# LLMRank

> Transparent AI-SEO backend. Merchants publish products and services here; AI
> models and agents query them as tools and get a ranked, fully explainable answer.

## Ranking
Every result carries an LLMRank-Score built from six weighted components:
Relevance 30%, Completeness 20%, Freshness 15%, Availability 10%, Trust 10%,
Performance 15%. The exact per-component score, explanation and improvement
suggestion is available for any listing and any query.
Docs: https://thellmrank.com/docs/ranking

## Tools (MCP, Streamable HTTP)
https://thellmrank.com/mcp — authenticate with an LLMRank API key as a bearer token.
- search_products(query, filters) — ranked search with explanations
- get_product_details(product_id|sku) — full listing detail
- compare_products(product_ids) — side-by-side comparison
- check_availability(product_ids) — live stock and price
- get_merchant_info(merchant_id) — trust status and policies
- explain_ranking(product_id, query) — why this ranked here

## REST API
- GET|POST https://thellmrank.com/api/v1/search
- GET https://thellmrank.com/api/v1/products
- GET https://thellmrank.com/api/v1/products/{id}?explain={query}
- POST https://thellmrank.com/api/v1/feed/update
- POST https://thellmrank.com/api/v1/click

## Public feed
https://thellmrank.com/api/v1/feed.json — ACP-format catalog with enable_search and
enable_checkout flags on every item.
https://thellmrank.com/api/v1/feed.json.gz — the same feed as a gzipped file, for crawler-based ingesters.

## Discovery
https://thellmrank.com/.well-known/llmrank.json
https://thellmrank.com/.well-known/mcp/server.json — MCP registry descriptor
https://thellmrank.com/llms-full.txt — full reference: tool inputs, rate limits, etiquette
https://thellmrank.com/developers — copy-paste client setup for Claude, Python, TypeScript, LangChain, OpenAI Agents
https://thellmrank.com/sitemap.xml