Save money on anything. Tell our shopping agent what you need. It checks every store and finds you the best deal.
Scored across 12 tools
Most tools are clearly partitioned: async starts (ask_deal_ai, check_deal) are explicitly paired with their pollers, and search_products / compare_products / get_product_details / get_product_prices / get_deals target distinct retrieval shapes. The one real overlap is check_deal vs get_product_prices, which both return every store's price for a single product; only the 'was' price check and verdict-versus-history framing separate them. ask_deal_ai also brushes against search_products since both do product research, though the conversational agent framing helps.
Consistent snake_case verb_noun pattern throughout (get_*, list_*, search_*, compare_*, check_*, watch_*, stop_*), which is easy to scan and predict. Minor deviations: ask_deal_ai embeds the server name rather than a resource, and the watch family is inconsistent internally (watch_price for the action, but list_watched / stop_watching for the same concept).
Twelve tools is well within a comfortable range and each one maps to a distinct shopping workflow: search, inspect, compare, price-check, deals feed, agent Q&A, and wish-list management. The async start/poll pairs justify their extra slots rather than bloating the surface.
Strong lifecycle coverage: discover -> inspect -> compare -> price check -> watch -> list -> unwatch, plus a conversational agent and a deals feed, with no dead ends for returned identifiers. Gaps are minor: no tool to toggle the account's price alerts on/off (descriptions only report the state), and no way to cancel an in-flight ask_deal_ai or check_deal.
Asks Deal.ai's shopping agent, for the signed-in Deal.ai account: it researches products, compares US stores' prices and recommends the best picks, as on deal.ai. Give search_id from an earlier answer to ask a follow-up in the same conversation. Answers take one to three minutes: this waits about 20 seconds and returns the answer if it's ready, else a search_id for get_deal_ai_answer. An answer has the agent's words and each pick's best store price with when it was seen, rating, deal.ai page and store link. The picks are watched for price drops for the account, as searches on deal.ai are.
Inputs: question (required), search_id.
Starts a deal check on a product link from any store, or a product's name, and optionally the price the shopper was offered. It finds this exact item (same model, size and generation) at every big US store right now, compares their prices and checks the store's "was" price. Returns a check_id and the check's deal.ai page at once; the check takes 20 to 70 seconds, so call get_deal_check for the result.
Inputs: product (required), price.
Sets two to four products (product_ids from search_products, or deal.ai links) side by side: each one's rating and best current price from a store with when it was seen, their specifications aligned by name where they share them, each product's deal.ai page, and deal.ai's comparison page for a pair where there is one.
Inputs: products (required).
The answer to a question asked with ask_deal_ai: still running (with what the agent is doing), didn't finish, a question the agent asks back, or the answer: the agent's words and each pick's best store price with when it was seen, rating, deal.ai page and store link.
Inputs: search_id (required).
The result of a deal check started with check_deal: still running, or the verdict in one sentence (cheaper elsewhere, the lowest price, the only store), each store's price for the exact item, the "was" price checked, and the check's deal.ai page.
Inputs: check_id (required).
Deal.ai's live deals, checked against other stores and ranked as deal.ai lists them: this hour's deals, or a store's, or those matching what was asked for, optionally by category, discount and price. Each deal's title, price, the earlier or usual price it's measured against and the discount, the store, when it was checked, its deal.ai page and a link to buy it.
Inputs: limit, query, store, category, +2 more.
For one product (a product_id from search_products, or its deal.ai link): its name, brand, model and category, its description, the fact sheet read from a store's page (highlights, what's in the box, returns, what buyers praise and complain about), its main specifications, rating, a few photos, variants and sizes, the best current price from a store with when it was seen, and the product's deal.ai page.
Inputs: product (required).
For one product (a product_id from search_products, or its deal.ai link): every store's current price Deal.ai has, cheapest first, with shipping, condition, stock and when each was seen; the lowest and highest prices tracked; recent daily lows; and the product's deal.ai page.
Inputs: product (required).
The products on the signed-in account's Deal.ai wish list, whose price drops it is e-mailed about: each one's best store price now and where, with when it was seen, its reference price (the price when it was saved or last e-mailed), and its deal.ai page; and whether the account's price alerts are on.
Inputs: no named parameters.
Searches Deal.ai's product price pages, live deals, buying guides and side-by-side comparisons. Returns each page's title, kind, latest price (with the day it was seen), saving or rating where it has one, a product_id for products, and its deal.ai link.
Inputs: query (required), kind, limit.
Removes a product from the signed-in account's Deal.ai wish list, and stops its price-drop e-mails, including those from a search that found it.
Inputs: product (required).
Saves a product to the signed-in account's Deal.ai wish list, so Deal.ai e-mails the account when its price drops (while the account's price alerts are on; this says whether they are). Returns added or already watched, the product's best store price now with when it was seen, and its deal.ai page.
Inputs: product (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Shopping search across 100M+ products, with every retailer's offer and live price in one place.
Find the best deals on Amazon, Flipkart and more Indian e-commerce stores via AI.
Lets AI agents find, try and buy software inside a budget you set. No passwords or card numbers.
Cheapest delivered price for any product: tech, home, shoes, beauty. Read from each store's checkout
Add one secure layer between your agents and this server.
