msw-packages
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches package descriptions, file trees, and source code from the author's official GitHub repository at
github.com/MSW-Git/MSWPackages. - [COMMAND_EXECUTION]: Instructs the agent to use command-line tools like
grepto inspect the local workspace for UUID collisions during the integration process. - [DYNAMIC_EXECUTION]: Integrates downloaded script files (
.mlua) and UI definitions (.ui) into the project workspace, which are subsequently loaded and executed by the game environment. - [INDIRECT_PROMPT_INJECTION]: Processes external content from GitHub which may contain embedded instructions.
- Ingestion points: External
README.mdand source files (.mlua,.ui) from theMSW-Git/MSWPackagesrepository. - Boundary markers: No specific delimiters or "ignore" instructions are used when processing the fetched remote content.
- Capability inventory: File system write access to
RootDesk/, network access viaWebFetch, and local workspace inspection viagrep. - Sanitization: No sanitization of the fetched text or code for potential prompt injection patterns is mentioned in the workflow.
Audit Metadata