retailerapi
Unified product data across major US retailers (Walmart, Amazon, eBay, Target, Best Buy, Lowe's, Home Depot): lookups, price history, sellers, reviews.
@retailerapi/mcp
Model Context Protocol server for retailerapi.com — a unified product-data API covering major US retailers. Two tools your AI agent can call directly: product lookups and live offers.
Works with Claude Desktop, Claude Code, Cursor, and any other MCP-compatible client over stdio.
Covered retailers: Walmart, Amazon, eBay, Target, Best Buy, Lowe's, Home Depot. Set include_cross_retailer=true on a product lookup to surface cached cells for every retailer we have for that UPC.
Quick start
1. Get an API key
Sign in at app.retailerapi.com and create a key on the API Keys page. Keys look like rk_live_…. Free tier is 1,000 lookups/month — no card.
2. Add the server to your MCP client
Claude Desktop
Edit claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/claude_desktop_config.json, Windows: %APPDATA%\Claude\claude_desktop_config.json) and add:
{
"mcpServers": {
"retailerapi": {
"command": "npx",
"args": ["-y", "@retailerapi/mcp"],
"env": {
"RETAILERAPI_KEY": "rk_live_your_key_here"
}
}
}
}
Restart Claude Desktop. The retailerapi tools will appear in the tool picker.
Claude Code
claude mcp add retailerapi npx -y @retailerapi/mcp \
--env RETAILERAPI_KEY=rk_live_your_key_here
Cursor
Add to ~/.cursor/mcp.json (or the project-level .cursor/mcp.json):
{
"mcpServers": {
"retailerapi": {
"command": "npx",
"args": ["-y", "@retailerapi/mcp"],
"env": {
"RETAILERAPI_KEY": "rk_live_your_key_here"
}
}
}
}
Generic stdio
RETAILERAPI_KEY=rk_live_your_key_here npx @retailerapi/mcp
The process speaks MCP over stdio (newline-delimited JSON-RPC on stdin/stdout). Logs go to stderr.
Tools
lookup_product
Resolve any identifier (UPC / EAN / ISBN / GTIN / Amazon ASIN / retailer item_id) into a normalized product summary. Base call (1 token) returns: title, brand, image, current price, identifiers, weight, dimensions, MSRP, description, categories, full price history, aggregated stats, retailer_links (free 'where to find it'), Bucket-1 facts (sold_tag, estimated_sales, is_best_seller, pack_count, hazmat), and computed marketplace fees (referral_fee_usd, wfs_fee_usd). Fees are FREE in base call — Keepa parity.
Set include_cross_retailer=true to add the cross_retailer block — a map keyed by retailer slug of cached per-retailer cells (price, in_stock, Bucket-1 fields) for every retailer we have for this UPC (+2 tokens). Read-only over our cache. Set include_seller_context=true to add live seller-side state (is_restricted, WFS eligibility) on marketplace retailers (+3 tokens).
To force a fresh scrape of a specific retailer (bypassing cache), call with retailer=<slug> and force_refresh=true. This is the only way to force fresh data from the API.
Barcode lookups also return a diagnostic _meta block with the source retailer for each top-level field (including weight_lbs_source and dimensions_source — useful when one retailer's catalog is missing physical specs and another retailer backfills them) and a data_quality_score (0.0–1.0).
Package vs assembled. Retailers that distinguish boxed-for-shipping weight from product weight populate weight_assembled_lbs + weight_package_lbs (and parallel dimensions_assembled + dimensions_package). Top-level weight_lbs / dimensions are the derived "best available" — assembled wins, package fills in, plain weight is the last resort. Retailers exposing only one weight populate weight_lbs and leave the explicit pair as null.
| Field | Type |
|---|---|
identifier | string (required) |
identifier_type | "UPC" | "EAN" | "ISBN" | "GTIN" | "ASIN" | "item_id" (optional — auto-detect if omitted) |
include_cross_retailer | boolean (optional — default false) — +2 tokens, read-only |
include_seller_context | boolean (optional — default false) — +3 tokens |
retailer | string (optional) — anchor to a specific retailer slug |
force_refresh | boolean (optional — default false) — only valid with retailer; bypass cache + force fresh scrape |
Example prompts:
- "Look up UPC 045496590161 — what's the brand, price, and Walmart referral fee?"
- "Find UPC 194629116676 across every retailer — who has it cheapest?"
- "What's the WFS fee on this product? Are there any seller restrictions on Amazon?"
get_offers
List current marketplace sellers on a product, including price, in-stock state, and which seller owns the buy box.
| Field | Type |
|---|---|
item_id | string (required) |
Example prompt: "Who has the buy box on item 1689065034 and what's the next-cheapest seller?"
Errors
Tool calls return structured JSON errors instead of crashing the agent:
| Status | Error code | Meaning |
|---|---|---|
| 401, 403 | unauthorized | API key invalid or missing scope. Check RETAILERAPI_KEY. |
| 404 | not_found | Product or item_id not found. |
| 429 | rate_limited | Quota or burst limit hit. Includes retry_after_seconds. |
| 5xx | upstream_error | Backend issue. Retry shortly. |
| — | missing_api_key | RETAILERAPI_KEY env var not set. Pointer to docs included. |
Environment
| Variable | Required | Default |
|---|---|---|
RETAILERAPI_KEY | yes | — |
RETAILERAPI_BASE_URL | no | https://api.retailerapi.com/v1 |
Develop locally
pnpm install
pnpm --filter @retailerapi/mcp build
RETAILERAPI_KEY=rk_live_… node packages/mcp/dist/index.js
The MCP Inspector (npx @modelcontextprotocol/inspector) is the easiest way to exercise the tools manually.
License
MIT
संबंधित सर्वर
WhisperGraph MCP
Open-source, self-hostable MCP server for WhisperGraph — a graph of 7.39B nodes / 39B edges mapping DNS, BGP, GeoIP, WHOIS, and threat intelligence. Six read-only tools (Cypher query + schema introspection + threat assessment), six resources, eight investigation prompts. stdio and Streamable HTTP transports.
HemmaBo Federation MCP Server
AI-native vacation rental infrastructure — search properties, check availability, get quotes, and complete direct bookings via MCP. Supports multi-language, real-time calendar sync, and dynamic staircase pricing across a federated host network.
Relay-gateway
Relay is a desktop application for managing Model Context Protocol (MCP) servers. It provides a user-friendly interface to configure, enable/disable, and export MCP servers for use with Claude Desktop and other AI applications.
SketchUp MCP Server
Control SketchUp with AI. MCP (Model Context Protocol) server that allows AI assistants like Claude, Cursor, and Gemini to programmatically create 3D models in SketchUp.
AstraCipher
Cryptographic identity MCP server for AI agents using W3C DIDs, Verifiable Credentials, and NIST post-quantum cryptography (ML-DSA-65 FIPS 204).
Fiken MCP Server
MCP server for Fiken — Norway's popular accounting software. Manage invoices, contacts, journal entries, and bank accounts via AI agents.
Mnemex
Mnemex is a Python MCP server that provides AI assistants with human-like memory dynamics through temporal decay and natural spaced repetition, storing memories locally in human-readable JSONL and Markdown formats.
Solematica
Solar estimates, provider comparison, energy prices and blog for the Italian market. 6 tools, 5 free. Compare providers, analyze roofs via Google Solar API, check PUN/ARERA prices.
Decision Anchor
Non-judgmental external anchoring for AI agent decisions. Records accountability boundaries.
Trading MCP Server
An intelligent trading assistant that fetches live stock prices using the Yahoo Finance API.