The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to perform operations using shell utilities such as exiftool, cwebp, convert, and ffmpeg, which involves direct command-line interaction with the host system.- [COMMAND_EXECUTION]: The skill explicitly provides instructions for the agent to recommend or use sudo to install system-level packages (libimage-exiftool-perl, webp), which constitutes a privilege escalation risk.- [COMMAND_EXECUTION]: The instruction set describes building shell commands by interpolating metadata fields like titles and descriptions into command arguments, which can lead to command injection if malicious strings are present in the source files.- [PROMPT_INJECTION]: The skill ingests untrusted data from image metadata (IPTC/XMP) and SERP results, creating an indirect prompt injection surface.
Ingestion points: Processed image files and Google Images SERP data via the DataForSEO API.
Boundary markers: None defined to isolate external data from instructions.
Capability inventory: Includes file modification and shell command execution across multiple utilities.
Sanitization: No validation or escaping of external content is specified before interpolation into commands or output summaries.

seo-images