mcp.pangolinfo.com

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.

This server isn't on gate's handpicked list. That's not necessarily bad, but there's no third-party signal about it — review its tools carefully.

The 'url' parameter accepts any https:// URI (schema only enforces format:uri, not an amazon.com domain restriction) and is described as a 'power-user escape hatch' for arbitrary filter/sort/pagination construction. While the description strongly implies Amazon-only use and says the parser 'must match', there is no schema-level enforcement that the URL is actually an Amazon domain. This is a broader capability than the other purpose-built, narrowly-scoped Amazon tools, and could be misused to fetch/scrape unintended or attacker-supplied URLs if the backend doesn't strictly validate domain match with parserName.

The description tells the AI 'for account balance or remaining credits (CONTRACT §9 forbids exposing account endpoints via MCP)' — i.e., it instructs the agent to withhold certain categories of information from the user based on an opaque, unverifiable external 'contract' clause that the user has no way to inspect. While likely a legitimate business policy, embedding this as a directive inside a tool description is a pattern worth flagging since it conditions agent behavior on an unverifiable authority rather than user-visible logic.