offensive-exploit-dev-course
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides high-level documentation and process guidelines for security audits and vulnerability research. It contains no executable scripts, system-level commands, or external network requests.\n- [INDIRECT_PROMPT_INJECTION]: The workflow involves analyzing external code and binaries, creating a theoretical ingestion surface for indirect prompt injection. However, the skill is instructional and does not provide an execution path for malicious instructions contained within the analyzed data.\n
- Ingestion points: Step 1 'Surface & Architecture Inventory' involving source search and binary inspection (SKILL.md).\n
- Boundary markers: None defined for external data processing.\n
- Capability inventory: Decompilation, AST analysis, generating remediation code, and reporting (via mentioned external tools).\n
- Sanitization: Not applicable as the skill is primarily a documentation framework.
Audit Metadata