seo-images
Fail
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONNO_CODE
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for running
sudo apt install, which constitutes a privilege escalation risk by requesting administrative access.\n- [COMMAND_EXECUTION]: The skill invokes several command-line tools (exiftool,cwebp,convert,ffmpeg) using parameters that could be derived from potentially untrusted user inputs or file metadata, creating a risk of command injection.\n- [EXTERNAL_DOWNLOADS]: The skill recommends downloading and installing system packages from external software repositories.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the parsing of untrusted data from external URLs and image metadata.\n - Ingestion points: HTML content from user-specified URLs and metadata fields (IPTC, XMP, EXIF) within image files.\n
- Boundary markers: No delimiters or specific instructions to ignore malicious embedded content are utilized.\n
- Capability inventory: The skill executes file-system modifications and external command-line utilities.\n
- Sanitization: No data validation or sanitization processes are described for the ingestion and processing of external data.\n- [NO_CODE]: The skill refers to external scripts (
scripts/parse_html.pyandscripts/iptc_ai_label.py) that are not provided in the skill package, making its precise behavior and security posture unverifiable.
Recommendations
- AI detected serious security threats
Audit Metadata