The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted image data from users which could potentially influence agent behavior if the agent processes the contents of these files as instructions. Ingestion points: input_path in scripts/pixel_art.py and base_image in scripts/pixel_art_video.py. Boundary markers: No specific delimiters or warnings for the agent are implemented when handling these external files. Capability inventory: The skill can write files via Pillow and execute external commands via ffmpeg. Sanitization: The skill does not perform explicit sanitization or metadata filtering on the input images.
[COMMAND_EXECUTION]: The skill executes external shell commands to facilitate video encoding. Evidence: scripts/pixel_art_video.py calls subprocess.run() to invoke ffmpeg for MP4 and GIF creation. Context: The execution is handled via a list of arguments without shell=True, which is a safer implementation. This functionality is essential for the skill's primary purpose of creating animated pixel art.

pixel-art