video-content-replace
Fail
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: HIGHCOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill's core execution logic involves interpolating user-controlled input (the
promptparameter) directly into a shell command template:meitu video-content-replace --video_url {url} --prompt "{修改描述}". If the user input contains shell metacharacters such as backticks, semicolons, or pipes, it could lead to arbitrary command execution on the host system. - [CREDENTIALS_UNSAFE]: The skill requires explicit read access to
~/.meitu/credentials.json, which containsMEITU_OPENAPI_ACCESS_KEYandMEITU_OPENAPI_SECRET_KEY. While these are vendor-specific credentials, the pattern of reading raw secret files from the home directory is a security concern. - [DATA_EXPOSURE & EXFILTRATION]: The skill has the combined capability to read sensitive credentials and execute shell commands with network access (via the
meituCLI or potentially injected commands). This creates a path for data exfiltration if the command injection vulnerability is exploited.
Recommendations
- AI detected serious security threats
Audit Metadata