create-hooks

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly logs and forwards raw command text (e.g., .tool_input.command and $ARGUMENTS) into shell commands, files, and prompt hooks, so any API keys or passwords embedded in those commands would be captured and emitted verbatim, creating a high exfiltration risk.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The skill explicitly enables executing arbitrary shell commands as hooks (with access to project and user paths) which can modify files or run privileged operations on the host and does not prohibit or constrain actions like editing system files or creating accounts, so it can be used to compromise machine state.