Security audit
Agent Code Guard
Security checks for vulnerabilities and agentic risk
Overview
This plugin transparently enforces a configurable guard that blocks selected agents from editing code or config files directly.
Install only if you want a startup policy that can block or gate tool calls for selected agents. Configure agentIds carefully, start with audit or requireApproval if you want to observe impact first, and do not treat this as a hard sandbox against hostile code execution.
SkillSpector was not run because this plugin release contains no bundled skills.
Static analysis
No suspicious patterns detected.