| File | Purpose |
|---|---|
README.md |
Project overview, installation instructions, and quick start guide |
.env |
Environment variables - API keys, passwords, configuration settings |
.env.example |
Template for environment variables (safe to commit to git) |
docker-compose.yml |
Main deployment file - orchestrates all services and dependencies |
setup.sh |
Automated setup script - creates directories, configures environment |
quick-start.sh |
Fast deployment script - starts services with health checks |
.gitignore |
Git ignore rules (excludes logs, secrets, temp files) |
| File | Purpose |
|---|---|
main.py |
Core memory server - FastAPI app with working/episodic/semantic/procedural memory |
requirements.txt |
Python dependencies (FastAPI, asyncpg, aioredis, chromadb, neo4j, etc.) |
Dockerfile |
Container configuration for the memory server |
config/memory.yaml |
Service-specific memory system configuration |
tests/test_memory_operations.py |
Unit tests for memory storage, retrieval, consolidation |
tests/test_api_endpoints.py |
API endpoint testing (store, query, consolidate) |
| File | Purpose |
|---|---|
main.py |
RAG with memory integration - retrieval-augmented generation using memory context |
requirements.txt |
Dependencies for document processing, embeddings, retrieval |
Dockerfile |
Container for RAG service |
models/embeddings/ |
Stored embedding models for document similarity |
| File | Purpose |
|---|---|
main.py |
Multi-agent coordination - manages agent interactions and shared memory |
requirements.txt |
Dependencies for agent management and coordination |
Dockerfile |
Container for agent orchestrator |
agents/base_agent.py |
Base agent class with memory integration |
agents/memory_agent.py |
Specialized agent that focuses on memory operations |
| File | Purpose |
|---|---|
main.py |
Analytics dashboard - memory usage statistics, performance metrics |
requirements.txt |
Dependencies for data visualization and analytics |
Dockerfile |
Container for analytics service |
static/ |
CSS, JavaScript, images for dashboard UI |
templates/ |
HTML templates for analytics dashboard |
| File | Purpose |
|---|---|
main.py |
System health monitoring - checks all services, sends alerts |
requirements.txt |
Dependencies for health checks and notifications |
Dockerfile |
Container for health monitoring service |
| File | Purpose |
|---|---|
redis-memory.conf |
Redis configuration - memory limits, persistence, security |
chroma-config.yaml |
ChromaDB configuration - vector database settings for semantic memory |
memory-config.yaml |
Main memory system config - consolidation rules, thresholds, decay rates |
neo4j-memory.conf |
Neo4j configuration - graph database settings for procedural memory |
logging.yaml |
Logging configuration - log levels, formats, output destinations |
| File | Purpose |
|---|---|
init-memory.sql |
PostgreSQL initialization - creates extensions, users, permissions |
memory-schema.sql |
Complete database schema - tables, indexes, functions, triggers |
timescale-init.sql |
TimescaleDB setup - time-series database for temporal memory analysis |
seed-data.sql |
Sample data - test agents, memories, relationships for development |
migrations/001_initial_schema.sql |
Database version 1 - initial schema creation |
migrations/002_add_indexes.sql |
Database version 2 - performance optimization indexes |
migrations/003_memory_analytics.sql |
Database version 3 - analytics tables and views |
| File | Purpose |
|---|---|
memory-nginx.conf |
Main Nginx config - API gateway, load balancing, SSL, rate limiting |
ssl/cert.pem |
SSL certificate for HTTPS (production) |
ssl/key.pem |
SSL private key for HTTPS (production) |
logs/ |
Nginx access/error logs directory |
conf.d/upstream.conf |
Upstream server definitions - backend service pools |
conf.d/security.conf |
Security headers and policies - CORS, CSP, etc. |
| File | Purpose |
|---|---|
prometheus-memory.yml |
Metrics collection config - scrape targets, intervals |
rules/memory-alerts.yml |
Alert rules - memory usage, performance thresholds |
rules/performance-rules.yml |
Performance alerts - response times, error rates |
| File | Purpose |
|---|---|
memory-dashboards/memory-overview.json |
Main dashboard - overall system health and memory usage |
memory-dashboards/agent-performance.json |
Agent metrics - individual agent performance and activity |
memory-dashboards/system-health.json |
Infrastructure dashboard - database, Redis, service health |
memory-dashboards/consolidation-metrics.json |
Memory consolidation - consolidation rates, effectiveness |
datasources/prometheus.yaml |
Prometheus data source configuration |
datasources/postgres.yaml |
PostgreSQL data source for memory analytics |
| Directory | Purpose |
|---|---|
knowledge-base/ |
Curated documents for RAG - policies, procedures, knowledge articles |
uploads/ |
User-uploaded documents - dynamic content for processing |
| Directory | Purpose |
|---|---|
sentence-transformers/ |
Pre-trained embedding models - cached transformers |
custom-models/ |
Fine-tuned models - domain-specific embeddings |
| Directory | Purpose |
|---|---|
memory-backups/ |
Memory system backups - periodic exports of all memories |
analytics-reports/ |
Generated reports - memory usage, agent performance |
| File | Purpose |
|---|---|
default-agent.yaml |
Default agent settings - standard memory behavior, thresholds |
specialized-agents/research-agent.yaml |
Research agent config - high memory retention, fact-focused |
specialized-agents/support-agent.yaml |
Support agent config - conversation memory, empathy settings |
specialized-agents/creative-agent.yaml |
Creative agent config - associative memory, inspiration focus |
memory-profiles/conservative-memory.yaml |
Conservative profile - slow forgetting, high importance threshold |
memory-profiles/aggressive-learning.yaml |
Aggressive profile - fast learning, low consolidation threshold |
memory-profiles/balanced-retention.yaml |
Balanced profile - standard memory decay and consolidation |
| File | Purpose |
|---|---|
memory-consolidation.yaml |
Consolidation workflow - when/how to move memories to long-term storage |
agent-coordination.yaml |
Multi-agent workflow - how agents share and coordinate memories |
data-processing.yaml |
Data pipeline - document ingestion, embedding, storage workflow |
| Directory | Purpose |
|---|---|
mcp-server/ |
Memory server logs - API requests, memory operations, errors |
memory-analytics/ |
Analytics service logs - dashboard access, report generation |
health-monitor/ |
Health monitoring logs - service checks, alerts sent |
consolidated/ |
Aggregated logs - combined logs from all services |
| File | Purpose |
|---|---|
backup-memories.sh |
Backup script - exports all memories to files |
restore-memories.sh |
Restore script - imports memories from backup files |
health-check.sh |
Manual health check - tests all services without docker |
performance-test.py |
Performance testing - load testing for memory operations |
data-migration.py |
Data migration - moves data between database versions |
| File | Purpose |
|---|---|
test_memory_operations.py |
Memory function tests - store, retrieve, decay, consolidation |
test_consolidation.py |
Consolidation logic tests - importance-based routing |
test_vector_search.py |
Vector similarity tests - embedding and search accuracy |
| File | Purpose |
|---|---|
test_api_endpoints.py |
API integration tests - full request/response cycles |
test_memory_flow.py |
Memory lifecycle tests - working β long-term storage flow |
test_multi_agent.py |
Multi-agent tests - agent coordination and memory sharing |
| File | Purpose |
|---|---|
load_test.py |
Load testing - concurrent requests, high memory volume |
memory_benchmark.py |
Memory performance - consolidation speed, query response times |
| File | Purpose |
|---|---|
sample_memories.json |
Test memory data - realistic memory examples for testing |
test_agents.yaml |
Test agent configs - agents configured for testing scenarios |
| File | Purpose |
|---|---|
memory-server-api.md |
Memory server API docs - endpoints, parameters, examples |
rag-service-api.md |
RAG service API docs - document processing, retrieval |
orchestrator-api.md |
Orchestrator API docs - agent coordination, memory sharing |
| File | Purpose |
|---|---|
production-setup.md |
Production deployment - SSL, scaling, security hardening |
kubernetes-deployment.md |
Kubernetes guide - K8s manifests, scaling, monitoring |
scaling-guide.md |
Scaling strategies - horizontal/vertical scaling approaches |
| File | Purpose |
|---|---|
memory-system-design.md |
System architecture - memory types, data flow, consolidation |
data-flow-diagrams.md |
Visual diagrams - how data moves through the system |
security-model.md |
Security design - authentication, authorization, data protection |
| File | Purpose |
|---|---|
basic-usage.py |
Getting started code - simple memory operations |
agent-integration.py |
Agent integration - building memory-aware agents |
custom-memory-types.py |
Extensibility - adding new memory types |
| Directory | Purpose |
|---|---|
| SSL certificate storage | Production certificates for HTTPS |
| File | Purpose |
|---|---|
api-keys.env.template |
API key template - secure format for API keys |
database-credentials.env.template |
Database credentials template - secure database access |
| File | Purpose |
|---|---|
access-control.yaml |
Access control rules - who can access what memories |
data-retention.yaml |
Data retention policy - how long to keep different memory types |
| File | Purpose |
|---|---|
memory-inspector.py |
Memory inspection tool - visualize memory contents, relationships |
performance-profiler.py |
Performance analysis - identify bottlenecks, optimize queries |
data-visualizer.py |
Memory visualization - graphs of memory networks, decay patterns |
migration-assistant.py |
Data migration helper - assists with schema changes, data moves |
docker-compose.yml,setup.sh,.envβ Get system running
services/mcp-memory-server/main.pyβ Heart of the memory system
database/memory-schema.sqlβ Complete database structure
config/memory-config.yamlβ Memory behavior settings
monitoring/grafana/memory-dashboards/β Visual system monitoring
agent-config/specialized-agents/β Different agent personalities
This comprehensive file structure provides everything needed for a production-ready, memory-enhanced AI platform with clear organization and purpose! π―