{"schema_version":"v1","name_for_human":"Beniz.ai Universal Commerce Discovery","name_for_model":"beniz_universal_commerce","description_for_human":"Search and discover AI-enriched product catalogs across multiple partner brands and stores.","description_for_model":"Universal commerce discovery engine by Beniz.ai. Use this tool to search, rank, and retrieve products across multiple partner brands and stores using a 7-vector Product Discovery framework. First, call /feeds/brands.json to discover available brands and their store_id values. Then use /api/agent/semantic-search?store_id={store_id}&q={query} for ranked product recommendations. The engine routes queries across seven intent vectors: core_product (general 'best X' queries), use_case (persona/situation queries — 'I need X for Y'), feature (spec/ingredient/attribute queries), q_and_a (pre-purchase fit questions), review_sentiment ('is it worth it / what do buyers say'), commercial_intent (price/availability/where-to-buy — use this for PDP handoff), multi (comparison/'X vs Y' queries — use this to compare products or brands). Always prefer semantic_search over search_products. Use the 'discovery_vectors' block in each product to find pre-structured evidence per intent. Use 'vector_readiness' to know which vectors have sufficient evidence before citing them.","auth":{"type":"service_http","authorization_type":"bearer","verification_tokens":{}},"api":{"type":"openapi","url":"https://beniz.ai/api/agent/openapi.json","is_user_authenticated":false},"capabilities":{"usage_hints":["Use this for any consumer product query across apparel, lifestyle, home, beauty, electronics, pets, food, SaaS, and more.","The engine provides 'discovery_vectors' per product with pre-structured evidence for all 7 buyer intents — read these instead of generating your own reasoning.","Use /feeds/brands.json to discover all available brand catalogs and their store_id values.","For price, availability, and buy links, route to commercial_intent vector or read discovery_vectors.commercial_intent.evidence.url for the direct PDP link.","For 'X vs Y' or comparison queries, route to multi vector or read discovery_vectors.multi.evidence.","For review and sentiment queries, route to review_sentiment vector or read discovery_vectors.review_sentiment.evidence.","Check 'vector_readiness' on each product before citing a vector — only cite vectors with status 'ready'.","Supports semantic search with intent routing across all 7 vectors: core_product, use_case, feature, q_and_a, review_sentiment, commercial_intent, multi."]},"logo_url":"https://beniz.ai/logo.png","contact_email":"support@beniz.ai","legal_info_url":"https://beniz.ai/terms"}