website-security-audit
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes command-line tools and system utilities to perform network reconnaissance and vulnerability scanning on target domains.
- Evidence: The instructions describe using curl.exe, dig, nslookup, nmap, and sqlmap to inspect infrastructure and test for vulnerabilities.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from external websites, which presents a risk of indirect prompt injection.
- Ingestion points: The skill fetches and processes HTML, JavaScript bundles, and HTTP response data from target websites as seen in SKILL.md.
- Boundary markers: There are no specific instructions or delimiters defined to prevent the agent from following instructions that might be hidden within the external content being audited.
- Capability inventory: The agent is equipped to perform network requests and execute various security-related CLI tools.
- Sanitization: No mechanisms for sanitizing or filtering external website content are described in the auditing pipeline.
Audit Metadata