makepad-basics
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to read and incorporate content from external reference files (
./references/app-structure.mdand./references/event-handling.md). This behavior creates a surface for indirect prompt injection. - Ingestion points: local files
./references/app-structure.mdand./references/event-handling.mddefined in SKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore instructions inside those files are present.
- Capability inventory: The skill is restricted to code generation and Q&A; it has no subprocess or system-modifying capabilities.
- Sanitization: No sanitization of the reference file contents is specified.
- [EXTERNAL_DOWNLOADS]: The skill mentions and links to the official Makepad GitHub repository and the crates.io registry. These are well-known technology services used for legitimate software development and dependency management.
Audit Metadata