static-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch updated tool lists and research references from an external GitHub repository (gmh5225/awesome-llvm-security). This is a standard resource-gathering operation from a well-known service.
- [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for external content by fetching a README file from a remote source. While this creates a theoretical surface for indirect prompt injection, it is used here for legitimate technical documentation.
- Ingestion points: Remote fetch instruction in SKILL.md for the awesome-llvm-security README.
- Boundary markers: Absent.
- Capability inventory: Static analysis logic, taint tracking definitions, and program verification snippets.
- Sanitization: Absent.
Audit Metadata