dot-dash

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill includes a "token" command that prints the bearer token (and points to the token file path), which would require the agent to read and output a secret value verbatim, creating an exfiltration risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). This skill explicitly captures and streams all active session transcripts and exposes a local WebSocket/HTTP dashboard that supports automatic prompt injection and uses a bearer token stored on disk — functionality that enables deliberate covert monitoring, credential exposure, and remote manipulation of user sessions (especially if the dashboard is reachable beyond localhost), and is therefore high-risk for abuse as a backdoor/surveillance tool.