baoyu-slide-deck
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [SAFE]: The skill performs its stated function using standard file operations and a multi-step user confirmation flow.
- [PROMPT_INJECTION]: The skill processes untrusted user markdown to generate image prompts, which is an indirect prompt injection surface.
- Ingestion points: Content is read from source files as described in SKILL.md.
- Boundary markers: The base-prompt.md uses markdown headers to separate style and content.
- Capability inventory: The skill can read/write files and execute scripts.
- Sanitization: No sanitization is performed on user content.
- [COMMAND_EXECUTION]: Local TypeScript scripts are used to merge images into PDF and PPTX files. The scripts scripts/merge-to-pdf.ts and scripts/merge-to-pptx.ts are executed via bun and perform only local file operations.
- [EXTERNAL_DOWNLOADS]: The skill uses pdf-lib and pptxgenjs libraries. These are standard, well-known Node.js packages used for document generation and are considered safe resources.
Audit Metadata