The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands including python -m markitdown, unzip, and grep to extract and inspect content from PowerPoint files during the editing and quality assurance processes.
[DYNAMIC_EXECUTION]: The skill generates JavaScript code snippets for the PptxGenJS library. This code is intended to be provided to the user for local execution to create presentation files.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and processes content from external PowerPoint files.
Ingestion points: Content is extracted from .pptx files using markitdown and unzip commands as described in the 'Quality Assurance' and 'Decision Tree' sections of SKILL.md.
Boundary markers: The instructions do not specify any delimiters or warnings to ignore potentially malicious instructions embedded within the extracted presentation content.
Capability inventory: The agent can execute shell commands (markitdown, unzip, grep) and generate executable JavaScript scripts.
Sanitization: There is no mention of sanitizing, escaping, or validating the text extracted from external files before it is processed or used by the agent.

pptx-slides