llvm-tooling
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches updated tool listings and resource references from the author's GitHub repository (gmh5225/awesome-llvm-security/refs/heads/main/README.md).
- [INDIRECT_PROMPT_INJECTION]: The skill establishes a data ingestion surface by directing the agent to incorporate external content from a remote README file into its context, which could potentially contain malicious instructions. * Ingestion points: Instructions in SKILL.md to fetch data from a remote URL. * Boundary markers: Not present; the agent is not instructed to ignore embedded commands in the fetched data. * Capability inventory: Templates for shell command execution (plugin compilation) and debugger script implementation. * Sanitization: Not present.
Audit Metadata