The Agent Skills Directory

[COMMAND_EXECUTION]: The skill operates by generating Python code using the python-pptx library and executing it to produce .pptx files. This dynamic execution is the primary mechanism for the skill's functionality.
Evidence: The documentation states, "I'll generate python-pptx code and execute it" and provides numerous examples of code that saves files to disk (e.g., prs.save('pitch_deck.pptx')).
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted user content and interpolates it into the logic used for slide generation without explicit boundary markers or sanitization.
Ingestion points: User-provided presentation descriptions, outlines, data, and images specified in the "How to Use" and "Example prompts" sections of SKILL.md.
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are used when processing external data.
Capability inventory: The skill can write files to the local file system using the python-pptx library.
Sanitization: There is no evidence of validation or filtering for the content provided by the user before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill references and requires the installation of the python-pptx library from standard package registries.
Evidence: Installation instructions provided in SKILL.md specify pip install python-pptx and link to the official documentation and GitHub repository.

pptx-manipulation