video-creator
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input for video topics and platforms which is then interpolated into prompts for generating scripts and social media copy.
- Ingestion points: User input variables such as {topic}, {platforms}, and {style} are used in Step 2.
- Boundary markers: The skill lacks explicit boundary markers or instructions to ignore embedded commands within user-provided content.
- Capability inventory: The skill possesses capabilities for file system writing (Write tool) and browser automation (Playwright MCP).
- Sanitization: No sanitization or validation of the user-provided topic or description is present before interpolation.
- [COMMAND_EXECUTION]: The skill leverages the Playwright MCP server to perform browser automation, including navigating to URLs, clicking elements, and typing text on social media platforms like Douyin and YouTube. These actions are performed to automate the publishing workflow as described in the skill's core functionality.
Audit Metadata