The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected. The skill contains instructional content and code snippets for interacting with well-known, trusted AI services using their official SDKs. All network operations are directed at legitimate API endpoints for OpenAI, Anthropic, Google, and xAI. While the documentation references future-dated model versions (e.g., GPT-5, Claude 4.6), these appear to be placeholders for future model releases rather than malicious obfuscation.

multimodal-llm