The Agent Skills Directory

[SAFE]: The skill's primary function is to generate standalone HTML files for presentations. It follows best practices by keeping all dependencies inline and avoiding external build tools.
[COMMAND_EXECUTION]: The skill uses Python scripts (scripts/extract-pptx.py and scripts/export-pptx.py) to handle PowerPoint conversions. These scripts use the standard python-pptx library for file parsing and do not exhibit dangerous subprocess or shell execution patterns.
[EXTERNAL_DOWNLOADS]: The HTML template and example files fetch fonts from trusted services including Google Fonts (fonts.googleapis.com) and Fontshare (api.fontshare.com). These are well-known, safe resources for typography.
[DATA_EXFILTRATION]: There are no network operations that exfiltrate user data. The 'Inline Editing' feature explicitly uses localStorage for client-side persistence, which is a safe and standard web practice for this use case.
[PROMPT_INJECTION]: The instructions focus on layout, design, and technical implementation. There are no attempts to override agent safety guidelines or extract system prompts.
[CREDENTIALS_UNSAFE]: No hardcoded credentials or sensitive API keys were found in the provided files.
[OBFUSCATION]: All code, CSS, and instructions are provided in clear text. No hidden URLs, multi-layer encoding, or zero-width character tricks were detected.

frontend-slides