Skip to main content
Glama

Logimu

seller

Read-onlyIdempotent

Amazon SELLER profile by seller id, optionally with the products the seller sells. USE WHEN the user asks about a third-party seller / merchant / storefront (who they are, how they are rated, what they sell) - the seller id is the 'A...' code in /sp?seller=... or the seller_id on any offer the product tool returns. RETURNS seller (name, rating, lifetime and 12-month rating counts, positive %, is_amazon), feedback_summary per period (30 d / 90 d / 12 m / lifetime: average rating + count - Amazon no longer publishes positive/neutral/negative splits), rating_histogram, recent_feedback (up to 5 entries as date / rating / struck ONLY - never the buyer's name or text), and business (legal name, registration / VAT numbers, address, phone, email, representative) where the seller publishes it on Amazon - all 11 marketplaces; au and ca pages carry no business block, so it is null there. With products=true, also seller_products: mode=live the storefront as Amazon shows it now (the serp object, SECONDS), mode=cache the products we OBSERVED the seller offering (25 per page, buy-box days, observed price range, <100 ms). MODES for the profile: cache = our stored observation - the seller page as our tracker last read it (the full profile; sellers are re-read about every 14 days), or the page fetched in the last 24 h, or for sellers not stored yet the warehouse rating record; live = fetch the seller page now (5-20 s, up to ~30 s); auto (default) = stored when younger than max_age_days (default 14), else live. tracking{} says since when we have observed the seller (observed_first_seen), when last (last_observed) and how often - OBSERVED, never 'founded' or 'active since'. With history=true also history{}: the seller's rating change points (positive %, star rating, counts per 30 d / 90 d / 12 m / lifetime) and storefront-size points, oldest first, UTC - change points, each value holds until the next; the series starts 2026-10-04 at the earliest, so expect 1-2 points per seller for now. HONESTY: a 404 means Amazon has no seller page for that id on that marketplace (or we never observed it, in cache mode); a 502 is a collection failure, not a fact about the seller - retry. An empty storefront does not confirm the seller exists. LOOKUP: give seller_id, or name (resolved through observed buy-box data to the id, free). COST: paid plan required - the tool is listed only for keys on a paid plan (no free lane). Profile: 1 credit live, 0.5 from cache, auto bills what was served; history=true +0.5 once when at least one point comes back. Products: 1 credit per storefront page fetched (live) or per page of 25 observed products (cache). Feedback, business details and tracking are included at no extra charge. Not found, unavailable, blocked, empty storefront, empty history: 0. DERIVED (2026-10-05): brands=true adds the operator footprint - the brands this seller was observed winning the buy box on, with buy-box days (+1 credit per page of 25); network=true adds the connected sellers that win on the same brands (+1 credit).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
modeNoProfile lane (see description). products=true uses products_mode.auto
nameNoInstead of seller_id: the seller's display name as Amazon prints it (e.g. 'Total Office Solutions'). Resolved through observed buy-box data to the seller id; a name we have never observed, or one with no id on record, returns guidance - nothing is billed.
pageNoProducts page (cache: 25 per page; live: Amazon's page 1-10).
brandsNoAlso return the operator footprint: brands this seller was observed winning the buy box on (16 months), 25 per page (brands_page). +1 credit per page.
countryNoAmazon marketplace. No Walmart - seller pages are Amazon-only.us
historyNoAlso return history{} (rating + storefront-size change points, oldest first, UTC). +0.5 credit when non-empty.
networkNoAlso return connected sellers: sellers winning the buy box on the same brands, strongest overlap first (max 50). +1 credit.
max_pageNolive products only: auto-paginate up to this page (max 10); billed per page fetched.
productsNoAlso return the seller's products (billed separately, see description).
seller_idNoAmazon seller id: 'A' + 8-24 letters/digits.
brands_pageNo
max_age_daysNoauto only: how old the stored observation may be before fetching live (default 14 = the tracker's cadence).
products_modeNocache = observed products (fast); live = the storefront now (seconds).cache
history_pointsNohistory=true: the latest N change points per series (1-5000).

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.8/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already mark this read-only/idempotent, and the description still adds substantial non-structured context: HTTP status semantics (404 = no page or never observed, 502 = collection failure to retry), the fact that an empty storefront does not confirm existence, caching/refresh cadence (~14 days), and a full per-operation credit model. Nothing here is derivable from the annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Long, but front-loaded with purpose, USE WHEN, RETURNS, and modes before the cost block. Most sentences carry unique operational information; a few (mode/products_mode behavior) restate what the schema enums already say, which keeps it from a 5.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

No output schema exists, so the description carries the full return burden and does so explicitly: seller fields, feedback_summary periods, rating_histogram, recent_feedback contents and privacy limits, business block behavior per marketplace, and the products/history add-ons. An agent has everything needed to call and interpret results.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 93%, so the schema already documents most parameters. The description still adds cross-parameter meaning the schema cannot: how mode=auto bills 'what was served', the live fetch timing (5-20s, up to ~30s), the cache page size of 25, and the +0.5/+1 surcharges for history/brands/network.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with a specific verb+resource: 'Amazon SELLER profile by seller id, optionally with the products the seller sells.' It also distinguishes itself from the sibling product tool by noting the seller id appears 'on any offer the product tool returns,' so an agent can route without opening either schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit 'USE WHEN the user asks about a third-party seller / merchant / storefront (who they are, how they are rated, what they sell)' plus a concrete LOOKUP rule (seller_id or name). Costs and modes are also tied to concrete situations.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.