llvm-obfuscation
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 a resource file (README.md) from the author's GitHub repository (
gmh5225/awesome-llvm-security). This is a reference to a vendor resource on a well-known service for the purpose of retrieving curated tool lists. - [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for external data by fetching a remote markdown file. This introduces a theoretical attack surface where instructions embedded in the external content could influence the agent.
- Ingestion points: The
README.mdfile located athttps://raw.githubusercontent.com/gmh5225/awesome-llvm-security/refs/heads/main/README.md. - Boundary markers: No specific delimiters or warnings to ignore instructions within the fetched data are provided.
- Capability inventory: The skill primarily involves information retrieval and providing technical guidance; it does not contain subprocess execution or file-write capabilities.
- Sanitization: No explicit sanitization or validation of the fetched content is described.
Audit Metadata