Pangolinfo Amazon Data MCP

What Pangolinfo Amazon Data MCP MCP server does

Pangolinfo Amazon Data MCP MCP server provides access to 19 data operations covering Amazon, WIPO, and PACER sources. The Amazon tools support product search, product details, reviews, seller storefronts, and Amazon page scraping. Additional operations cover category and niche analysis, bestseller and new-release lists, AI-oriented search results, keyword trends, Alexa for Shopping, local map points of interest, WIPO trademark searches, and PACER patent litigation searches.

The repository includes a Python package named pangolinfo-mcp. It exposes these operations through the PangolinfoMCPClient, so Python code can use the service without an AI client or an MCP SDK. A generic call_tool method is available when code needs to invoke a tool by name, and list_tools or capabilities can expose the available tool set.

How it works

Create a client with a permanent API key, then call the corresponding synchronous method with the required arguments. For example, product details use an ASIN, category methods use identifiers such as a category ID, and text-based searches accept a query or keyword. Results are returned to Python, where the application can inspect or process them. The client also supports use as a context manager and provides close() for cleanup.

Pangolinfo Amazon Data MCP MCP server can also be consumed directly through MCP over Streamable HTTP at https://mcp.pangolinfo.com/mcp. MCP clients send the API key as a Bearer token in the configuration. Clients that require a local stdio process can use the pangolinfo-amazon-data-mcp npm bridge, which forwards requests to the hosted service.

Setup and configuration

Install the Python package with:

pip install pangolinfo-mcp

Obtain a permanent API key from the Pangolinfo Account Center at tool.pangolinfo.com. Python applications pass that key when constructing PangolinfoMCPClient:

from pangolinfo_mcp import PangolinfoMCPClient

with PangolinfoMCPClient(api_key="your_key") as client:
    result = client.search_amazon("wireless earbuds")

For direct MCP connections, configure the hosted URL and an Authorization header containing Bearer followed by the key. The key may alternatively be supplied as an api_key query parameter. MCP configuration is not hot-reloaded, so clients must restart or reconnect after credentials change.

Tools and capabilities

The 19 documented methods are grouped as follows:

• Amazon core: product search, product details, reviews, seller products, and URL scraping.
• Category and niche analysis: niche filters, category filters and search, category children and trees, batch category paths, bestsellers, and new releases.
• Search and SERP AI: AI search, keyword trend time series, and Alexa for Shopping.
• Maps and IP compliance: local map data, WIPO trademark search, and PACER patent litigation search.
• Utility: live capability discovery.

The Python client also supports a generic tool call such as call_tool("search_amazon", {"keyword": "laptop", "page": 2}).

Limitations and notes

The documented Python interface is synchronous; asynchronous support is described as planned rather than currently available. The package has no MCP SDK dependency and requires httpx. Use of the service requires a permanent API key. The hosted MCP configuration must be reloaded by reconnecting the client after an API key change. The supplied material does not specify individual request limits, response schemas, regional coverage, or pricing details beyond stating that the client uses the same engine and pricing as the REST API and MCP server.