The Agent Skills Directory

[PROMPT_INJECTION]: The skill documentation describes a workflow for processing external artifacts, which creates a surface for Indirect Prompt Injection.
Ingestion points: The skill is designed to process 'any sort of artifact' (SKILL.md).
Boundary markers: There are no instructions to the agent to treat the content of these artifacts as untrusted or to ignore instructions embedded within them.
Capability inventory: The technical details mention using python-pptx for file generation and modification (SKILL.md).
Sanitization: The skill lacks instructions for sanitizing or validating input from the processed artifacts.
[NO_CODE]: The skill contains only documentation (SKILL.md) and license information (LICENSE.txt), with no executable scripts or code files included in the package.

brand-guidelines