Agent MCP

# Agent-MCP/agent_mcp/utils/prompt_templates.py from typing import Dict, Optional, Any from ..core.config import logger # Predefined prompt templates for different agent types PROMPT_TEMPLATES = { "worker_with_rag": """This is your agent token: {agent_token} Ask the project RAG agent at least 5-7 questions to understand what you need to do. I want you to critically think when asking a question, then criticize yourself before asking that question. How you criticize yourself is by proposing an idea, criticizing it, and based on that criticism you pull through with that idea. It's better to add too much context versus too little. Add all these context entries to the agent mcp. ACT AUTO --worker --memory""", "basic_worker": """You are {agent_id} worker agent. Your Agent Token: {agent_token} Query the project knowledge graph to understand: 1. Overall system architecture 2. Your specific responsibilities 3. Integration points with other components 4. Coding standards and patterns to follow 5. Current implementation status Begin implementation following the established patterns. AUTO --worker --memory""", "frontend_worker": """You are {agent_id} frontend worker agent. Your Agent Token: {agent_token} Query the project knowledge graph to understand: 1. UI/UX requirements and design system 2. Frontend architecture and component structure 3. State management patterns 4. Integration with backend APIs 5. Testing and validation requirements Focus on component-based development with visual validation. AUTO --worker --playwright""", "admin_agent": """You are the admin agent. Admin Token: {admin_token} Your role is to: - Coordinate all development work - Create and manage worker agents - Maintain project context - Assign tasks based on agent specializations Query the project RAG for current status and begin coordination.""", "custom": "{custom_prompt}", "testing_agent": """You are {agent_id} - a CRITICAL TESTING AGENT. Your Agent Token: {agent_token} TASK JUST COMPLETED BY: {completed_by_agent} - Task ID: {completed_task_id} - Title: {completed_task_title} - Description: {completed_task_description} YOUR MISSION: HEAVY CRITICISM & VALIDATION 1. Query project RAG to understand what was implemented 2. Test REAL functionality - NO MOCK DATA ALLOWED 3. Check for actual working implementation 4. Look for edge cases, errors, incomplete features 5. Validate integration points work correctly TESTING APPROACH: - For frontend: Use Playwright to actually interact with components - For backend: Test real API endpoints with real data - For database: Verify actual data operations - Check error handling, edge cases, security RESPONSE PROTOCOL: - If ALL tests pass: Send message "wait, continue" to {completed_by_agent} - If ANY test fails: 1. Update task {completed_task_id} status back to "pending" using update_task_status 2. Add detailed notes about what failed and what needs to be fixed 3. Send message to {completed_by_agent} with specific failure details and next steps CRITICAL: Always update task status to "pending" if tests fail - this ensures the original agent can continue working on fixes. BE RUTHLESSLY CRITICAL. Better to catch issues now than deploy broken code. AUTO --worker --memory""", } def get_prompt_template(template_name: str) -> Optional[str]: """Get a prompt template by name.""" return PROMPT_TEMPLATES.get(template_name) def format_prompt(template_name: str, **kwargs) -> Optional[str]: """ Format a prompt template with the provided variables. Args: template_name: Name of the template to use **kwargs: Variables to substitute in the template Returns: Formatted prompt string or None if template not found """ template = get_prompt_template(template_name) if not template: logger.error(f"Prompt template '{template_name}' not found") return None try: return template.format(**kwargs) except KeyError as e: logger.error(f"Missing required variable {e} for template '{template_name}'") return None except Exception as e: logger.error(f"Error formatting template '{template_name}': {e}") return None def create_custom_prompt(prompt_text: str, **kwargs) -> str: """ Create a custom prompt with variable substitution. Args: prompt_text: The custom prompt text **kwargs: Variables to substitute Returns: Formatted custom prompt """ try: return prompt_text.format(**kwargs) except Exception as e: logger.warning(f"Error formatting custom prompt: {e}, returning as-is") return prompt_text def get_available_templates() -> Dict[str, str]: """Get all available prompt templates with descriptions.""" descriptions = { "worker_with_rag": "Worker agent with RAG querying and critical thinking instructions", "basic_worker": "Standard worker agent with basic project querying", "frontend_worker": "Frontend-focused worker with UI/UX emphasis", "admin_agent": "Admin agent for coordination and management", "custom": "Custom prompt template (requires custom_prompt parameter)", "testing_agent": "Critical testing agent that validates completed work with heavy criticism", } return descriptions def validate_template_variables(template_name: str, variables: Dict[str, Any]) -> bool: """ Validate that all required variables are provided for a template. Args: template_name: Name of the template to validate variables: Dictionary of variables to check Returns: True if all required variables are present, False otherwise """ template = get_prompt_template(template_name) if not template: return False # Extract variable names from template import re required_vars = set(re.findall(r"\{(\w+)\}", template)) provided_vars = set(variables.keys()) missing_vars = required_vars - provided_vars if missing_vars: logger.error( f"Missing required variables for template '{template_name}': {missing_vars}" ) return False return True def build_agent_prompt( agent_id: str, agent_token: str, admin_token: str, template_name: str = "basic_worker", custom_prompt: str = None, **extra_vars, ) -> Optional[str]: """ Build a complete agent prompt with all necessary information. Args: agent_id: The agent's ID agent_token: The agent's token admin_token: The admin token template_name: Name of the template to use custom_prompt: Custom prompt text (for custom template) **extra_vars: Additional variables for template substitution Returns: Complete formatted prompt or None if error """ # Prepare base variables variables = { "agent_id": agent_id, "agent_token": agent_token, "admin_token": admin_token, **extra_vars, } # Handle custom template if template_name == "custom" and custom_prompt: variables["custom_prompt"] = custom_prompt # Validate and format if not validate_template_variables(template_name, variables): return None return format_prompt(template_name, **variables)