The Agent Skills Directory

[SAFE]: The skill instructions and connectivity guides reference official domains (patsnap.com, connect.patsnap.com) that belong to the skill's stated author, representing legitimate service integration.
[SAFE]: The documentation for setting up MCP servers correctly uses generic placeholders (sk-xxxxxxxxxxxx) and directs users to the official portal for API key generation, adhering to safe credential handling practices.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize large volumes of untrusted data from external sources, which presents a risk of indirect prompt injection.
Ingestion points: Data retrieved from external patent databases, clinical trial results, and web searches via tools such as ls_patent_fetch, ls_clinical_trial_fetch, and ls_news_fetch.
Boundary markers: The agent is not instructed to use specific delimiters or protective boundary markers when processing external document content.
Capability inventory: The agent has the capability to perform multi-step analysis, integrate cross-domain data, and generate detailed research reports based on external inputs.
Sanitization: There are no explicit instructions or tool-level mechanisms described for sanitizing or filtering the content of retrieved research documents before they are processed by the agent.

pharmaceuticals-exploration