The Agent Skills Directory

[COMMAND_EXECUTION]: The utility script scripts/jimureport_gen.py invokes subprocess.run to call scripts/jimureport_creator.py. This behavior is localized within the skill's own directory structure and is used purely for internal workflow orchestration.
[CREDENTIALS_UNSAFE]: The file scripts/jimureport_core.py contains a hardcoded string dd05f1c54d63749eda95f9fa6d49v442a as a fallback signing secret. This is a well-known default value for the JimuReport platform's request authentication and does not represent an exposure of sensitive user credentials.
[SAFE]: The skill communicates exclusively with vendor-controlled domains (e.g., api.jeecg.com) for legitimate reporting functions.
[SAFE]: Clear instructions are provided within SKILL.md to ensure that sensitive information like database passwords are requested directly from the user rather than being stored or retrieved from the agent's long-term memory.

jimureport