The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill functionality includes storing and processing sensitive personal identification information (PII), specifically "passport numbers," within its data management structure. Handling high-sensitivity government identifiers increases the risk of identity exposure if the agent's context or stored JSON files are compromised.
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection due to its handling of untrusted data inputs and file system capabilities.
Ingestion points: The skill ingests user-provided travel plans, destination details, medication lists, and medical history, which can originate from untrusted external sources.
Boundary markers: The instructions do not define clear delimiters or "ignore embedded instructions" warnings when interpolating these data points into health reports or medical cards.
Capability inventory: The skill is authorized to use Read, Write, Grep, and Glob tools to interact with the file system, specifically targeting data/ and data-example/ directories.
Sanitization: No specific sanitization or validation logic is outlined to prevent malicious instructions embedded in user data from influencing the agent's output or tool usage.
[SAFE]: The skill references health risk data and preventive guidelines from well-known services and organizations, including the World Health Organization (WHO) and the U.S. Centers for Disease Control and Prevention (CDC). References to these authoritative sources are documented neutrally as a standard feature of the health analysis functionality.

travel-health-analyzer