browser-cdp

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly shows commands to eval localStorage.getItem("token") and document.cookie to extract auth tokens/cookies, which would require the agent to read and return secret values verbatim (high exfiltration risk).

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (medium risk: 0.60). This skill explicitly instructs killing user Chrome processes and deleting/overwriting a Chrome debug profile (pkill/taskkill, --reset clears ~/chrome-debug-profile), which are destructive host-state changes — even though it requires user consent and has a detect-only mode, it still pushes the agent to perform potentially data‑lossing operations on the machine.