chengfeng-video-styles-installer
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The documentation instructs users to install the skill using
npxdirectly from the author's GitHub repository, which involves downloading and executing remote code during the setup process. - [COMMAND_EXECUTION]: The installation script (
bin/install.js) performs multiple file system operations, such as creating directories, copying files, and managing symbolic links in the user's home directory (~/.claudeand~/.codex). - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted user data (text, images, and videos) to generate its output, creating a potential surface for indirect prompt injection.
- Ingestion points: User-provided video scripts, copy, and reference media files processed by the agent.
- Boundary markers: None identified; the instructions do not specify techniques to isolate user input from the agent's logic.
- Capability inventory: The agent can write files to the local system (generating HTML samples).
- Sanitization: No explicit sanitization or validation of the input content is performed beyond the agent's standard operating procedures.
- [DYNAMIC_EXECUTION]: The core functionality of the skill is to have the agent dynamically generate and execute HTML/JavaScript code (utilizing Canvas, SVG, or three.js) at runtime based on the user's specific request.
Audit Metadata