The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's setup guide instructs the user to execute the claude mcp add command. This is a standard platform-specific operation to register Model Context Protocol servers. The command points to legitimate vendor-owned endpoints under the patsnap.com domain.
[EXTERNAL_DOWNLOADS]: The skill references multiple external endpoints (e.g., connect.patsnap.com) to fetch pharmaceutical data, clinical trials, and patent information. These resources are owned by the skill author, Patsnap, and are required for the skill's documented functionality.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface for indirect prompt injection because it ingests and processes untrusted data from academic papers, patents, and web search results. However, the skill implements a highly structured, modular workflow for processing this information, which serves as a mitigation against adversarial instructions embedded in the source data.

pharmaceuticals-exploration-zhcn