baoyu-slide-deck
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands to manage configuration files and execute merging scripts. Specifically, it uses
testto check for the existence ofEXTEND.mdand project directories, andnpx -y bunto execute the TypeScript scripts located in thescripts/directory. - [EXTERNAL_DOWNLOADS]: The use of
npx -y bunto run the merging scripts may trigger the download of thebunpackage and its dependencies (likepdf-libandpptxgenjs) from the NPM registry if they are not already present in the environment. - [INDIRECT_PROMPT_INJECTION]: The skill ingests arbitrary user content to generate slide outlines and image generation prompts, which presents a surface for indirect prompt injection.
- Ingestion points: User-provided content files (
source.md) or text pasted directly by the user. - Boundary markers: The instructions do not define clear delimiters or specific instructions to ignore embedded commands within the user-provided content during processing.
- Capability inventory: The skill has the ability to read and write to the local filesystem and call external image generation skills (e.g., "nano banana pro").
- Sanitization: No explicit sanitization, validation, or escaping of user-supplied text is performed before it is interpolated into analysis results or image prompts.
Audit Metadata