topaz

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly instructs the agent to ask the user to paste their Pixazo API key into the conversation (and to read it from the conversation), then save/use it—causing the secret to be present in the LLM context and handled directly, which is an exfiltration risk.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs the agent to fetch https://www.pixazo.ai/models/topaz.md (and its HTML form https://www.pixazo.ai/models/topaz) at runtime to obtain exact parameter names/accepted values, meaning remote content would directly control how the agent constructs requests/instructions.