17-Follower-Memory-Summary-Prompt

Follower Memory Summary Prompt

Overview

Purpose: Summarize shared scenes and experiences for NPC followers to maintain memory coherence Source File: summaries.py Model Usage: Context summarization for follower memory system Implementation: ChatBotRPG follower tracking system

Context

This prompt is used to generate memory summaries for NPC followers when they accompany the player across multiple scenes. It ensures followers retain coherent memories of shared experiences while managing context window limitations.

Prompt Structure

Core Objective

Summarize conversation history and shared scenes from the follower’s perspective, maintaining:

• Memory coherence across scene transitions
• Follower-specific observations and reactions
• Key events and interactions witnessed
• Emotional state and relationship development

Expected Inputs

• Previous scene transcripts
• Follower character sheet and personality
• Shared events and interactions
• Current relationship status with player

Expected Outputs

• Concise summary of shared experiences (follower POV)
• Key emotional moments
• Notable decisions or turning points
• Context for future interactions

Usage Pattern

# From summaries.py - Follower memory summarization
def summarize_follower_memory(follower, scene_history):
    """
    Generate summary of shared experiences for follower NPC.
 
    Args:
        follower: NPC follower character data
        scene_history: List of previous scenes with this follower
 
    Returns:
        str: Summarized memory from follower perspective
    """
    prompt = construct_follower_summary_prompt(
        follower_data=follower,
        scenes=scene_history
    )
 
    summary = llm_inference(prompt)
 
    return summary

Related Prompts

• 16-Context-Summarization-Prompt - General conversation history summarization
• 01-Character-Generation-Prompt - Character personality definition
• 03-Scene-Setting-Prompt - Scene context management

Implementation Notes

Memory Management

• Summarization triggered on scene transitions
• Maintains follower continuity across locations
• Reduces token usage for long-term followers
• Preserves relationship development arc

Technical Details

• Context Window: Manages follower memory within limits
• Trigger Condition: Scene change with active follower
• Storage: Persisted in follower character state
• Update Frequency: Per scene transition or every N turns

Related Files

• character-schema - Character data structure
• architecture - System architecture overview
• state-management - State persistence patterns

---

Extraction Status

This prompt was identified in the ChatBotRPG codebase during prompt-forensics-agent analysis. Full source code details available in summaries.py.

Usage Context

Part of the follower/companion system that allows NPCs to travel with the player across multiple scenes while maintaining coherent memory and personality.