seo-images
Warn
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on the execution of multiple command-line utilities, including exiftool, cwebp, ImageMagick, and FFmpeg, to perform file processing and metadata manipulation tasks.- [COMMAND_EXECUTION]: The instructions direct the agent to recommend administrative privilege escalation using the
sudocommand to install system-level dependencies on the host environment.- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by ingesting and processing untrusted data from external web pages (such as alt text and file names) which are then used in analytical reports and potentially as arguments in shell commands. - Ingestion points: External website content (fetched from user-provided URLs) and competitive SERP data retrieved via the DataForSEO API.
- Boundary markers: The instructions lack guidance on implementing boundary markers or delimiters to separate untrusted external content from agent instructions.
- Capability inventory: The skill has broad capabilities, including local file system modification, shell command execution, and external tool invocation.
- Sanitization: There are no provided instructions for sanitizing or escaping external strings before they are interpolated into shell command templates, which could lead to command injection.
Audit Metadata