The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by ingesting untrusted data from external sources like codebases, articles, and notes to generate slide content.
Ingestion points: External text and code provided by the user as source material for generating slides (SKILL.md).
Boundary markers: The instructions do not specify the use of clear delimiters or boundary markers to isolate external content from the agent's instructions.
Capability inventory: The agent is empowered to write HTML files to the local file system and use platform tools to publish these files (SKILL.md).
Sanitization: The risk is mitigated by explicit instructions to distill content rather than quote it verbatim and by the mandatory use of textContent and createElement to prevent XSS vulnerabilities in the output.
[SAFE]: The skill enforces robust security requirements for its generated artifacts, including a strict prohibition on using innerHTML with variable data and a requirement to prefix localStorage keys to prevent cross-artifact data access.
[EXTERNAL_DOWNLOADS]: The skill allows generated HTML to fetch font assets from Google Fonts, which is a well-known and trusted external service.
[SAFE]: The mechanism for publishing decks uses a local, single-session loopback handshake for communication, ensuring that artifact submission remains within the controlled environment of the agent's platform.

html-slideshow-deck