The Agent Skills Directory

[SAFE]: No security issues detected. The skill consists of documentation for integrating the Gemini Omni Flash video generation model.
[EXTERNAL_DOWNLOADS]: The skill references official Google AI documentation and blog posts from ai.google.dev and blog.google. These references are informative and point to well-known, established domains.
[CREDENTIALS_UNSAFE]: The skill correctly identifies the need for GEMINI_API_KEY or GOOGLE_API_KEY via environment variables (env_any) in the metadata, following secure practices for secret management rather than hardcoding credentials.
[INDIRECT_PROMPT_INJECTION]: The skill describes a tool that processes user-supplied prompts and reference images. While this creates a potential surface for indirect prompt injection at the model level, the skill documentation itself is benign and does not contain implementation code that would introduce vulnerabilities.

gemini-omni