dreamina-cli

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs using a headless OAuth device flow and explicitly shows running a wrapper with "--device-code <device_code>" and also exposes fully-expanded cli_args in wrapper output, which requires the agent to read and then emit short-lived device codes or other auth tokens verbatim (an exfiltration risk).

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

• Third-party content exposure detected (low risk: 0.10). The runtime workflow invokes local wrapper scripts (e.g., scripts/* calling dreamina_wrapper.main_for_command), which only ingest outsider text when it is passed as explicit CLI parameters like --prompt or --name; there is no required workflow step that passively reads outsider-authored free text from an external feed/queue/task list.

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

• Potentially malicious external URL detected (high risk: 0.90). The README contains an installation command that fetches and pipes remote code to a shell (https://jimeng.jianying.com/cli), which is a high-risk remote-code-execution dependency used to install the required dreamina CLI for this skill.