The Agent Skills Directory

[SAFE]: No malicious patterns were detected. The skill instructions focus on valid use cases for the Slidev framework, including project initialization, content generation, and export functionality using official tools.
[EXTERNAL_DOWNLOADS]: The skill instructs the agent to perform standard project setup using well-known tools like npm and npx to fetch the Slidev framework and its dependencies, such as playwright for PDF export, from official registries.
[PROMPT_INJECTION]: The skill identifies a potential attack surface for indirect prompt injection, where untrusted data provided by a user is processed to generate slides. This is handled through structured templates and instruction rules.
Ingestion points: User-provided topic descriptions and content for conversion (e.g., in references/prompt-templates.md).
Boundary markers: Utilizes labeled sections (e.g., 'Requirements', 'Rules') to delimit the agent's tasks.
Capability inventory: Employs standard file system writes and execution of the slidev CLI.
Sanitization: Relies on prompt-level logical constraints rather than explicit data sanitization.

ai-slidev