The Agent Skills Directory

[SAFE]: The skill functions as a visual renderer as described, with no suspicious or hidden behaviors detected.
[COMMAND_EXECUTION]: The skill reads a local HTML template and writes the final output to the user's Downloads directory, which is appropriate for its stated purpose.
[EXTERNAL_DOWNLOADS]: The skill includes the ability to fetch content from user-provided URLs to generate presentations.
[PROMPT_INJECTION]: The skill processes external data (files and URLs) which is a potential surface for indirect prompt injection; however, the risk is mitigated by explicit instructions to treat input text as 'truth' without modification and the inclusion of HTML escaping in the output template to prevent script execution.
[DATA_EXFILTRATION]: No unauthorized data transmission was detected; the skill's operations are confined to processing user-provided input and generating local output.

ljg-present