The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were identified in the skill's instructions or supporting documentation.
[COMMAND_EXECUTION]: The analysis files mention the use of curl and opencli for research purposes to analyze video cover styles on Bilibili. These are documented as past research steps and do not constitute commands for the agent to execute during the skill's operational runtime. Bilibili is recognized as a well-known service.
[PROMPT_INJECTION]: The skill processes user-provided script files or topic descriptions to extract visual keywords. Although the instructions do not specify boundary markers for this untrusted input, the resulting output is limited to static text for image generation prompts, presenting a low risk for exploitation via indirect prompt injection.

laohan-fengmianqiuzhi