The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill acts as an aggregator for external data retrieved from MCP servers and web search results. This creates a surface for indirect prompt injection where instructions embedded in pharmaceutical literature or trial data could theoretically influence the agent's output.
Ingestion points: Data returned from ls_X_search and ls_X_fetch tools, as well as results from the web search tool.
Boundary markers: The instructions do not specify explicit delimiters (like XML tags or markdown blocks) to isolate retrieved research data from the system prompt.
Capability inventory: The skill primarily focuses on data retrieval and report generation; it does not contain dangerous capabilities such as file writing, credential access, or arbitrary command execution.
Sanitization: No specific sanitization or validation logic is defined for the external data before it is integrated into the final report.

disease-investigation