detecting-directory-listing

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves a clear security purpose (detecting directory listings) and includes built-in safeguards such as authorization checks for remote targets and timeouts for network operations.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests HTML and XML content from external URLs to check for server fingerprints. While this provides a theoretical surface for indirect prompt injection if a malicious server returns crafted content, the script only performs regex matching on a small sample of the body and does not execute the content, minimizing risk. Findings are documented with evidence containing only metadata (content type, sample length) rather than raw content.
  • [COMMAND_EXECUTION]: The skill utilizes legitimate shell commands (python3, curl) as specified in the allowed-tools configuration to perform its auditing tasks. It includes restrictions against dangerous operations like rm or editing sensitive system files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 02:08 PM
Security Audit — agent-trust-hub — detecting-directory-listing