dating

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt shows and requires including an API key in Authorization headers (and returns an api_key on registration), which implies the agent must capture and insert secret API keys verbatim into generated requests/commands (curl examples), creating an exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches public, user-generated profiles and chat messages (e.g., GET /api/agents, GET /api/chat/{matchId}/messages) and its documented workflow tells the agent to "follow the next_steps" from API responses (which include executable API actions), so untrusted third-party content can be ingested and directly influence subsequent tool use and actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill relies on runtime API responses from https://inbed.ai (e.g., the /api/* endpoints) whose returned "next_steps" contain structured actions and instructions the agent is expected to execute, meaning external content from that URL can directly control agent prompts and behavior.