blightbow/evennia_ai
2
0
Fork 0
Code Issues 9 Pull requests Projects Releases Packages Wiki Activity Actions
Page: Home
Pages
Architecture Commands and API Architecture Context System Architecture Core Engine Architecture Event Sourcing Architecture Generative Reflection Architecture Helpers Architecture Journal System Architecture LLM Interaction Architecture LLM Providers Architecture Logging Architecture Memory and Sleep Architecture Overview Architecture Persona Protection Architecture Prompt System Architecture RAG Implementation Architecture Resilience System Architecture Safety System Architecture Self Management Architecture Sub Agent Delegation Architecture Task Assessment Architecture Token Management Architecture Tool System Configuration Reference Context and Memory Flow Analysis Data Flow 01 Context Compaction Data Flow 02 ReAct Loop Data Flow 03 Memory Consolidation Data Flow 04 Message Classification Data Flow 05 Entity Profile System Data Flow 06 Tool Execution Data Flow 07 Sleep Mode Transitions Data Flow 08 LLM Provider Interaction Data Flow 09 Self Management Operations Home LLM Decision Patterns Research Foundations User Guide 00 Index User Guide 01 Getting Started User Guide 02 Configuration and Customization User Guide 03 Advanced Capabilities User Guide 04 Troubleshooting
9 Home
Blightbow edited this page 2025-12-09 04:12:38 -05:00
Table of Contents

AI Assistant System Documentation

Welcome to the documentation for the Evennia AI Assistant system.

User Guides

Start here if you're new to the AI Assistant system.

Document Description
User-Guide-00-Index Reading order and navigation by role
User-Guide-01-Getting-Started 5-minute first assistant setup
User-Guide-02-Configuration-and-Customization Intermediate customization guide
User-Guide-03-Advanced-Capabilities Advanced features and extension points
User-Guide-04-Troubleshooting Common problems and solutions

Architecture Documentation

Structural documentation covering system design, layers, and component relationships.

Document Description
Architecture-Overview Five-layer architecture, design principles, component relationships
Architecture-Core-Engine AssistantScript, tick loop, ReAct loop, message classification
Architecture-Context-System 7 context types, component IDs, three-layer execution pattern
Architecture-Memory-and-Sleep Three-tier memory, sleep phases, wake conditions
Architecture-Tool-System Tool categories, registry, validation, token advisory
Architecture-Self-Management Projects, goals, session memory, self-monitoring
Architecture-Journal-System Importance scoring, reflection triggers, journal lifecycle
Architecture-Commands-and-API Command mixins (7), API mixins (11), registry patterns
Architecture-Helpers Cross-layer utilities in helpers.py (2,559 lines)
Architecture-LLM-Providers UnifiedLLMClient, provider abstraction, response models
Architecture-Prompt-System PromptComponent IDs, templates, registry, runtime assembly
Architecture-Task-Assessment Event classification, context signals, LLM/heuristic assessment
Architecture-Sub-Agent-Delegation Personality insulation, delegate discovery, budget management
Architecture-Resilience-System Circuit breaker pattern, tool result caching
Architecture-Event-Sourcing AssistantAggregate, event replay, time-travel debugging
Architecture-Safety-System Emergency stop, consecutive error tracking, recovery
Architecture-Logging Execution log ring buffer, metrics aggregation
Architecture-RAG-Implementation Qdrant client, embedding providers, collection operations
Architecture-LLM-Interaction Context type routing, message building, response parsing
Architecture-Generative-Reflection Three-step reflection process, persona protection
Architecture-Token-Management Token counting, budget analysis, prompt health validation

Data Flow Documentation

Detailed execution sequences showing how data moves through the system.

Document Description
Data-Flow-01-Context-Compaction Sleep/emergency compaction, pre-compaction fact extraction
Data-Flow-02-ReAct-Loop Multi-action execution, termination conditions
Data-Flow-03-Memory-Consolidation Sleep phase memory operations
Data-Flow-04-Message-Classification 8-rule message classification system
Data-Flow-05-Entity-Profile-System O-Mem patterns, Pf→Pa consolidation
Data-Flow-06-Tool-Execution Individual tool execution, caching, validation
Data-Flow-07-Sleep-Mode-Transitions Sleep entry, phase changes, wake conditions
Data-Flow-08-LLM-Provider-Interaction API calls, retries, timeout handling
Data-Flow-09-Self-Management-Operations Project swaps, goal decomposition, session memory

LLM Guidance

Decision patterns and cognitive tool selection for AI assistant NPCs.

Document Description
LLM-Decision-Patterns Memory tier selection, goals vs projects, token pressure responses

Reference Documentation

Document Description
Configuration-Reference All configurable attributes with types and defaults
Research-Foundations Academic papers and design influences

Design Documents

Document Description
Context-and-Memory-Flow-Analysis Original research and design analysis

Quick Links

  • Source code: evennia/contrib/base_systems/ai/
  • Test suite: evennia/contrib/base_systems/ai/tests/
  • Deprecated docs: docs/claude/AI_ARCHITECTURE_V3.md (see wiki instead)

Last updated: 2025-12-09