> For the complete documentation index, see [llms.txt](https://ai.ortusbooks.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://ai.ortusbooks.com/main-components/agents/rag.md).

# RAG & Document Loading

Agents can leverage document loaders and vector memory to access knowledge bases and provide grounded, factual responses.

## 🔄 Agent RAG Workflow

{% @mermaid/diagram content="graph TB
Q\[User Query] --> A\[Agent]
A --> R\[Retrieve from Vector Memory]
R --> D\[Relevant Documents]
D --> C\[Inject into Context]
C --> AI\[AI Model]
AI --> RESP\[Grounded Response]

```
style A fill:#BD10E0
style R fill:#4A90E2
style D fill:#50E3C2
style AI fill:#7ED321" %}
```

## Basic RAG Agent

```javascript
// Step 1: Create vector memory
vectorMemory = aiMemory( "chroma", {
    collection       : "product_docs",
    embeddingProvider: "openai",
    embeddingModel   : "text-embedding-3-small"
} )

// Step 2: Ingest documents
result = aiDocuments( "/docs/products", {
    type      : "directory",
    recursive : true,
    extensions: [ "md", "txt", "pdf" ]
} ).toMemory(
    memory  : vectorMemory,
    options : { chunkSize: 1000, overlap: 200 }
)

println( "Loaded #result.documentsIn# documents as #result.chunksOut# chunks" )

// Step 3: Create agent with vector memory
agent = aiAgent(
    name        : "Product Support",
    description : "Product documentation specialist",
    instructions: "Answer questions using the product documentation. Always cite sources.",
    memory      : vectorMemory
)

// Step 4: Query — agent automatically retrieves relevant docs
response = agent.run( "How do I configure SSL certificates?" )
```

## Multi-Source RAG Agent

Combine multiple knowledge bases by passing an array of memories:

```javascript
productDocs = aiMemory( "chroma", { collection: "product_docs" } )
apiDocs     = aiMemory( "chroma", { collection: "api_docs" } )
faqMemory   = aiMemory( "chroma", { collection: "faq" } )

// Ingest each source
aiDocuments( "/docs/products", { type: "directory" } ).toMemory( productDocs )
aiDocuments( "/docs/api",      { type: "directory" } ).toMemory( apiDocs )
aiDocuments( "/docs/faq.md",   { type: "markdown"  } ).toMemory( faqMemory )

// Agent searches across all knowledge bases
agent = aiAgent(
    name    : "Knowledge Assistant",
    memories: [ productDocs, apiDocs, faqMemory ]
)

response = agent.run( "Explain the authentication API" )
```

## RAG Agent with Real-Time Tools

Combine document retrieval with live data access:

```javascript
docMemory = aiMemory( "chroma", { collection: "documentation" } )
aiDocuments( "/docs", { type: "directory" } ).toMemory( docMemory )

statusTool = aiTool(
    "check_system_status",
    "Check current system status and metrics",
    () => getCurrentSystemStatus()
)

agent = aiAgent(
    name        : "System Assistant",
    instructions: "Use docs for general questions. Use the status tool for real-time data.",
    memory      : docMemory,
    tools       : [ statusTool ]
)

// Agent selects the right approach per query
agent.run( "What are the system requirements?" )  // → uses docs
agent.run( "Is the system currently running?" )   // → uses statusTool
```

## 📚 For Full RAG Capabilities

This page covers agent-level RAG setup. For deeper detail on:

* **Document loaders** (PDF, CSV, JSON, XML, HTTP, web crawlers, SQL, etc.) — see [Document Loaders](/rag/document-loaders.md)
* **Chunking strategies, batch loading, async ingestion** — see [RAG Guide](/rag/rag.md)
* **Vector memory configuration** (Pinecone, Chroma, Qdrant, pgvector, etc.) — see [Vector Memory Systems](/main-components/memory/vector-memory.md)

## Related Pages

* [Memory Management](/main-components/agents/memory.md) — Memory types and per-call identity routing
* [Vector Memory Systems](/main-components/memory/vector-memory.md) — Full vector store configuration
* [Document Loaders](/rag/document-loaders.md) — All loader types
* [RAG Guide](/rag/rag.md) — End-to-end RAG implementation


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://ai.ortusbooks.com/main-components/agents/rag.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.