Enhancing AI workflow

.flet

@@ -1,59 +0,0 @@
-# Flet Build Configuration
-# This file provides additional build settings for the Flet application
-
-# App metadata
-[app]
-name = "GitHub Pulse"
-description = "GitHub automation workflows with AI assistance"
-author = "TySP-Dev"
-
-# Build settings
-[build]
-# Include these files/directories in the build
-include_packages = [
-    "app_components",
-    "assets"
-]
-
-# Exclude unnecessary files from build
-exclude = [
-    "*.pyc",
-    "__pycache__",
-    "*.pyo",
-    "*.pyd",
-    ".git",
-    ".gitignore",
-    "venv",
-    ".env",
-    "*.md",
-    "BUILD.md",
-    "SETUP.md",
-    "*.example"
-]
-
-# Asset optimization
-[assets]
-# Optimize images during build
-optimize_images = true
-# Include app_components/assets directory
-directories = [
-    "assets",
-    "app_components/assets"
-]
-
-# Platform-specific settings
-[android]
-adaptive_icon_background = "#1976D2"
-adaptive_icon_foreground = "assets/icon.png"
-
-[ios]
-info_plist_version = "1.0"
-
-[macos]
-info_plist_version = "1.0"
-
-[windows]
-console = false  # Hide console window in production
-
-[linux]
-categories = ["Development", "Utility"]

.gitignore

@@ -187,6 +187,9 @@ linux/
 macos/
 windows/
 web/
+.flet
+build/
+pyproject.toml
 
 # Configuration (generated during build, not in git)
 # Note: pyproject.toml is now tracked for proper builds

src/app_components/__init__.py

@@ -30,6 +30,7 @@ from .settings_dialog import SettingsDialog
 from .main_gui import MainGUI
 from .utils import Logger, PRNumberManager, ContentBuilders
 from .workflow import WorkflowManager, WorkflowItem, GitHubRepoFetcher
+from .ai_action_planner import AIActionPlanner, ActionPlan
 
 __all__ = [
     'ConfigManager',
@@ -43,6 +44,8 @@ __all__ = [
     'WorkflowManager',
     'WorkflowItem',
     'GitHubRepoFetcher',
+    'AIActionPlanner',
+    'ActionPlan',
     '__version__',
     '__author__',
     '__app_name__',

src/app_components/ai_action_planner.py

@@ -0,0 +1,617 @@
+"""
+AI Action Planner
+Generates and executes action plans for GitHub issues and PRs using AI
+"""
+
+import json
+import re
+import requests
+from typing import List, Dict, Any, Optional, Callable
+from pathlib import Path
+
+
+class ActionPlan:
+    """Represents an AI-generated action plan"""
+
+    def __init__(self, title: str, steps: List[Dict[str, Any]], context: Dict[str, Any]):
+        self.title = title
+        self.steps = steps  # List of {description, file_path, changes, completed}
+        self.context = context  # PR/Issue context
+        self.completed_steps = []
+        self.failed_steps = []
+
+    def to_dict(self) -> Dict[str, Any]:
+        """Convert plan to dictionary"""
+        return {
+            'title': self.title,
+            'steps': self.steps,
+            'context': self.context,
+            'completed_steps': self.completed_steps,
+            'failed_steps': self.failed_steps
+        }
+
+    @classmethod
+    def from_dict(cls, data: Dict[str, Any]) -> 'ActionPlan':
+        """Create plan from dictionary"""
+        plan = cls(data['title'], data['steps'], data['context'])
+        plan.completed_steps = data.get('completed_steps', [])
+        plan.failed_steps = data.get('failed_steps', [])
+        return plan
+
+
+class OllamaProvider:
+    """Simple Ollama API provider for AI action planning"""
+
+    def __init__(self, base_url: str, model: str, logger):
+        self.base_url = base_url.rstrip('/')
+        self.model = model
+        self.logger = logger
+
+    def generate(self, prompt: str) -> Optional[str]:
+        """Generate a response from Ollama"""
+        try:
+            response = requests.post(
+                f"{self.base_url}/api/generate",
+                json={
+                    "model": self.model,
+                    "prompt": prompt,
+                    "stream": False
+                },
+                timeout=120
+            )
+            response.raise_for_status()
+            result = response.json()
+            return result.get('response', '')
+        except Exception as e:
+            self.logger.log(f"❌ Ollama API error: {str(e)}")
+            return None
+
+    def make_change(self, file_content: str, old_text: str, new_text: str, file_path: str, custom_instructions: str = None) -> Optional[str]:
+        """Make changes to file content using Ollama"""
+        # Try direct replacement first
+        if old_text and old_text.strip() in file_content:
+            return file_content.replace(old_text.strip(), new_text.strip())
+
+        # Use Ollama to make intelligent changes
+        prompt = f"""You are a code modification assistant. Modify the following file according to the instructions.
+
+File: {file_path}
+
+Current Content:
+```
+{file_content}
+```
+
+Instructions: {new_text}
+{f'Additional context: {custom_instructions}' if custom_instructions else ''}
+
+Return ONLY the complete modified file content. Do not include explanations or markdown code blocks."""
+
+        return self.generate(prompt)
+
+
+class AIActionPlanner:
+    """Generates and executes action plans using AI"""
+
+    def __init__(self, ai_manager, logger, config_manager):
+        self.ai_manager = ai_manager
+        self.logger = logger
+        self.config_manager = config_manager
+
+    def generate_plan(self, item, custom_instructions: str = "") -> Optional[ActionPlan]:
+        """
+        Generate an action plan for a PR or Issue
+
+        Args:
+            item: The PR or Issue (WorkflowItem object or dict)
+            custom_instructions: Optional user-provided instructions
+
+        Returns:
+            ActionPlan object or None if generation failed
+        """
+        # Handle both WorkflowItem objects and dictionaries
+        if hasattr(item, 'item_type'):
+            # It's a WorkflowItem object
+            item_type = item.item_type
+            item_number = item.number
+            title = item.title
+            body = item.body or ''
+            repo = getattr(item, 'repo', None)
+        else:
+            # It's a dictionary
+            item_type = item.get('type', 'unknown')
+            item_number = item.get('number')
+            title = item.get('title', 'Untitled')
+            body = item.get('body', '')
+            repo = item.get('repo')
+
+        self.logger.log(f"🤖 Generating action plan for {item_type} #{item_number}...")
+
+        # Get AI provider
+        config = self.config_manager.get_config()
+        ai_provider_name = config.get('AI_PROVIDER', 'none').lower()
+
+        if ai_provider_name == 'none' or not ai_provider_name:
+            self.logger.log("❌ No AI provider configured. Please configure in Settings.")
+            return None
+
+        # Get provider instance
+        provider = self._get_ai_provider(ai_provider_name, config)
+        if not provider:
+            return None
+
+        # Generate the plan using AI
+        try:
+            self.logger.log(f"📤 Calling AI provider: {type(provider).__name__}")
+            plan_text = self._call_ai_for_plan(provider, item_type, title, body, custom_instructions)
+
+            if not plan_text:
+                self.logger.log("❌ AI did not generate a plan (empty response)")
+                return None
+
+            self.logger.log(f"📥 Received response from AI ({len(plan_text)} characters)")
+            self.logger.log(f"📄 Response preview: {plan_text[:200]}...")
+
+            # Parse the plan
+            self.logger.log("🔍 Parsing AI response into steps...")
+            steps = self._parse_plan(plan_text)
+
+            if not steps:
+                self.logger.log("❌ Could not parse action steps from AI response")
+                return None
+
+            # Get repo from item or config
+            if repo is None:
+                repo = config.get('GITHUB_REPO', '')
+
+            plan = ActionPlan(
+                title=f"Action Plan for {item_type.upper()} #{item_number}: {title}",
+                steps=steps,
+                context={
+                    'item_type': item_type,
+                    'item_number': item_number,
+                    'item_title': title,
+                    'item_body': body,
+                    'repo': repo
+                }
+            )
+
+            self.logger.log(f"✅ Generated plan with {len(steps)} steps")
+            return plan
+
+        except Exception as e:
+            self.logger.log(f"❌ Error generating plan: {str(e)}")
+            return None
+
+    def _get_ai_provider(self, provider_name: str, config: Dict[str, Any]):
+        """Get the AI provider instance"""
+        try:
+            if provider_name in ['claude', 'anthropic']:
+                # Try both CLAUDE_API_KEY and ANTHROPIC_API_KEY for compatibility
+                api_key = config.get('CLAUDE_API_KEY')
+                if not api_key:
+                    api_key = config.get('ANTHROPIC_API_KEY')
+                if not api_key:
+                    self.logger.log("❌ Claude API key not found in secure storage (tried both CLAUDE_API_KEY and ANTHROPIC_API_KEY)")
+                    return None
+                self.logger.log("ℹ️  Initializing Claude provider...")
+                from . import ai_manager
+                provider = ai_manager.ClaudeProvider(api_key, self.logger)
+                self.logger.log("✅ Claude provider initialized successfully")
+                return provider
+
+            elif provider_name in ['chatgpt', 'openai']:
+                api_key = config.get('OPENAI_API_KEY')
+                if not api_key:
+                    self.logger.log("❌ OpenAI API key not found in secure storage")
+                    return None
+                self.logger.log("ℹ️  Initializing ChatGPT provider...")
+                from . import ai_manager
+                provider = ai_manager.ChatGPTProvider(api_key, self.logger)
+                self.logger.log("✅ ChatGPT provider initialized successfully")
+                return provider
+
+            elif provider_name == 'ollama':
+                # Ollama doesn't need an API key, uses URL from config
+                ollama_url = config.get('OLLAMA_URL', 'http://localhost:11434')
+                ollama_model = config.get('OLLAMA_MODEL', 'llama2')
+                self.logger.log(f"ℹ️  Using Ollama at {ollama_url} with model {ollama_model}")
+                # Create a simple Ollama provider wrapper
+                return OllamaProvider(ollama_url, ollama_model, self.logger)
+
+            else:
+                self.logger.log(f"❌ Unsupported AI provider: {provider_name}")
+                return None
+
+        except Exception as e:
+            self.logger.log(f"❌ Error creating AI provider: {str(e)}")
+            return None
+
+    def _call_ai_for_plan(self, provider, item_type: str, title: str, body: str, custom_instructions: str) -> Optional[str]:
+        """Call AI to generate an action plan"""
+
+        prompt = f"""You are an expert software engineer tasked with creating an actionable plan to address a GitHub {item_type}.
+
+{item_type.upper()} Title: {title}
+
+{item_type.upper()} Description:
+{body}
+
+{"Additional Instructions: " + custom_instructions if custom_instructions else ""}
+
+Please create a detailed action plan with specific, executable steps. For each step, specify:
+1. What needs to be done (clear description)
+2. Which file(s) need to be modified (if applicable)
+3. What changes should be made (if applicable)
+
+Format your response as a JSON array of steps, where each step has:
+- "description": A clear description of what to do
+- "file_path": Path to the file to modify (or null if not file-specific)
+- "changes": Description of changes to make (or null if not applicable)
+- "action_type": One of ["modify_file", "create_file", "delete_file", "investigate", "test", "document"]
+
+Example format:
+```json
+[
+  {{
+    "description": "Fix the authentication bug in login handler",
+    "file_path": "src/auth/login.py",
+    "changes": "Update the password validation logic to handle special characters correctly",
+    "action_type": "modify_file"
+  }},
+  {{
+    "description": "Add unit tests for authentication",
+    "file_path": "tests/test_auth.py",
+    "changes": "Add test cases for special characters in passwords",
+    "action_type": "create_file"
+  }}
+]
+```
+
+IMPORTANT: Return ONLY the JSON array, no other text before or after."""
+
+        try:
+            if isinstance(provider, OllamaProvider):
+                # Use Ollama
+                self.logger.log(f"🤖 Calling Ollama AI to generate plan...")
+                return provider.generate(prompt)
+
+            elif hasattr(provider, '_generate_updated_document'):
+                # Use Claude's document generation
+                self.logger.log(f"🤖 Calling Claude AI to generate plan...")
+                import anthropic
+                client = anthropic.Anthropic(api_key=provider.api_key)
+
+                message = client.messages.create(
+                    model="claude-sonnet-4-5",
+                    max_tokens=4096,
+                    messages=[{"role": "user", "content": prompt}]
+                )
+
+                return message.content[0].text
+
+            elif hasattr(provider, 'client'):
+                # Use OpenAI/ChatGPT
+                self.logger.log(f"🤖 Calling ChatGPT AI to generate plan...")
+                response = provider.client.chat.completions.create(
+                    model="gpt-4",
+                    messages=[{"role": "user", "content": prompt}],
+                    max_tokens=4096
+                )
+
+                self.logger.log(f"✅ ChatGPT response received")
+                return response.choices[0].message.content
+
+            else:
+                self.logger.log(f"❌ Unknown provider type: {type(provider).__name__}")
+                return None
+
+        except Exception as e:
+            self.logger.log(f"❌ AI API call failed: {str(e)}")
+            import traceback
+            self.logger.log(f"❌ Traceback: {traceback.format_exc()}")
+            return None
+
+    def _parse_plan(self, plan_text: str) -> List[Dict[str, Any]]:
+        """Parse the AI-generated plan text into structured steps"""
+
+        try:
+            # Extract JSON from response (might be wrapped in markdown)
+            json_match = re.search(r'```json\s*(\[.*?\])\s*```', plan_text, re.DOTALL)
+            if json_match:
+                json_text = json_match.group(1)
+            else:
+                # Try to find JSON array directly
+                json_match = re.search(r'\[.*\]', plan_text, re.DOTALL)
+                if json_match:
+                    json_text = json_match.group(0)
+                else:
+                    self.logger.log("⚠️  Could not find JSON in AI response")
+                    return []
+
+            # Parse JSON
+            steps = json.loads(json_text)
+
+            # Validate and clean up steps
+            validated_steps = []
+            for i, step in enumerate(steps):
+                if isinstance(step, dict):
+                    validated_step = {
+                        'step_number': i + 1,
+                        'description': step.get('description', f'Step {i+1}'),
+                        'file_path': step.get('file_path'),
+                        'changes': step.get('changes'),
+                        'action_type': step.get('action_type', 'investigate'),
+                        'completed': False,
+                        'status': 'pending'
+                    }
+                    validated_steps.append(validated_step)
+
+            self.logger.log(f"✅ Successfully parsed {len(validated_steps)} steps from AI response")
+            return validated_steps
+
+        except json.JSONDecodeError as e:
+            self.logger.log(f"❌ Failed to parse JSON: {str(e)}")
+            self.logger.log(f"Response was: {plan_text[:500]}...")
+            return []
+        except Exception as e:
+            self.logger.log(f"❌ Error parsing plan: {str(e)}")
+            return []
+
+    def execute_plan(
+        self,
+        plan: ActionPlan,
+        local_repo_path: str,
+        progress_callback: Optional[Callable[[int, int, str], None]] = None,
+        log_callback: Optional[Callable[[str], None]] = None
+    ) -> Dict[str, Any]:
+        """
+        Execute an action plan
+
+        Args:
+            plan: The ActionPlan to execute
+            local_repo_path: Path to local git repository
+            progress_callback: Callback function(current_step, total_steps, message)
+            log_callback: Callback function for logging thought process
+
+        Returns:
+            Dictionary with execution results
+        """
+        def log(message):
+            """Helper to log to both logger and callback"""
+            self.logger.log(message)
+            if log_callback:
+                log_callback(message)
+
+        log(f"▶️  Starting execution of plan: {plan.title}")
+
+        if not local_repo_path or not Path(local_repo_path).exists():
+            log(f"❌ Local repository path not found: {local_repo_path}")
+            return {'success': False, 'error': 'Invalid local repository path'}
+
+        total_steps = len(plan.steps)
+        completed = 0
+        failed = 0
+
+        for i, step in enumerate(plan.steps):
+            step_num = step['step_number']
+
+            # Mark step as in-progress
+            step['status'] = 'in_progress'
+            if progress_callback:
+                progress_callback(i + 1, total_steps, f"Executing step {step_num}...")
+
+            log(f"\n📍 Step {step_num}/{total_steps}: {step['description']}")
+
+            try:
+                result = self._execute_step(step, local_repo_path, plan.context, log)
+
+                if result['success']:
+                    step['completed'] = True
+                    step['status'] = 'completed'
+                    plan.completed_steps.append(step_num)
+                    completed += 1
+                    log(f"✅ Step {step_num} completed")
+                else:
+                    step['status'] = 'failed'
+                    step['error'] = result.get('error', 'Unknown error')
+                    plan.failed_steps.append(step_num)
+                    failed += 1
+                    log(f"❌ Step {step_num} failed: {result.get('error')}")
+
+            except Exception as e:
+                step['status'] = 'failed'
+                step['error'] = str(e)
+                plan.failed_steps.append(step_num)
+                failed += 1
+                log(f"❌ Step {step_num} failed with exception: {str(e)}")
+
+        log(f"\n📊 Execution complete: {completed}/{total_steps} steps successful, {failed} failed")
+
+        # If we made changes successfully, commit and push them
+        if completed > 0:
+            try:
+                log("\n🔧 Committing and pushing changes...")
+
+                # Get PR/Issue info from context
+                item_type = plan.context.get('item_type', 'item')
+                item_number = plan.context.get('item_number', 'unknown')
+                item_title = plan.context.get('item_title', 'changes')
+
+                # Commit message
+                commit_msg = f"AI: Execute action plan for {item_type} #{item_number}\n\n{item_title}\n\nAutomated changes by GitHub Pulse AI"
+
+                # Get current branch (should be the PR branch)
+                import subprocess
+                result = subprocess.run(
+                    ['git', 'rev-parse', '--abbrev-ref', 'HEAD'],
+                    cwd=local_repo_path,
+                    capture_output=True,
+                    text=True,
+                    timeout=10
+                )
+                current_branch = result.stdout.strip() if result.returncode == 0 else 'main'
+                log(f"📍 Current branch: {current_branch}")
+
+                # Stage all changes
+                log("📝 Staging changes...")
+                subprocess.run(['git', 'add', '-A'], cwd=local_repo_path, check=True, timeout=10)
+
+                # Check if there are changes to commit
+                result = subprocess.run(
+                    ['git', 'diff', '--cached', '--quiet'],
+                    cwd=local_repo_path,
+                    timeout=10
+                )
+
+                if result.returncode != 0:  # There are changes
+                    # Commit
+                    log("💾 Committing changes...")
+                    subprocess.run(
+                        ['git', 'commit', '-m', commit_msg],
+                        cwd=local_repo_path,
+                        check=True,
+                        timeout=10
+                    )
+
+                    # Push
+                    log(f"🚀 Pushing to {current_branch}...")
+                    subprocess.run(
+                        ['git', 'push', 'origin', current_branch],
+                        cwd=local_repo_path,
+                        check=True,
+                        timeout=30
+                    )
+                    log(f"✅ Changes pushed to {current_branch}")
+                else:
+                    log("ℹ️  No changes to commit")
+
+            except subprocess.TimeoutExpired:
+                log("⚠️  Git operation timed out")
+            except subprocess.CalledProcessError as e:
+                log(f"⚠️  Git operation failed: {e}")
+            except Exception as e:
+                log(f"⚠️  Error during git commit/push: {str(e)}")
+
+        return {
+            'success': failed == 0,
+            'completed': completed,
+            'failed': failed,
+            'total': total_steps,
+            'plan': plan
+        }
+
+    def _execute_step(self, step: Dict[str, Any], local_repo_path: str, context: Dict[str, Any], log=None) -> Dict[str, Any]:
+        """Execute a single step of the plan"""
+
+        action_type = step.get('action_type', 'investigate')
+        file_path = step.get('file_path')
+        changes = step.get('changes')
+
+        # Use log function if provided, otherwise fall back to logger
+        log_func = log if log else self.logger.log
+
+        if action_type == 'modify_file' and file_path:
+            return self._modify_file(file_path, changes, local_repo_path, log_func)
+
+        elif action_type == 'create_file' and file_path:
+            return self._create_file(file_path, changes, local_repo_path, log_func)
+
+        elif action_type == 'delete_file' and file_path:
+            return self._delete_file(file_path, local_repo_path, log_func)
+
+        else:
+            # For investigate, test, document actions, just mark as completed
+            # (requires manual intervention)
+            log_func(f"ℹ️  Manual action required: {step['description']}")
+            return {'success': True, 'message': 'Manual action logged'}
+
+    def _modify_file(self, file_path: str, changes: str, local_repo_path: str, log=None) -> Dict[str, Any]:
+        """Modify a file using AI"""
+
+        log_func = log if log else self.logger.log
+        full_path = Path(local_repo_path) / file_path
+
+        if not full_path.exists():
+            return {'success': False, 'error': f'File not found: {file_path}'}
+
+        try:
+            log_func(f"📝 Reading file: {file_path}")
+            # Read current content
+            with open(full_path, 'r', encoding='utf-8') as f:
+                current_content = f.read()
+
+            # Get AI provider to make changes
+            config = self.config_manager.get_config()
+            provider_name = config.get('AI_PROVIDER', 'none').lower()
+            provider = self._get_ai_provider(provider_name, config)
+
+            if not provider:
+                return {'success': False, 'error': 'AI provider not available'}
+
+            # Use AI to make the changes
+            log_func(f"🤖 Using AI to modify {file_path}...")
+            log_func(f"🔍 Analyzing changes needed...")
+            updated_content = provider.make_change(
+                file_content=current_content,
+                old_text=current_content[:200] + "...",  # Context
+                new_text=changes,  # What to change
+                file_path=str(full_path),
+                custom_instructions=changes
+            )
+
+            if updated_content and updated_content != current_content:
+                log_func(f"💾 Writing changes to {file_path}...")
+                # Write updated content
+                with open(full_path, 'w', encoding='utf-8') as f:
+                    f.write(updated_content)
+
+                log_func(f"✅ Successfully modified {file_path}")
+                return {'success': True, 'file': file_path}
+            else:
+                return {'success': False, 'error': 'AI could not generate changes'}
+
+        except Exception as e:
+            return {'success': False, 'error': f'Error modifying file: {str(e)}'}
+
+    def _create_file(self, file_path: str, content: str, local_repo_path: str, log=None) -> Dict[str, Any]:
+        """Create a new file"""
+
+        log_func = log if log else self.logger.log
+        full_path = Path(local_repo_path) / file_path
+
+        if full_path.exists():
+            return {'success': False, 'error': f'File already exists: {file_path}'}
+
+        try:
+            log_func(f"📄 Creating new file: {file_path}")
+            # Create parent directories if needed
+            full_path.parent.mkdir(parents=True, exist_ok=True)
+
+            # Create file with content
+            with open(full_path, 'w', encoding='utf-8') as f:
+                f.write(content or f"# TODO: Implement {file_path}\n")
+
+            log_func(f"✅ Created {file_path}")
+            return {'success': True, 'file': file_path}
+
+        except Exception as e:
+            return {'success': False, 'error': f'Error creating file: {str(e)}'}
+
+    def _delete_file(self, file_path: str, local_repo_path: str, log=None) -> Dict[str, Any]:
+        """Delete a file"""
+
+        log_func = log if log else self.logger.log
+        full_path = Path(local_repo_path) / file_path
+
+        if not full_path.exists():
+            return {'success': False, 'error': f'File not found: {file_path}'}
+
+        try:
+            log_func(f"🗑️  Deleting file: {file_path}")
+            full_path.unlink()
+            log_func(f"✅ Deleted {file_path}")
+            return {'success': True, 'file': file_path}
+
+        except Exception as e:
+            return {'success': False, 'error': f'Error deleting file: {str(e)}'}

src/app_components/ai_manager.py

@@ -742,6 +742,12 @@ Current file content:
 class ChatGPTProvider(AIProvider):
     """ChatGPT/GPT-4 provider using OpenAI API"""
 
+    def __init__(self, api_key: str, logger: Logger):
+        """Initialize ChatGPT provider with OpenAI client"""
+        super().__init__(api_key, logger)
+        import openai
+        self.client = openai.OpenAI(api_key=api_key)
+
     def make_change(self, file_content: str, old_text: str, new_text: str, file_path: str, custom_instructions: str = None) -> Optional[str]:
         """Make smart, targeted changes based on reference text and suggestions
 
@@ -774,8 +780,8 @@ class ChatGPTProvider(AIProvider):
         """Generate updated document content using ChatGPT"""
 
         try:
-            import openai
-            client = openai.OpenAI(api_key=self.api_key)
+            # Use the client initialized in __init__
+            client = self.client
 
             # Build custom instructions text
             if custom_instructions and custom_instructions.strip():
@@ -3358,16 +3364,19 @@ class AIManager:
 
             # Anthropic/Claude
             elif provider_name in ['claude', 'anthropic']:
+                # Try both CLAUDE_API_KEY and ANTHROPIC_API_KEY for compatibility
+                api_key = config.get('CLAUDE_API_KEY', '')
+                if not api_key:
                     api_key = config.get('ANTHROPIC_API_KEY', '')
                 if not api_key:
-                    return "Error: Anthropic API key not configured"
+                    return "Error: Claude API key not configured (tried both CLAUDE_API_KEY and ANTHROPIC_API_KEY)"
 
                 try:
                     import anthropic
                     client = anthropic.Anthropic(api_key=api_key)
 
                     response = client.messages.create(
-                        model=config.get('ANTHROPIC_MODEL', 'claude-3-5-sonnet-20241022'),
+                        model=config.get('ANTHROPIC_MODEL', 'claude-sonnet-4-5'),
                         max_tokens=2000,
                         messages=[
                             {"role": "user", "content": prompt}

src/app_components/main_gui.py

@@ -104,6 +104,15 @@ class MainGUI:
         # Initialize logger
         self.logger = None  # Will be set after UI is created
 
+        # AI Action Plan state
+        self.current_action_plan = None
+        self.plan_display_ref = ft.Ref[ft.Column]()
+        self.plan_progress_ref = ft.Ref[ft.ProgressBar]()
+        self.plan_status_ref = ft.Ref[ft.Text]()
+        self.execute_plan_button_ref = ft.Ref[ft.ElevatedButton]()
+        self.generate_plan_button_ref = ft.Ref[ft.ElevatedButton]()
+        self.ai_instructions_ref = ft.Ref[ft.TextField]()
+
         # Register settings change listener for live updates
         self.config_manager.register_listener(self._on_settings_changed)
 
@@ -457,6 +466,11 @@ class MainGUI:
                     icon=ft.icons.DIFFERENCE,
                     content=self._create_diff_tab()
                 ),
+                ft.Tab(
+                    text="AI Action Plan",
+                    icon=ft.icons.AUTO_AWESOME,
+                    content=self._create_ai_plan_tab()
+                ),
             ],
             expand=True,
         )
@@ -1538,6 +1552,10 @@ Description:
         # Store the active item
         self.active_workflow_item = item
 
+        # Enable the Generate Plan button when an item is selected
+        if self.generate_plan_button_ref.current:
+            self.generate_plan_button_ref.current.disabled = False
+
         # Determine display labels
         repo_label = "Target" if item.repo_source == "target" else "Fork"
         repo_color = ft.colors.BLUE if item.repo_source == "target" else ft.colors.PURPLE
@@ -2683,6 +2701,409 @@ Description:
             self.diff_text_ref.current.value = diff_content
             self.page.update()
 
+    def _create_ai_plan_tab(self) -> ft.Container:
+        """Create the AI Action Plan tab"""
+
+        # Plan display area (initially empty)
+        plan_display = ft.Column(
+            ref=self.plan_display_ref,
+            controls=[
+                ft.Container(
+                    content=ft.Column([
+                        ft.Icon(ft.icons.AUTO_AWESOME, size=64, color=ft.colors.BLUE_400),
+                        ft.Text(
+                            "AI Action Plan",
+                            size=24,
+                            weight=ft.FontWeight.BOLD,
+                            color=ft.colors.BLUE_400,
+                        ),
+                        ft.Text(
+                            "Select a PR or Issue and generate an AI action plan",
+                            size=14,
+                            color=ft.colors.GREY_600,
+                            text_align=ft.TextAlign.CENTER,
+                        ),
+                    ], horizontal_alignment=ft.CrossAxisAlignment.CENTER, spacing=10),
+                    alignment=ft.alignment.center,
+                    expand=True,
+                )
+            ],
+            spacing=10,
+            scroll=ft.ScrollMode.AUTO,
+            expand=True,
+        )
+
+        # AI Instructions input
+        ai_instructions = ft.TextField(
+            ref=self.ai_instructions_ref,
+            label="Additional Instructions (Optional)",
+            hint_text="Add any specific requirements or context for the AI...",
+            multiline=True,
+            min_lines=2,
+            max_lines=4,
+        )
+
+        # Buttons row
+        buttons_row = ft.Row(
+            [
+                ft.ElevatedButton(
+                    ref=self.generate_plan_button_ref,
+                    text="Generate Plan",
+                    icon=ft.icons.PSYCHOLOGY,
+                    on_click=self._on_generate_plan_click,
+                    disabled=True,  # Enable when item is selected
+                ),
+                ft.ElevatedButton(
+                    ref=self.execute_plan_button_ref,
+                    text="Execute Plan",
+                    icon=ft.icons.PLAY_ARROW,
+                    on_click=self._on_execute_plan_click,
+                    disabled=True,  # Enable when plan is generated
+                    style=ft.ButtonStyle(
+                        color=ft.colors.WHITE,
+                        bgcolor=ft.colors.GREEN_700,
+                    ),
+                ),
+            ],
+            spacing=10,
+        )
+
+        # Progress bar (indeterminate/animated when visible)
+        progress_bar = ft.ProgressBar(
+            ref=self.plan_progress_ref,
+            visible=False,
+            color=ft.colors.BLUE_400,
+            bgcolor=ft.colors.BLUE_GREY_800,
+        )
+
+        # Status text
+        status_text = ft.Text(
+            ref=self.plan_status_ref,
+            value="",
+            color=ft.colors.BLUE_400,
+            size=12,
+        )
+
+        # Main layout
+        return ft.Container(
+            content=ft.Column(
+                [
+                    ai_instructions,
+                    buttons_row,
+                    progress_bar,
+                    status_text,
+                    ft.Divider(),
+                    plan_display,
+                ],
+                spacing=10,
+                expand=True,
+            ),
+            padding=20,
+            expand=True,
+        )
+
+    def _on_generate_plan_click(self, e):
+        """Generate AI action plan for current item"""
+        if not self.active_workflow_item:
+            self._show_snackbar("Please select a PR or Issue first", error=True)
+            return
+
+        # Run in thread to avoid blocking UI
+        import threading
+        thread = threading.Thread(target=self._generate_plan_async)
+        thread.start()
+
+    def _generate_plan_async(self):
+        """Generate plan asynchronously"""
+        try:
+            # Update UI - disable both buttons during generation
+            if self.generate_plan_button_ref.current:
+                self.generate_plan_button_ref.current.disabled = True
+            if self.execute_plan_button_ref.current:
+                self.execute_plan_button_ref.current.disabled = True
+            if self.plan_status_ref.current:
+                self.plan_status_ref.current.value = "🤖 Generating action plan..."
+            if self.plan_progress_ref.current:
+                self.plan_progress_ref.current.visible = True
+            self.page.update()
+
+            # Create action planner
+            from .ai_action_planner import AIActionPlanner
+            planner = AIActionPlanner(self.ai_manager, self.logger, self.config_manager)
+
+            # Get custom instructions
+            custom_instructions = ""
+            if self.ai_instructions_ref.current:
+                custom_instructions = self.ai_instructions_ref.current.value or ""
+
+            # Generate plan
+            plan = planner.generate_plan(self.active_workflow_item, custom_instructions)
+
+            if plan:
+                self.current_action_plan = plan
+                self._display_action_plan(plan)
+
+                if self.plan_status_ref.current:
+                    self.plan_status_ref.current.value = f"✅ Plan generated with {len(plan.steps)} steps"
+                if self.execute_plan_button_ref.current:
+                    self.execute_plan_button_ref.current.disabled = False
+            else:
+                if self.plan_status_ref.current:
+                    self.plan_status_ref.current.value = "❌ Failed to generate plan"
+
+        except Exception as ex:
+            import traceback
+            self.logger.log(f"❌ Error generating plan: {str(ex)}")
+            self.logger.log(f"❌ Traceback: {traceback.format_exc()}")
+            if self.plan_status_ref.current:
+                self.plan_status_ref.current.value = f"❌ Error: {str(ex)}"
+
+        finally:
+            # Re-enable Generate Plan button
+            if self.generate_plan_button_ref.current:
+                self.generate_plan_button_ref.current.disabled = False
+            # Only enable Execute Plan if we have a valid plan
+            if self.execute_plan_button_ref.current and self.current_action_plan:
+                self.execute_plan_button_ref.current.disabled = False
+            if self.plan_progress_ref.current:
+                self.plan_progress_ref.current.visible = False
+            self.page.update()
+
+    def _display_action_plan(self, plan):
+        """Display the generated action plan"""
+        if not self.plan_display_ref.current:
+            return
+
+        steps_widgets = []
+
+        # Plan header
+        steps_widgets.append(
+            ft.Container(
+                content=ft.Column([
+                    ft.Text(plan.title, size=18, weight=ft.FontWeight.BOLD),
+                    ft.Text(
+                        f"Context: {plan.context.get('item_type', 'unknown').upper()} #{plan.context.get('item_number', '?')}",
+                        size=12,
+                        color=ft.colors.GREY_600
+                    ),
+                ]),
+                padding=10,
+                bgcolor=ft.colors.BLUE_GREY_900,
+                border_radius=5,
+            )
+        )
+
+        # Individual steps
+        for step in plan.steps:
+            step_num = step['step_number']
+            description = step['description']
+            file_path = step.get('file_path', 'N/A')
+            action_type = step.get('action_type', 'unknown')
+            status = step.get('status', 'pending')
+
+            # Status icon and color based on step status
+            if status == 'completed':
+                icon = ft.icons.CHECK_CIRCLE
+                icon_color = ft.colors.GREEN
+                bg_color = None  # No special background
+            elif status == 'failed':
+                icon = ft.icons.ERROR
+                icon_color = ft.colors.RED
+                bg_color = None
+            elif status == 'in_progress':
+                icon = ft.icons.TIMELAPSE  # Animated-looking icon
+                icon_color = ft.colors.BLUE_400
+                bg_color = ft.colors.BLUE_GREY_800  # Highlight in-progress steps
+            else:  # pending
+                icon = ft.icons.RADIO_BUTTON_UNCHECKED
+                icon_color = ft.colors.GREY
+                bg_color = None
+
+            step_widget = ft.Container(
+                content=ft.Row([
+                    ft.Icon(icon, color=icon_color, size=20),
+                    ft.Column([
+                        ft.Text(f"Step {step_num}: {description}", weight=ft.FontWeight.BOLD, size=14),
+                        ft.Text(f"Action: {action_type}", size=11, color=ft.colors.BLUE_400),
+                        ft.Text(f"File: {file_path}", size=11, color=ft.colors.GREY_600) if file_path != 'N/A' else ft.Container(),
+                    ], spacing=2, expand=True),
+                ], spacing=10),
+                padding=10,
+                border=ft.border.all(1, ft.colors.GREY_700),
+                border_radius=5,
+                bgcolor=bg_color,  # Highlight in-progress steps
+            )
+            steps_widgets.append(step_widget)
+
+        # Update display
+        self.plan_display_ref.current.controls = steps_widgets
+        self.page.update()
+
+    def _on_execute_plan_click(self, e):
+        """Execute the current action plan"""
+        if not self.current_action_plan:
+            self._show_snackbar("No plan to execute", error=True)
+            return
+
+        # Run in thread
+        import threading
+        thread = threading.Thread(target=self._execute_plan_async)
+        thread.start()
+
+    def _execute_plan_async(self):
+        """Execute plan asynchronously"""
+        try:
+            self.logger.log("🔧 Starting _execute_plan_async...")
+
+            # Update UI - disable both buttons during execution
+            if self.execute_plan_button_ref.current:
+                self.execute_plan_button_ref.current.disabled = True
+            if self.generate_plan_button_ref.current:
+                self.generate_plan_button_ref.current.disabled = True
+            if self.plan_status_ref.current:
+                self.plan_status_ref.current.value = "▶️ Executing plan..."
+            if self.plan_progress_ref.current:
+                self.plan_progress_ref.current.visible = True
+            self.page.update()
+
+            # Get local repo path
+            config = self.config_manager.get_config()
+            local_repo_path = config.get('LOCAL_REPO_PATH', '')
+            self.logger.log(f"🔧 Local repo path: {local_repo_path}")
+
+            if not local_repo_path:
+                self.logger.log("❌ LOCAL_REPO_PATH is not set")
+                self._show_snackbar("Please set LOCAL_REPO_PATH in settings", error=True)
+                if self.plan_status_ref.current:
+                    self.plan_status_ref.current.value = "❌ LOCAL_REPO_PATH not set in settings"
+                    self.page.update()
+                return
+
+            # Create action planner
+            from .ai_action_planner import AIActionPlanner
+            planner = AIActionPlanner(self.ai_manager, self.logger, self.config_manager)
+            self.logger.log(f"🔧 Created AIActionPlanner, about to execute plan with {len(self.current_action_plan.steps)} steps")
+
+            # Execute with progress callback
+            def progress_callback(current, total, message):
+                if self.plan_status_ref.current:
+                    self.plan_status_ref.current.value = f"▶️ {message} ({current}/{total})"
+                # Update the plan display in real-time to show progress
+                if self.current_action_plan:
+                    self._display_action_plan(self.current_action_plan)
+                self.page.update()
+
+            # Execute with logging callback for thought process
+            def log_callback(message):
+                # Log to the processing log
+                self.logger.log(message)
+                # Also show in status if it's important
+                if any(keyword in message for keyword in ["🤖", "✅", "❌", "📝", "🔍"]):
+                    if self.plan_status_ref.current:
+                        self.plan_status_ref.current.value = message
+                    self.page.update()
+
+            result = planner.execute_plan(
+                self.current_action_plan,
+                local_repo_path,
+                progress_callback,
+                log_callback
+            )
+
+            self.logger.log(f"🔧 execute_plan returned: {result}")
+
+            # Update display
+            self._display_action_plan(self.current_action_plan)
+
+            # Show completion dialog
+            if result['success']:
+                self._show_completion_dialog(result)
+            else:
+                if self.plan_status_ref.current:
+                    self.plan_status_ref.current.value = f"⚠️ Plan completed with errors: {result['failed']}/{result['total']} steps failed"
+
+        except Exception as ex:
+            import traceback
+            self.logger.log(f"❌ Error executing plan: {str(ex)}")
+            self.logger.log(f"❌ Traceback: {traceback.format_exc()}")
+            if self.plan_status_ref.current:
+                self.plan_status_ref.current.value = f"❌ Error: {str(ex)}"
+
+        finally:
+            # Re-enable both buttons after execution
+            if self.execute_plan_button_ref.current:
+                self.execute_plan_button_ref.current.disabled = False
+            if self.generate_plan_button_ref.current:
+                self.generate_plan_button_ref.current.disabled = False
+            if self.plan_progress_ref.current:
+                self.plan_progress_ref.current.visible = False
+            self.page.update()
+
+    def _show_completion_dialog(self, result):
+        """Show dialog after plan execution with options to push/create PR"""
+        context = self.current_action_plan.context
+        item_type = context.get('item_type', 'unknown')
+        item_number = context.get('item_number', '?')
+
+        def close_dialog(e):
+            dialog.open = False
+            self.page.update()
+
+        def push_changes(e):
+            dialog.open = False
+            self.page.update()
+            self._push_changes_async()
+
+        def create_pr(e):
+            dialog.open = False
+            self.page.update()
+            self._create_pr_from_plan_async()
+
+        dialog = ft.AlertDialog(
+            title=ft.Text("✅ Plan Execution Complete!"),
+            content=ft.Column([
+                ft.Text(f"Successfully completed {result['completed']}/{result['total']} steps"),
+                ft.Divider(),
+                ft.Text("What would you like to do next?", weight=ft.FontWeight.BOLD),
+                ft.Text(f"• Push changes to the {item_type} branch", size=12),
+                ft.Text(f"• Create/Update PR for these changes", size=12),
+                ft.Text(f"• Review changes manually", size=12),
+            ], tight=True, spacing=10),
+            actions=[
+                ft.TextButton("Review Later", on_click=close_dialog),
+                ft.ElevatedButton(
+                    "Push Changes",
+                    icon=ft.icons.CLOUD_UPLOAD,
+                    on_click=push_changes,
+                ),
+                ft.ElevatedButton(
+                    "Create/Update PR",
+                    icon=ft.icons.MERGE_TYPE,
+                    on_click=create_pr,
+                    style=ft.ButtonStyle(bgcolor=ft.colors.GREEN_700),
+                ),
+            ],
+            actions_alignment=ft.MainAxisAlignment.END,
+        )
+
+        self.page.dialog = dialog
+        dialog.open = True
+        self.page.update()
+
+    def _push_changes_async(self):
+        """Push changes to git repository"""
+        # This will be implemented to push changes
+        self.logger.log("🚀 Pushing changes to repository...")
+        # TODO: Implement git push logic
+        self._show_snackbar("Push functionality coming soon!")
+
+    def _create_pr_from_plan_async(self):
+        """Create or update PR from executed plan"""
+        # This will be implemented to create/update PR
+        self.logger.log("📝 Creating/updating PR...")
+        # TODO: Implement PR creation logic
+        self._show_snackbar("PR creation functionality coming soon!")
+
 
 class Logger:
     """Logger class for Flet"""

src/app_components/settings_dialog.py

@@ -771,6 +771,7 @@ class SettingsDialog:
             if not path_str:
                 path_str = str(Path.home() / "Downloads" / "github_repos")
 
+            print(f"🔍 Scanning for repos in: {path_str}")
             base_path = Path(path_str)
 
             if not base_path.exists():
@@ -783,26 +784,70 @@ class SettingsDialog:
             # Scan for git repositories
             repos = []
             try:
-                for owner_dir in base_path.iterdir():
-                    if not owner_dir.is_dir():
+                # First, check if repos are directly in the base path (flat structure)
+                for item in base_path.iterdir():
+                    if not item.is_dir():
                         continue
 
-                    for repo_dir in owner_dir.iterdir():
+                    git_dir = item / ".git"
+                    if git_dir.exists():
+                        # This is a git repo directly in the base path
+                        # Try to get the remote origin to get owner/repo format
+                        try:
+                            import subprocess
+                            result = subprocess.run(
+                                ['git', 'config', '--get', 'remote.origin.url'],
+                                cwd=str(item),
+                                capture_output=True,
+                                text=True,
+                                timeout=5
+                            )
+
+                            if result.returncode == 0:
+                                url = result.stdout.strip()
+                                # Parse GitHub URL to get owner/repo
+                                if 'github.com' in url:
+                                    # Handle both HTTPS and SSH URLs
+                                    if url.startswith('https://'):
+                                        # https://github.com/owner/repo.git
+                                        parts = url.replace('https://github.com/', '').replace('.git', '').split('/')
+                                        if len(parts) >= 2:
+                                            repo_name = f"{parts[0]}/{parts[1]}"
+                                            repos.append(repo_name)
+                                            continue
+                                    elif url.startswith('git@'):
+                                        # git@github.com:owner/repo.git
+                                        parts = url.replace('git@github.com:', '').replace('.git', '').split('/')
+                                        if len(parts) >= 2:
+                                            repo_name = f"{parts[0]}/{parts[1]}"
+                                            repos.append(repo_name)
+                                            continue
+                        except:
+                            pass
+
+                        # Fallback: use directory name
+                        repos.append(f"local/{item.name}")
+                    else:
+                        # Check if this is an owner directory with repos inside (nested structure)
+                        for repo_dir in item.iterdir():
                             if not repo_dir.is_dir():
                                 continue
 
                             git_dir = repo_dir / ".git"
                             if git_dir.exists():
-                            repo_name = f"{owner_dir.name}/{repo_dir.name}"
+                                repo_name = f"{item.name}/{repo_dir.name}"
                                 repos.append(repo_name)
 
             except Exception as e:
                 print(f"Error scanning repos: {e}")
+                import traceback
+                traceback.print_exc()
 
             # Update dropdown
             if self.detected_repos_dropdown_ref.current:
                 if repos:
                     repos.sort()
+                    print(f"✅ Found {len(repos)} repo(s): {', '.join(repos)}")
                     self.detected_repos_dropdown_ref.current.options = [
                         ft.dropdown.Option(repo) for repo in repos
                     ]
@@ -811,6 +856,7 @@ class SettingsDialog:
                     else:
                         self.detected_repos_dropdown_ref.current.value = f'{len(repos)} repo(s) found - select one'
                 else:
+                    print(f"❌ No git repositories found in {path_str}")
                     self.detected_repos_dropdown_ref.current.value = 'No git repositories found'
                     self.detected_repos_dropdown_ref.current.options = []
 

src/main.py

@@ -3,6 +3,10 @@ GitHub Pulse
 Main application entry point
 
 This application provides GitHub automation workflows with AI assistance.
+
+Note: You may see a Flutter engine warning when closing the app:
+  "embedder.cc (2519): 'FlutterEngineRemoveView' returned 'kInvalidArguments'"
+This is a harmless known issue with Flet/Flutter and can be safely ignored.
 """
 
 import sys