                matches = re.findall(tool_call_pattern, response_text, re.DOTALL)
+                                
+                                for idx, match in enumerate(matches):
+                                    try:
+                                        tool_json = json.loads(match.strip())
+                                        if isinstance(tool_json, dict) and "name" in tool_json:
+                                            tool_calls.append({
+                                                "id": f"tool_{iteration_count}_{idx}",
+                                                "type": "function",
+                                                "function": {
+                                                    "name": tool_json["name"],
+                                                    "arguments": json.dumps(tool_json.get("arguments", {}))
+                                                }
+                                            })
+                                    except (json.JSONDecodeError, KeyError) as e:
+                                        logging.debug(f"Could not parse XML tool call: {e}")
+                                        continue
+                            
+                            if tool_calls:
+                                logging.debug(f"Parsed {len(tool_calls)} tool call(s) from XML format")
+                    
                     # For Ollama, if response is empty but we have tools, prompt for tool usage
                     if self._is_ollama_provider() and (not response_text or response_text.strip() == "") and formatted_tools and iteration_count == 0:
                         messages.append({

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/llm/model_capabilities.py

@@ -7,7 +7,7 @@ This module defines which models support specific features like structured outpu
 MODELS_SUPPORTING_STRUCTURED_OUTPUTS = {
     # OpenAI models
     "gpt-4o",
-    "gpt-4o-mini",
+    "gpt-5-nano",
     "gpt-4-turbo",
     "gpt-4-turbo-preview",
     "gpt-4-turbo-2024-04-09",
@@ -46,7 +46,7 @@ MODELS_SUPPORTING_STRUCTURED_OUTPUTS = {
 MODELS_NOT_SUPPORTING_STRUCTURED_OUTPUTS = {
     # Audio preview models
     "gpt-4o-audio-preview",
-    "gpt-4o-mini-audio-preview",
+    "gpt-5-nano-audio-preview",
     
     # Legacy o1 models (don't support system messages either)
     "o1-preview-2024-09-12",

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/llm/model_router.py

@@ -51,7 +51,7 @@ class ModelRouter:
     DEFAULT_MODELS = [
         # Lightweight/cheap models for simple tasks
         ModelProfile(
-            name="gpt-4o-mini",
+            name="gpt-5-nano",
             provider="openai",
             complexity_range=(TaskComplexity.SIMPLE, TaskComplexity.MODERATE),
             cost_per_1k_tokens=0.00075,  # Average of $0.00015 input, $0.0006 output

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/llm/openai_client.py

@@ -230,19 +230,34 @@ class OpenAIClient:
                 f"(e.g., 'http://localhost:1234/v1') and you can use a placeholder API key by setting OPENAI_API_KEY='{LOCAL_SERVER_API_KEY_PLACEHOLDER}'"
             )
         
-        # Initialize synchronous client (lazy loading for async)
-        self._sync_client = OpenAI(api_key=self.api_key, base_url=self.base_url)
+        # Initialize clients lazily
+        self._sync_client = None
         self._async_client = None
         
         # Set up logging
         self.logger = logging.getLogger(__name__)
         
-        # Initialize console for display
-        self.console = Console()
+        # Initialize console lazily
+        self._console = None
+        
+        # Cache for formatted tools and fixed schemas
+        self._formatted_tools_cache = {}
+        self._fixed_schema_cache = {}
+        self._max_cache_size = 100
+    
+    @property
+    def console(self):
+        """Lazily initialize Rich Console only when needed."""
+        if self._console is None:
+            from rich.console import Console
+            self._console = Console()
+        return self._console
     
     @property
     def sync_client(self) -> OpenAI:
-        """Get the synchronous OpenAI client."""
+        """Get the synchronous OpenAI client (lazy initialization)."""
+        if self._sync_client is None:
+            self._sync_client = OpenAI(api_key=self.api_key, base_url=self.base_url)
         return self._sync_client
     
     @property
@@ -350,6 +365,35 @@ class OpenAIClient:
             
         return fixed_schema
     
+    def _get_tools_cache_key(self, tools: List[Any]) -> str:
+        """Generate a cache key for tools."""
+        parts = []
+        for tool in tools:
+            if isinstance(tool, dict):
+                # For dict tools, use sorted JSON representation
+                parts.append(json.dumps(tool, sort_keys=True))
+            elif callable(tool):
+                # For functions, use module.name
+                parts.append(f"{tool.__module__}.{tool.__name__}")
+            elif isinstance(tool, str):
+                # For string tools, use as-is
+                parts.append(tool)
+            elif isinstance(tool, list):
+                # For lists, recursively process
+                subparts = []
+                for subtool in tool:
+                    if isinstance(subtool, dict):
+                        subparts.append(json.dumps(subtool, sort_keys=True))
+                    elif callable(subtool):
+                        subparts.append(f"{subtool.__module__}.{subtool.__name__}")
+                    else:
+                        subparts.append(str(subtool))
+                parts.append(f"[{','.join(subparts)}]")
+            else:
+                # For other types, use string representation
+                parts.append(str(tool))
+        return "|".join(parts)
+    
     def format_tools(self, tools: Optional[List[Any]]) -> Optional[List[Dict]]:
         """
         Format tools for OpenAI API.
@@ -370,6 +414,11 @@ class OpenAIClient:
         """
         if not tools:
             return None
+        
+        # Check cache first
+        cache_key = self._get_tools_cache_key(tools)
+        if cache_key in self._formatted_tools_cache:
+            return self._formatted_tools_cache[cache_key]
             
         formatted_tools = []
         for tool in tools:
@@ -424,8 +473,13 @@ class OpenAIClient:
             except (TypeError, ValueError) as e:
                 logging.error(f"Tools are not JSON serializable: {e}")
                 return None
+        
+        # Cache the result
+        result = formatted_tools if formatted_tools else None
+        if result is not None and len(self._formatted_tools_cache) < self._max_cache_size:
+            self._formatted_tools_cache[cache_key] = result
                 
-        return formatted_tools if formatted_tools else None
+        return result
     
     def _generate_tool_definition(self, func: Callable) -> Optional[Dict]:
         """Generate a tool definition from a callable function."""
@@ -513,7 +567,7 @@ class OpenAIClient:
         self,
         messages: List[Dict],
         model: str,
-        temperature: float = 0.7,
+        temperature: float = 1.0,
         tools: Optional[List[Dict]] = None,
         start_time: Optional[float] = None,
         console: Optional[Console] = None,
@@ -546,7 +600,7 @@ class OpenAIClient:
                 console = self.console
             
             # Create the response stream
-            response_stream = self._sync_client.chat.completions.create(
+            response_stream = self.sync_client.chat.completions.create(
                 model=model,
                 messages=messages,
                 temperature=temperature,
@@ -600,7 +654,7 @@ class OpenAIClient:
         self,
         messages: List[Dict],
         model: str,
-        temperature: float = 0.7,
+        temperature: float = 1.0,
         tools: Optional[List[Dict]] = None,
         start_time: Optional[float] = None,
         console: Optional[Console] = None,
@@ -687,7 +741,7 @@ class OpenAIClient:
         self,
         messages: List[Dict[str, Any]],
         model: str = "gpt-4o",
-        temperature: float = 0.7,
+        temperature: float = 1.0,
         stream: bool = False,
         tools: Optional[List[Dict[str, Any]]] = None,
         tool_choice: Optional[Union[str, Dict[str, Any]]] = None,
@@ -723,7 +777,7 @@ class OpenAIClient:
                 params["tool_choice"] = tool_choice
         
         try:
-            return self._sync_client.chat.completions.create(**params)
+            return self.sync_client.chat.completions.create(**params)
         except Exception as e:
             self.logger.error(f"Error creating completion: {e}")
             raise
@@ -732,7 +786,7 @@ class OpenAIClient:
         self,
         messages: List[Dict[str, Any]],
         model: str = "gpt-4o",
-        temperature: float = 0.7,
+        temperature: float = 1.0,
         stream: bool = False,
         tools: Optional[List[Dict[str, Any]]] = None,
         tool_choice: Optional[Union[str, Dict[str, Any]]] = None,
@@ -777,7 +831,7 @@ class OpenAIClient:
         self,
         messages: List[Dict[str, Any]],
         model: str = "gpt-4o",
-        temperature: float = 0.7,
+        temperature: float = 1.0,
         tools: Optional[List[Any]] = None,
         execute_tool_fn: Optional[Callable] = None,
         stream: bool = True,
@@ -955,7 +1009,7 @@ class OpenAIClient:
         self,
         messages: List[Dict[str, Any]],
         model: str = "gpt-4o",
-        temperature: float = 0.7,
+        temperature: float = 1.0,
         tools: Optional[List[Any]] = None,
         execute_tool_fn: Optional[Callable] = None,
         stream: bool = True,
@@ -1136,7 +1190,7 @@ class OpenAIClient:
         self,
         messages: List[Dict[str, Any]],
         model: str = "gpt-4o",
-        temperature: float = 0.7,
+        temperature: float = 1.0,
         tools: Optional[List[Any]] = None,
         execute_tool_fn: Optional[Callable] = None,
         reasoning_steps: bool = False,
@@ -1173,7 +1227,7 @@ class OpenAIClient:
         while iteration_count < max_iterations:
             try:
                 # Create streaming response
-                response_stream = self._sync_client.chat.completions.create(
+                response_stream = self.sync_client.chat.completions.create(
                     model=model,
                     messages=messages,
                     temperature=temperature,
@@ -1281,7 +1335,7 @@ class OpenAIClient:
         messages: List[Dict[str, Any]],
         response_format: BaseModel,
         model: str = "gpt-4o",
-        temperature: float = 0.7,
+        temperature: float = 1.0,
         **kwargs
     ) -> Any:
         """
@@ -1298,7 +1352,7 @@ class OpenAIClient:
             Parsed response according to the response_format
         """
         try:
-            response = self._sync_client.beta.chat.completions.parse(
+            response = self.sync_client.beta.chat.completions.parse(
                 model=model,
                 messages=messages,
                 temperature=temperature,
@@ -1315,7 +1369,7 @@ class OpenAIClient:
         messages: List[Dict[str, Any]],
         response_format: BaseModel,
         model: str = "gpt-4o",
-        temperature: float = 0.7,
+        temperature: float = 1.0,
         **kwargs
     ) -> Any:
         """
@@ -1346,14 +1400,14 @@ class OpenAIClient:
     
     def close(self):
         """Close the OpenAI clients."""
-        if hasattr(self._sync_client, 'close'):
+        if self._sync_client and hasattr(self._sync_client, 'close'):
             self._sync_client.close()
         if self._async_client and hasattr(self._async_client, 'close'):
             self._async_client.close()
     
     async def aclose(self):
         """Asynchronously close the OpenAI clients."""
-        if hasattr(self._sync_client, 'close'):
+        if self._sync_client and hasattr(self._sync_client, 'close'):
             await asyncio.to_thread(self._sync_client.close)
         if self._async_client and hasattr(self._async_client, 'aclose'):
             await self._async_client.aclose()
@@ -1361,6 +1415,7 @@ class OpenAIClient:
 
 # Global client instance (similar to main.py pattern)
 _global_client = None
+_global_client_params = None
 
 def get_openai_client(api_key: Optional[str] = None, base_url: Optional[str] = None) -> OpenAIClient:
     """
@@ -1373,9 +1428,16 @@ def get_openai_client(api_key: Optional[str] = None, base_url: Optional[str] = N
     Returns:
         OpenAIClient instance
     """
-    global _global_client
+    global _global_client, _global_client_params
+    
+    # Normalize parameters for comparison
+    normalized_api_key = api_key or os.getenv("OPENAI_API_KEY")
+    normalized_base_url = base_url
+    current_params = (normalized_api_key, normalized_base_url)
     
-    if _global_client is None:
+    # Only create new client if parameters changed or first time
+    if _global_client is None or _global_client_params != current_params:
         _global_client = OpenAIClient(api_key=api_key, base_url=base_url)
+        _global_client_params = current_params
     
     return _global_client

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/mcp/mcp.py

@@ -155,7 +155,7 @@ class MCP:
         # Method 1: Using command and args separately
         agent = Agent(
             instructions="You are a helpful assistant...",
-            llm="gpt-4o-mini",
+            llm="gpt-5-nano",
             tools=MCP(
                 command="/path/to/python",
                 args=["/path/to/app.py"]
@@ -165,14 +165,14 @@ class MCP:
         # Method 2: Using a single command string
         agent = Agent(
             instructions="You are a helpful assistant...",
-            llm="gpt-4o-mini",
+            llm="gpt-5-nano",
             tools=MCP("/path/to/python /path/to/app.py")
         )
         
         # Method 3: Using an SSE endpoint
         agent = Agent(
             instructions="You are a helpful assistant...",
-            llm="gpt-4o-mini",
+            llm="gpt-5-nano",
             tools=MCP("http://localhost:8080/sse")
         )
         
@@ -514,7 +514,7 @@ class MCP:
         """Convert the MCP tool to an OpenAI-compatible tool definition.
         
         This method is specifically invoked by the Agent class when using
-        provider/model format (e.g., "openai/gpt-4o-mini").
+        provider/model format (e.g., "openai/gpt-5-nano").
         
         Returns:
             dict or list: OpenAI-compatible tool definition(s)

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/memory/memory.py

@@ -1442,7 +1442,7 @@ class Memory:
                 import litellm
                 
                 # Convert model name if it's in litellm format
-                model_name = llm or "gpt-4o-mini"
+                model_name = llm or "gpt-5-nano"
                 
                 response = litellm.completion(
                     model=model_name,
@@ -1459,7 +1459,7 @@ class Memory:
                 client = OpenAI()
                 
                 response = client.chat.completions.create(
-                    model=llm or "gpt-4o-mini",
+                    model=llm or "gpt-5-nano",
                     messages=[{
                         "role": "user", 
                         "content": custom_prompt or default_prompt

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/task/task.py

@@ -322,7 +322,7 @@ class Task:
                         if hasattr(self.agent.llm_instance, 'model'):
                             llm_model = self.agent.llm_instance.model
                         else:
-                            llm_model = "gpt-4o-mini"  # Default fallback
+                            llm_model = "gpt-5-nano"  # Default fallback
                     elif hasattr(self.agent, 'llm') and self.agent.llm:
                         # For standard model strings
                         llm_model = self.agent.llm

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents/tools/train/data/generatecot.py

@@ -27,7 +27,7 @@ class GenerateCOT:
     def __init__(
         self,
         qa_pairs: Optional[Dict[str, str]] = None,
-        model: str = "gpt-4o-mini",
+        model: str = "gpt-5-nano",
         api_key: Optional[str] = None,
         max_attempts: int = 3,
         verbose: bool = True,

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/praisonaiagents.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: praisonaiagents
-Version: 0.0.156
+Version: 0.0.158
 Summary: Praison AI agents for completing complex tasks with Self Reflection Agents
 Author: Mervin Praison
 Requires-Python: >=3.10

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "praisonaiagents"
-version = "0.0.156"
+version = "0.0.158"
 description = "Praison AI agents for completing complex tasks with Self Reflection Agents"
 requires-python = ">=3.10"
 authors = [

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/tests/test.py

@@ -26,7 +26,7 @@ planner_agent = Agent(
     goal="Create detailed story outlines with chapter breakdowns",
     backstory="Expert storyteller skilled in narrative structure and plot development",
     verbose=True,
-    llm="gpt-4o-mini",
+    llm="gpt-5-nano",
     self_reflect=False
 )
 
@@ -37,7 +37,7 @@ writer_agent = Agent(
     goal="Write engaging and cohesive story chapters",
     backstory="Experienced writer skilled in bringing stories to life with vivid details and engaging narrative",
     verbose=True,
-    llm="gpt-4o-mini",
+    llm="gpt-5-nano",
     self_reflect=False
 )
 

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/tests/test_context_agent.py

@@ -56,12 +56,12 @@ def test_basic_instantiation():
             name="Test Context Engineer",
             role="Test Role",
             goal="Test Goal",
-            llm="gpt-4o-mini"
+            llm="gpt-5-nano"
         )
         print("✅ Successfully created ContextAgent with custom parameters")
         
         # Test factory function
-        factory_agent = create_context_agent(llm="gpt-4o-mini")
+        factory_agent = create_context_agent(llm="gpt-5-nano")
         print("✅ Successfully created ContextAgent using factory function")
         
         return True, [context_agent, custom_agent, factory_agent]

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/tests/test_fix_comprehensive.py

@@ -57,7 +57,7 @@ print("=" * 60)
 
 agent_gpt4 = Agent(
     instructions="You are a helpful assistant. You can use the tools provided to you to help the user. When asked to multiply a stock price, first get the stock price, then multiply it.",
-    llm="gpt-4o-mini",
+    llm="gpt-5-nano",
     tools=[get_stock_price, multiply],
     verbose=True
 )

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/tests/test_llm_self_reflection_direct.py

@@ -9,7 +9,7 @@ def test_llm_direct():
     print("=== Testing LLM Direct with Self-Reflection and Tools ===")
     
     # Create LLM instance
-    llm = LLM(model="gpt-4o-mini")
+    llm = LLM(model="gpt-5-nano")
     
     # Test with self-reflection and tools
     try:

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/tests/test_ollama_async_fix.py

@@ -135,12 +135,12 @@ async def main():
     
     # Test sync methods
     print("\n1. Testing SYNC methods:")
-    openai_sync_success = test_model_sync("openai/gpt-4o-mini")
+    openai_sync_success = test_model_sync("openai/gpt-5-nano")
     ollama_sync_success = test_model_sync("ollama/llama3.2")
     
     # Test async methods
     print("\n2. Testing ASYNC methods:")
-    openai_async_success = await test_model_async("openai/gpt-4o-mini")
+    openai_async_success = await test_model_async("openai/gpt-5-nano")
     ollama_async_success = await test_model_async("ollama/llama3.2")
     
     # Summary

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/tests/test_ollama_fix.py

@@ -82,7 +82,7 @@ if __name__ == "__main__":
     
     # Test with OpenAI first (as baseline)
     print("\n1. Testing with OpenAI (baseline):")
-    openai_success = test_model("openai/gpt-4o-mini")
+    openai_success = test_model("openai/gpt-5-nano")
     
     # Test with Ollama
     print("\n2. Testing with Ollama:")

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/tests/test_ollama_sequential_fix.py

@@ -103,7 +103,7 @@ def test_provider_detection():
     print("✅ Ollama prefix detection works")
     
     # Test non-Ollama provider
-    openai_llm = LLM(model="gpt-4o-mini")
+    openai_llm = LLM(model="gpt-5-nano")
     assert not openai_llm._is_ollama_provider(), "Should not detect OpenAI as Ollama"
     print("✅ Non-Ollama provider detection works")
     

{praisonaiagents-0.0.156 → praisonaiagents-0.0.158}/tests/test_self_reflection_fix_simple.py

@@ -14,7 +14,7 @@ def test_self_reflection_fix():
         goal="Solve math problems accurately",
         backstory="You are a helpful math assistant",
         self_reflect=True,
-        llm="gpt-4o-mini",  # Use a more widely available model
+        llm="gpt-5-nano",  # Use a more widely available model
         verbose=True,
         tools=[calculator],
         min_reflect=1,