jimeng-opencli-text2video

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the opencli command-line tool, which is the primary mechanism for interacting with the Jimeng service for video generation and history retrieval.
  • [PROMPT_INJECTION]: The instructions include specific negative constraints, such as prohibiting the use of the dreamina CLI, to ensure the agent operates within the intended scope for standard members.
  • [PROMPT_INJECTION]: The skill possesses an indirect injection surface by passing user-provided prompts directly to shell commands.
  • Ingestion points: The <prompt> argument in opencli jimeng generate-video (SKILL.md).
  • Boundary markers: No specific delimiters or safety warnings are used to wrap the interpolated prompt.
  • Capability inventory: Execution of shell commands via the opencli binary (SKILL.md).
  • Sanitization: The instructions do not specify any sanitization or validation logic for the input prompt.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 03:33 AM
Security Audit — agent-trust-hub — jimeng-opencli-text2video