seedance-prompt

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (Step 6 in SKILL.md and the README) instructs the agent to fetch and run the Dreamina CLI installer via a public URL (curl -fsSL https://jimeng.jianying.com/cli | bash), i.e., downloading and executing a third‑party script as part of normal operation, which exposes it to untrusted external content that can alter tool behavior.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly runs a runtime install command that fetches and executes remote code—curl -fsSL https://jimeng.jianying.com/cli | bash—to install the Dreamina CLI which the skill uses to generate videos, so the URL is a runtime external dependency that executes remote code.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly instructs the agent to execute system commands and, if the CLI is missing, to run a remote installer via "curl ... | bash" (i.e., perform an installation that modifies the host), which causes the agent to change the machine's state and potentially install software without requiring explicit user-only action.