Memory API

Persistent storage, RAG, and knowledge management

This document covers AgenticGoKit's Memory API, which enables agents to store and retrieve information persistently. The Memory system is essential for building agents with long-term memory, knowledge bases, and RAG (Retrieval-Augmented Generation) capabilities.

📋 Core Concepts

Memory Interface

The core interface for agent memory systems:

type Memory interface {
    // Personal memory operations
    Store(ctx context.Context, content string, tags ...string) error
    Query(ctx context.Context, query string, limit ...int) ([]core.Result, error)

    // Key-value storage
    Remember(ctx context.Context, key string, value any) error
    Recall(ctx context.Context, key string) (any, error)

    // Chat history
    AddMessage(ctx context.Context, role, content string) error
    GetHistory(ctx context.Context, limit ...int) ([]core.Message, error)

    // Knowledge base (RAG)
    IngestDocument(ctx context.Context, doc core.Document) error
    IngestDocuments(ctx context.Context, docs []core.Document) error
    SearchKnowledge(ctx context.Context, query string, options ...core.SearchOption) ([]core.KnowledgeResult, error)
    SearchAll(ctx context.Context, query string, options ...core.SearchOption) (*core.HybridResult, error)
    BuildContext(ctx context.Context, query string, options ...core.ContextOption) (*core.RAGContext, error)

    // Session management
    NewSession() string
    SetSession(ctx context.Context, sessionID string) context.Context
    ClearSession(ctx context.Context) error
    Close() error
}

🚀 Basic Usage

Creating Memory System

// Create memory configuration
config := agentflow.AgentMemoryConfig{
    Provider:   "pgvector",
    Connection: "postgres://user:password@localhost:5432/agentflow",
    Dimensions: 1536,
    Embedding: agentflow.EmbeddingConfig{
        Provider: "openai",
        APIKey:   "your-api-key",
        Model:    "text-embedding-3-small",
    },
}

// Initialize memory
memory, err := agentflow.NewMemory(config)
if err != nil {
    log.Fatal(err)
}
defer memory.Close()

For complete documentation including RAG operations, document ingestion, and all memory providers, see the Agent API reference.