sanitizers
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFE
Full Analysis
- Prompt Injection (SAFE): No patterns detected that attempt to override agent instructions or bypass safety filters. The instructions are purely technical and educational.
- Data Exposure & Exfiltration (SAFE): No hardcoded credentials or sensitive file paths were found. Network operations are absent.
- Obfuscation (SAFE): The content is clear and uses no encoding techniques (like Base64 or zero-width characters) to hide intent.
- Unverifiable Dependencies & Remote Code Execution (SAFE): The skill does not download or execute external scripts or packages. It references standard compiler tools (gcc, clang) and official LLVM documentation.
- Privilege Escalation (SAFE): No commands for elevating privileges (e.g., sudo) or modifying system-level configurations are present.
- Persistence Mechanisms (SAFE): No attempts to modify shell profiles, cron jobs, or startup services.
- Metadata Poisoning (SAFE): The skill metadata correctly describes the tool's purpose without deceptive instructions.
- Indirect Prompt Injection (LOW): While the skill guides the agent in processing user-provided sanitizer reports, it does so through a manual diagnostic process rather than automated ingestion and execution, presenting a minimal risk surface.
- Dynamic Execution (SAFE): The skill contains no dynamic code generation, runtime compilation of untrusted code, or unsafe deserialization.
Audit Metadata