context-firewall

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The analyzed files contain no executable code, scripts, or binary assets. The skill is entirely instructional, providing a design pattern for managing AI context windows during complex operations.
  • [SAFE]: The skill's instructions focus on architectural best practices such as task decomposition, state persistence, and context isolation. No malicious patterns, prompt injections, or unauthorized data access attempts were identified during the analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 05:12 PM
Security Audit — agent-trust-hub — context-firewall