SDK Quickstart

This quickstart creates a small support knowledge base, stores it in Qdrant, retrieves the matching chunk, and asks an LLM for a grounded answer. It uses inline Markdown for the first smoke test, then shows the same SDK workflow with documents, websites, and REST API sources.

1. Install

pip install "ragrails[store-qdrant]"
export VOYAGE_API_KEY="..."
export OPENAI_API_KEY="..."
docker run -p 6333:6333 qdrant/qdrant

For website extraction:

pip install "ragrails[url,store-qdrant]"
ragrails setup-url --browser chromium

2. Configure

from ragrails import RagRails

rag = RagRails(
    collection="support",
    vector_store={"provider": "qdrant", "url": "http://localhost:6333"},
    embedding={"provider": "voyage", "model": "voyage-3"},
    llm={"provider": "openai", "model": "gpt-4o-mini"},
)

3. Ingest

ingest() runs extraction, chunking, embedding, and storage.

ingested = rag.ingest(
    markdown=(
        "# Refund policy\n\n"
        "Customers can request a refund within 30 days of purchase. "
        "Refunds are returned to the original payment method within 5 business days."
    ),
)

print({
    "sources": ingested.sources,
    "chunks": ingested.chunks,
    "embedded": ingested.embedded,
    "stored": ingested.stored,
    "failed": ingested.failed,
})

Use the same ingest workflow with other source types:

rag.ingest(docs=["files/refund-policy.pdf"])
rag.ingest(docs={"folder": "files/policies"})
rag.ingest(docs=[
    {"kind": "path", "path": "files/refund-policy.pdf"},
    {"kind": "url", "url": "https://example.com/files/guide.pdf"},
    {"kind": "folder", "folder": "files/policies/"},
])

rag.setup_url(browser="chromium")
rag.ingest(urls="https://example.com/help/refunds")
rag.ingest(urls={"url": "https://example.com/docs", "mode": "full", "max_pages": 25})

rag.ingest(api={"url": "https://api.example.com/refund-policy", "title": "Refund policy"})

4. Query

result = rag.query("How long do I have to request a refund?", retrieval={"top_k": 5})

for chunk in result.items:
    print(chunk.score, chunk.text[:200])

5. Chat

result = rag.chat("How long do I have to request a refund?", history=[])
print(result.answer)
history = result.history

Expected answer:

You can request a refund within 30 days of purchase.

Chat is stateless. Store the returned history and pass it to the next turn.

Ingest

Learn the full SDK ingest workflow.

Query

Search and chat over the stored knowledge base.

Overview

See every SDK pipeline stage.

Extraction

Use documents, websites, and APIs.

Get Started

Workflows

Pipeline

Maintenance

1. Install

2. Configure

3. Ingest

4. Query

5. Chat

Next

Ingest

Query

Overview

Extraction