Agent Conversation Flow Designer

You are an expert in designing conversational AI agent architectures, specialising in multi-turn dialogue systems and context management.

Design a conversation flow for this agent specification:

## Agent Requirements
[paste agent specification/requirements here]

## Design Framework

### 1. Conversation Mapping
- Identify all conversation entry points
- Map user intent categories
- Define conversation goals and outcomes
- Chart decision trees and branching logic

### 2. State Management Design
- Design conversation state schema
- Define state transition triggers
- Plan context preservation strategy
- Design session management approach

### 3. Turn-Taking Architecture
- Model user input patterns
- Design agent response strategies
- Plan interruption handling
- Define conversation repair mechanisms

### 4. Memory and Context
- Design short-term memory structure
- Plan long-term context storage
- Define context retrieval mechanisms
- Design context summarisation strategy

### 5. Error and Edge Cases
- Plan misunderstanding recovery
- Design clarification request patterns
- Handle out-of-scope queries
- Plan conversation reset mechanisms

## Output Format

**CONVERSATION ARCHITECTURE**
```
Entry Points: [list all ways users can start]
Primary Flows: [main conversation paths]
Fallback Flows: [error recovery paths]
Exit Points: [conversation termination scenarios]
```

**STATE SCHEMA**
```json
{
  "conversation_state": {},
  "user_context": {},
  "agent_memory": {},
  "session_data": {}
}
```

**FLOW DIAGRAMS**
- Visual representation of conversation flows
- State transition mappings
- Decision point documentation
- Context handoff points

**IMPLEMENTATION GUIDE**
- Technology stack recommendations
- Integration patterns with LLMs
- Testing strategy for conversation flows
- Monitoring and analytics setup

**EDGE CASE HANDLING**
- Common failure modes and solutions
- Context loss recovery strategies
- Multi-language conversation patterns
- Concurrent conversation management

Focus on creating robust, production-ready conversation architectures that maintain coherent context across extended interactions while handling real-world conversation complexities.