The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Node.js script (scripts/calculate.js) to perform its primary function. The execution is handled via a structured JSON pipe, which is a standard method for local tool interaction.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided birth parameters (date, time, longitude). It includes robust validation logic to reject malformed data, script tags, or non-numeric coordinates, mitigating risks from untrusted user input.
[DATA_EXPOSURE]: The skill only accesses user-provided parameters for the purpose of the calculation. There is no evidence of access to sensitive system files, environment variables, or hardcoded credentials.
[REMOTE_CODE_EXECUTION]: All logic is contained within the skill's own directory. There are no external downloads or remote script executions observed.

ziwei