compiler-development
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches supplemental curated lists, toolchain configurations, and tutorial resources from the author's own GitHub repository (gmh5225/awesome-llvm-security).
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill instructs the agent to fetch and process data from a remote markdown file on GitHub (SKILL.md).
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present when ingesting this external content.
- Capability inventory: The skill provides templates for LLVM IR generation and Just-In-Time (JIT) execution (SKILL.md).
- Sanitization: There is no evidence of sanitization or validation for the content fetched from the remote repository.
- [DYNAMIC_EXECUTION]: The skill provides technical implementation patterns for Just-In-Time (JIT) compilation and symbol execution using LLVM ORC (LLJIT), allowing for the execution of machine code generated at runtime (SKILL.md).
Audit Metadata