praisonaiagents 0.0.88__tar.gz → 0.0.90__tar.gz

{praisonaiagents-0.0.88 → praisonaiagents-0.0.90}/PKG-INFO

@@ -1,8 +1,9 @@
 Metadata-Version: 2.4
 Name: praisonaiagents
-Version: 0.0.88
+Version: 0.0.90
 Summary: Praison AI agents for completing complex tasks with Self Reflection Agents
 Author: Mervin Praison
+Requires-Python: >=3.10
 Requires-Dist: pydantic
 Requires-Dist: rich
 Requires-Dist: openai

{praisonaiagents-0.0.88 → praisonaiagents-0.0.90}/praisonaiagents/agents/autoagents.py

@@ -8,10 +8,10 @@ It automatically handles agent creation, task setup, and execution flow.
 from .agents import PraisonAIAgents
 from ..agent.agent import Agent
 from ..task.task import Task
-from typing import List, Any, Optional, Dict
+from typing import List, Any, Optional, Dict, Tuple
 import logging
 import os
-from pydantic import BaseModel
+from pydantic import BaseModel, ConfigDict
 from ..main import display_instruction, display_tool_call, display_interaction, client
 
 # Define Pydantic models for structured output
@@ -22,6 +22,7 @@ class TaskConfig(BaseModel):
     tools: List[str]
 
 class AgentConfig(BaseModel):
+    model_config = ConfigDict(arbitrary_types_allowed=True)
     name: str
     role: str
     goal: str
@@ -30,6 +31,7 @@ class AgentConfig(BaseModel):
     tasks: List[TaskConfig]
 
 class AutoAgentsConfig(BaseModel):
+    model_config = ConfigDict(arbitrary_types_allowed=True)
     main_instruction: str
     process_type: str
     agents: List[AgentConfig]
@@ -255,7 +257,7 @@ Return the configuration in a structured JSON format matching the AutoAgentsConf
             logging.error(f"Error generating configuration: {e}")
             raise
 
-    def _create_agents_and_tasks(self, config: AutoAgentsConfig) -> tuple[List[Agent], List[Task]]:
+    def _create_agents_and_tasks(self, config: AutoAgentsConfig) -> Tuple[List[Agent], List[Task]]:
         """Create agents and tasks from configuration"""
         agents = []
         tasks = []
@@ -283,7 +285,7 @@ Return the configuration in a structured JSON format matching the AutoAgentsConf
                 respect_context_window=self.respect_context_window,
                 code_execution_mode=self.code_execution_mode,
                 embedder_config=self.embedder_config,
-                knowledge_sources=self.knowledge_sources,
+                knowledge=self.knowledge_sources,
                 use_system_prompt=self.use_system_prompt,
                 cache=self.cache,
                 allow_delegation=self.allow_delegation,

{praisonaiagents-0.0.88 → praisonaiagents-0.0.90}/praisonaiagents/main.py

@@ -4,7 +4,7 @@ import json
 import logging
 from typing import List, Optional, Dict, Any, Union, Literal, Type
 from openai import OpenAI
-from pydantic import BaseModel
+from pydantic import BaseModel, ConfigDict
 from rich import print
 from rich.console import Console
 from rich.panel import Panel
@@ -365,6 +365,7 @@ class ReflectionOutput(BaseModel):
 client = OpenAI(api_key=(os.environ["OPENAI_API_KEY"] if os.environ.get("OPENAI_API_KEY") else "xxxx"))
 
 class TaskOutput(BaseModel):
+    model_config = ConfigDict(arbitrary_types_allowed=True)
     description: str
     summary: Optional[str] = None
     raw: str

{praisonaiagents-0.0.88 → praisonaiagents-0.0.90}/praisonaiagents/process/process.py

@@ -1,7 +1,7 @@
 import logging
 import asyncio
 from typing import Dict, Optional, List, Any, AsyncGenerator
-from pydantic import BaseModel
+from pydantic import BaseModel, ConfigDict
 from ..agent.agent import Agent
 from ..task.task import Task
 from ..main import display_error, client
@@ -9,6 +9,7 @@ import csv
 import os
 
 class LoopItems(BaseModel):
+    model_config = ConfigDict(arbitrary_types_allowed=True)
     items: List[Any]
 
 class Process:
@@ -30,6 +31,44 @@ class Process:
         self.task_retry_counter: Dict[str, int] = {} # Initialize retry counter
         self.workflow_finished = False # ADDED: Workflow finished flag
 
+    def _build_task_context(self, current_task: Task) -> str:
+        """Build context for a task based on its retain_full_context setting"""
+        if not (current_task.previous_tasks or current_task.context):
+            return ""
+            
+        context = "\nInput data from previous tasks:"
+        
+        if current_task.retain_full_context:
+            # Original behavior: include all previous tasks
+            for prev_name in current_task.previous_tasks:
+                prev_task = next((t for t in self.tasks.values() if t.name == prev_name), None)
+                if prev_task and prev_task.result:
+                    context += f"\n{prev_name}: {prev_task.result.raw}"
+                    
+            # Add data from context tasks
+            if current_task.context:
+                for ctx_task in current_task.context:
+                    if ctx_task.result and ctx_task.name != current_task.name:
+                        context += f"\n{ctx_task.name}: {ctx_task.result.raw}"
+        else:
+            # New behavior: only include the most recent previous task
+            if current_task.previous_tasks:
+                # Get the most recent previous task (last in the list)
+                prev_name = current_task.previous_tasks[-1]
+                prev_task = next((t for t in self.tasks.values() if t.name == prev_name), None)
+                if prev_task and prev_task.result:
+                    context += f"\n{prev_name}: {prev_task.result.raw}"
+                    
+            # For context tasks, still include the most recent one
+            if current_task.context:
+                # Get the most recent context task with a result
+                for ctx_task in reversed(current_task.context):
+                    if ctx_task.result and ctx_task.name != current_task.name:
+                        context += f"\n{ctx_task.name}: {ctx_task.result.raw}"
+                        break  # Only include the most recent one
+                        
+        return context
+
     def _find_next_not_started_task(self) -> Optional[Task]:
         """Fallback mechanism to find the next 'not started' task."""
         fallback_attempts = 0
@@ -147,25 +186,8 @@ Description length: {len(current_task.description)}
             """)
 
             # Add context from previous tasks to description
-            if current_task.previous_tasks or current_task.context:
-                context = "\nInput data from previous tasks:"
-
-                # Add data from previous tasks in workflow
-                for prev_name in current_task.previous_tasks:
-                    prev_task = next((t for t in self.tasks.values() if t.name == prev_name), None)
-                    if prev_task and prev_task.result:
-                        # Handle loop data
-                        if current_task.task_type == "loop":
-                            context += f"\n{prev_name}: {prev_task.result.raw}"
-                        else:
-                            context += f"\n{prev_name}: {prev_task.result.raw}"
-
-                # Add data from context tasks
-                if current_task.context:
-                    for ctx_task in current_task.context:
-                        if ctx_task.result and ctx_task.name != current_task.name:
-                            context += f"\n{ctx_task.name}: {ctx_task.result.raw}"
-
+            context = self._build_task_context(current_task)
+            if context:
                 # Update task description with context
                 current_task.description = current_task.description + context
 
@@ -778,25 +800,8 @@ Description length: {len(current_task.description)}
             """)
 
             # Add context from previous tasks to description
-            if current_task.previous_tasks or current_task.context:
-                context = "\nInput data from previous tasks:"
-
-                # Add data from previous tasks in workflow
-                for prev_name in current_task.previous_tasks:
-                    prev_task = next((t for t in self.tasks.values() if t.name == prev_name), None)
-                    if prev_task and prev_task.result:
-                        # Handle loop data
-                        if current_task.task_type == "loop":
-                            context += f"\n{prev_name}: {prev_task.result.raw}"
-                        else:
-                            context += f"\n{prev_name}: {prev_task.result.raw}"
-
-                # Add data from context tasks
-                if current_task.context:
-                    for ctx_task in current_task.context:
-                        if ctx_task.result and ctx_task.name != current_task.name:
-                            context += f"\n{ctx_task.name}: {ctx_task.result.raw}"
-
+            context = self._build_task_context(current_task)
+            if context:
                 # Update task description with context
                 current_task.description = current_task.description + context
 

{praisonaiagents-0.0.88 → praisonaiagents-0.0.90}/praisonaiagents/task/task.py

@@ -39,7 +39,8 @@ class Task:
         memory=None,
         quality_check=True,
         input_file: Optional[str] = None,
-        rerun: bool = False # Renamed from can_rerun and logic inverted, default True for backward compatibility
+        rerun: bool = False, # Renamed from can_rerun and logic inverted, default True for backward compatibility
+        retain_full_context: bool = False # By default, only use previous task output, not all previous tasks
     ):
         # Add check if memory config is provided
         if memory is not None or (config and config.get('memory_config')):
@@ -78,6 +79,7 @@ class Task:
         self.memory = memory
         self.quality_check = quality_check
         self.rerun = rerun # Assigning the rerun parameter
+        self.retain_full_context = retain_full_context
 
         # Set logger level based on config verbose level
         verbose = self.config.get("verbose", 0)
@@ -213,9 +215,20 @@ class Task:
                 logger.info(f"Task {self.id}: Calculating quality metrics for output: {task_output.raw[:100]}...")
 
                 # Get quality metrics from LLM
+                # Determine which LLM model to use based on agent configuration
+                llm_model = None
+                if self.agent:
+                    if getattr(self.agent, '_using_custom_llm', False) and hasattr(self.agent, 'llm_instance'):
+                        # For custom LLM instances (like Ollama)
+                        llm_model = self.agent.llm_instance
+                    elif hasattr(self.agent, 'llm') and self.agent.llm:
+                        # For standard model strings
+                        llm_model = self.agent.llm
+                
                 metrics = self.memory.calculate_quality_metrics(
                     task_output.raw,
-                    self.expected_output
+                    self.expected_output,
+                    llm=llm_model
                 )
                 logger.info(f"Task {self.id}: Quality metrics calculated: {metrics}")
 

{praisonaiagents-0.0.88 → praisonaiagents-0.0.90}/praisonaiagents.egg-info/PKG-INFO

@@ -1,8 +1,9 @@
 Metadata-Version: 2.4
 Name: praisonaiagents
-Version: 0.0.88
+Version: 0.0.90
 Summary: Praison AI agents for completing complex tasks with Self Reflection Agents
 Author: Mervin Praison
+Requires-Python: >=3.10
 Requires-Dist: pydantic
 Requires-Dist: rich
 Requires-Dist: openai

{praisonaiagents-0.0.88 → praisonaiagents-0.0.90}/praisonaiagents.egg-info/SOURCES.txt

@@ -46,4 +46,5 @@ praisonaiagents/tools/wikipedia_tools.py
 praisonaiagents/tools/xml_tools.py
 praisonaiagents/tools/yaml_tools.py
 praisonaiagents/tools/yfinance_tools.py
-praisonaiagents/tools/train/data/generatecot.py
+praisonaiagents/tools/train/data/generatecot.py
+tests/test.py

{praisonaiagents-0.0.88 → praisonaiagents-0.0.90}/pyproject.toml

@@ -4,8 +4,9 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "praisonaiagents"
-version = "0.0.88"
+version = "0.0.90"
 description = "Praison AI agents for completing complex tasks with Self Reflection Agents"
+requires-python = ">=3.10"
 authors = [
     { name="Mervin Praison" }
 ]
@@ -53,4 +54,7 @@ all = [
     "praisonaiagents[llm]",
     "praisonaiagents[mcp]",
     "praisonaiagents[api]"
-]
+]
+
+[tool.setuptools]
+packages = ["praisonaiagents"]

praisonaiagents-0.0.90/tests/test.py

@@ -0,0 +1,147 @@
+from typing import List, Dict
+from pydantic import BaseModel
+from praisonaiagents import Agent, Task, PraisonAIAgents
+from praisonaiagents.main import TaskOutput
+
+# Pydantic models for structured output
+class Chapter(BaseModel):
+    chapter_number: int
+    title: str
+    outline: str
+
+class StoryOutline(BaseModel):
+    title: str
+    total_chapters: int
+    chapters: List[Chapter]
+
+class ChapterContent(BaseModel):
+    chapter_number: int
+    title: str
+    content: str
+
+# Create the story planner agent
+planner_agent = Agent(
+    name="Story Planner",
+    role="Creative Writing Architect",
+    goal="Create detailed story outlines with chapter breakdowns",
+    backstory="Expert storyteller skilled in narrative structure and plot development",
+    verbose=True,
+    llm="gpt-4o-mini",
+    self_reflect=False
+)
+
+# Create the writer agent
+writer_agent = Agent(
+    name="Story Writer",
+    role="Creative Writer",
+    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",
+    self_reflect=False
+)
+
+# Story title
+story_title = "The Curious Cat"
+
+# Step 1: Create outline task
+outline_task = Task(
+    name="story_outline",
+    description=f"""Create a detailed story outline for a story titled '{story_title}'.
+    The outline should include:
+    1. A compelling story arc
+    2. Clear chapter breakdown
+    3. Brief description of what happens in each chapter
+    4. Character introductions and development points
+    
+    Provide the response in this exact JSON format:
+    {{
+        "title": "Story title",
+        "total_chapters": number,
+        "chapters": [
+            {{
+                "chapter_number": 1,
+                "title": "Chapter title",
+                "outline": "Detailed chapter outline"
+            }}
+        ]
+    }}
+    """,
+    expected_output="Complete story outline",
+    agent=planner_agent,
+    output_pydantic=StoryOutline
+)
+
+# Create initial workflow for outline
+outline_workflow = PraisonAIAgents(
+    agents=[planner_agent],
+    tasks=[outline_task],
+    verbose=True,
+    process="sequential"
+)
+
+# Get the outline
+outline_result = outline_workflow.start()
+story_outline = outline_result["task_results"][0].pydantic
+
+# Step 2: Create tasks for each chapter
+chapter_tasks = []
+chapters_content = []
+
+# Create a task for each chapter
+for chapter in story_outline.chapters:
+    chapter_task = Task(
+        name=f"write_chapter_{chapter.chapter_number}",
+        description=f"""Write chapter {chapter.chapter_number}: {chapter.title}
+        Chapter outline: {chapter.outline}
+        
+        Write a complete, engaging chapter following this outline.
+        Return in this exact JSON format:
+        {{
+            "chapter_number": {chapter.chapter_number},
+            "title": "{chapter.title}",
+            "content": "The full chapter content..."
+        }}
+        """,
+        expected_output="Complete chapter content",
+        agent=writer_agent,
+        output_pydantic=ChapterContent,
+        context=[outline_task]
+    )
+    chapter_tasks.append(chapter_task)
+
+# Create sequential workflow for chapters
+chapter_workflow = PraisonAIAgents(
+    agents=[writer_agent],
+    tasks=chapter_tasks,
+    verbose=True,
+    process="sequential"
+)
+
+# Run the chapter workflow
+chapter_results = chapter_workflow.start()
+
+# Collect and validate all chapter contents
+for task_result in chapter_results["task_results"].values():
+    if task_result and task_result.pydantic:
+        chapters_content.append(task_result.pydantic)
+
+# Sort chapters by chapter number
+chapters_content.sort(key=lambda x: x.chapter_number)
+
+# Format final result
+story = {
+    "title": story_outline.title,
+    "outline": story_outline.model_dump(),
+    "chapters": [chapter.model_dump() for chapter in chapters_content]
+}
+
+# Print results
+print("\n=== Story Generation Results ===")
+print(f"\nTitle: {story['title']}")
+print("\nOutline:")
+print(story['outline'])
+print("\nChapters:")
+for chapter in story['chapters']:
+    print(f"\nChapter {chapter['chapter_number']}: {chapter['title']}")
+    print(chapter['content'][:200] + "...")  # Print first 200 chars of each chapter