claude-sonnet-5

SUSPICIOUS. The skill's purpose and capabilities mostly align, and it does not show malware-like install behavior, but it routes prompts and credentials through a Frevana intermediary endpoint instead of official OpenRouter/Anthropic APIs, with limited public verification for the exact endpoint. That makes the data flow and trust model moderately risky rather than benign.