The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill is composed of purely instructional and historical reference markdown files.
[NO_CODE]: The skill does not include any scripts, binaries, or logic that would be executed at runtime. It functions as a set of static prompts for character persona and methodology.
[PROMPT_INJECTION]: The role-play instructions are well-structured and include specific behavioral constraints (e.g., prohibiting the encouragement of radical reforms without consideration of consequences). No attempts to bypass system safety guidelines or extract underlying prompts were detected.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user queries through a historical lens. However, since it has no access to external tools, file system write operations, or network capabilities, the risk associated with processing untrusted data is minimal and confined to the generated text response.

zhangjuzheng-perspective