The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands with utilities such as exiftool, cwebp, ffmpeg, and convert to process images and metadata. This introduces a command injection risk if user-provided file paths or metadata parsed from external sources contain malicious shell characters.
[REMOTE_CODE_EXECUTION]: The skill references and executes external scripts, including scripts/parse_html.py and iptc_ai_label.py, using a custom runner. As these scripts are not included in the provided file set, their logic and safety cannot be verified.
[EXTERNAL_DOWNLOADS]: The documentation instructs users to install several system packages (libimage-exiftool-perl, webp) via apt install to support the skill's functionality.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its handling of untrusted external content.
Ingestion points: HTML content retrieved from URLs and IPTC/XMP metadata extracted from image files.
Boundary markers: No explicit markers or instructions are provided to the agent to treat external content as untrusted data.
Capability inventory: The skill has the capability to execute shell commands, perform format conversions, and write metadata to local files.
Sanitization: There are no documented procedures for validating or sanitizing the data retrieved from external sources before it is used to generate reports or passed to command-line tools.

seo-images