The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides and instructs the agent to use several Python scripts (e.g., scripts/emit_entrance.py, scripts/emit_compound.py, scripts/validate.py) to generate and check PowerPoint animation XML fragments. These scripts are local to the skill and perform standard data processing and output tasks.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data through its scripts, which constitutes a potential ingestion surface for indirect prompt injection.
Ingestion points: The validate.py script accepts XML input via standard input or local files, and the emit_compound.py script accepts JSON data via command-line arguments or local files.
Boundary markers: The skill does not provide specific instructions or markers to the agent to distinguish between animation data and potential embedded instructions.
Capability inventory: The scripts are restricted to data parsing and printing generated XML to the standard output. They do not have network access or the ability to write to the file system.
Sanitization: Inputs are parsed using the standard json and lxml libraries.

pptx-animation