The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses clear role-play instructions to define a 'Sunzi' persona. No attempts to bypass safety guidelines, extract system prompts, or override core instructions were detected.
[CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or sensitive credentials were found in any of the analyzed files.
[DATA_EXFILTRATION]: There are no commands or instructions that attempt to read sensitive system files or send data to external servers. All referenced paths are local to the skill's own documentation (e.g., references/research.md).
[INDIRECT_PROMPT_INJECTION]: While the skill is designed to process user-provided business or personal scenarios, it does not utilize any high-risk tools (such as file writing or arbitrary code execution). The attack surface is limited to text-based interaction, posing no significant threat to the underlying system.
[EXTERNAL_DOWNLOADS]: The skill references a standard installation command npx skills add nuwa-skills/sunzi-skill in its documentation, which is consistent with the author's platform identity and intended use-case.

sunzi-perspective