offensive-windows-privesc
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external untrusted data such as source code, binaries, and system configurations to identify privilege escalation vulnerabilities.\n
- Ingestion points: The workflow involves inventorying entrypoints, trust boundaries, and configuration files in the target environment.\n
- Boundary markers: The skill does not define delimiters or specific instructions to ignore malicious content embedded within audited files.\n
- Capability inventory: The agent is directed to use static and binary analysis tools on potentially attacker-controlled input.\n
- Sanitization: No sanitization or validation logic is specified for data retrieved during the discovery phase.\n- [NO_CODE]: The skill consists entirely of documentation and metadata; no scripts or executable files are provided in the package.
Audit Metadata