Security audit
Inference AIops
Security checks for vulnerabilities and agentic risk
Overview
This cluster-management plugin needs review because its confirmation promises overstate the protections applied to disruptive operations.
Start with access restricted to read endpoints. Require a preview and explicit approval before cluster changes; do not assume MCP writes provide double confirmation or that every change has an undo. SkillSpector skipped assessment on a no-bundled-skills premise, but the plugin manifest and inspected files contain a skill.
SkillSpector was not run because this plugin release contains no bundled skills.
Static analysis
No suspicious patterns detected.