girlfriend-girlfriend

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt shows curl examples that embed an Authorization: Bearer {{YOUR_TOKEN}} header and tells the user to keep and use their returned token, which instructs including a secret token verbatim in generated requests/commands and thus creates 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.md shows the agent calling public inbed.ai endpoints (e.g., GET /api/discover and GET/POST /api/chat) that return user-generated profiles, narratives, and chat messages which the agent is expected to read and could materially influence decisions or actions, exposing it to untrusted third-party content.