The Agent Skills Directory

[SAFE]: The skill contains instructional content for providing astrology-based wellness guidance using internal references and tools. No malicious patterns, obfuscation, or unauthorized data access were found.- [INDIRECT_PROMPT_INJECTION]: The skill processes output from an internal shared evaluator, which represents a potential surface for indirect prompt injection. However, the skill explicitly defines boundaries, instructing the agent to use only cited markers and deferring all medical concerns to qualified professionals. These constraints mitigate the risk of the agent being misled by external data.

health