Semantic search and AI-powered Q&A over pre-indexed scripture, philosophy, and classical literature. Quran, Bible, Plato, Confucius, and 26 more — no ingestion pipeline, no fine-tuning. One API call.
import requests r = requests.post( "https://api.sourcetruth.io/v1/answer", headers={"X-API-Key": "st_..."}, json={ "question": "What does the Quran say about patience?", "corpus_codes": ["quran", "sahih-bukhari"], }, ) print(r.json()["answer"]) # → Grounded answer with Quran citations
const res = await fetch( "https://api.sourcetruth.io/v1/answer", { method: "POST", headers: { "X-API-Key": "st_...", "Content-Type": "application/json", }, body: JSON.stringify({ question: "What does the Quran say about patience?", corpus_codes: ["quran", "sahih-bukhari"], }), } ); const { answer } = await res.json();
curl -X POST \ https://api.sourcetruth.io/v1/answer \ -H "X-API-Key: st_..." \ -H "Content-Type: application/json" \ -d '{ "question": "What does the Quran say about patience?", "corpus_codes": ["quran", "sahih-bukhari"] }'
We did the hard part — sourcing, cleaning, chunking, and embedding 30 canonical corpora so you don't have to.
Vector + full-text search with reciprocal rank fusion. Returns ranked passages with tradition, language, and citation metadata. Free on all tiers.
Retrieval-augmented generation via Claude. Every factual claim is tied to a specific retrieved passage — no hallucinated scripture references, ever.
Islam, Buddhism, Christianity, Judaism, Hinduism, Sikhism, and more. Arabic Quran, Sanskrit epics, classical Greek philosophy — ready now.
Skip the ETL, chunking, embedding, and vector DB setup. One API key and you're querying 87,000+ chunks across 13 traditions in under a minute.
Query in English and retrieve Arabic Quran text, Classical Chinese, Tibetan, Punjabi/Gurmukhi, and Sanskrit. Cross-lingual retrieval built in.
Issue a free API key in one POST call — no credit card, no dashboard signup. Upgrade to paid tiers in-app via Stripe when you need more volume.
Every corpus is sourced from authoritative open-access editions and verified for commercial use. Pass corpus_codes to scope any query.
Full corpus list with licence details at GET /v1/corpora
No infrastructure to provision. No data to prepare. Works with any HTTP client.
One POST /v1/keys call with your name and email. No credit card. Your key arrives in the response — prefix st_.
Call /v1/search with a natural language query. Optionally scope to a tradition, language, or specific corpus code. Results in under 100ms.
Every result includes the passage text, canonical citation (e.g. 2:255, SN 56.11), tradition, language, and relevance score.
Start free. Scale when you need to. Upgrade or cancel in one API call.
/search requests/day