Security audit
TypeSafe AI (Jev System One)
Security checks for vulnerabilities and agentic risk
Overview
This plugin appears to do what it claims: add TypeSafe AI decision checks for triage, tool approval, compaction, and routing, with disclosed external API use and in-memory caching.
Before installing, users should understand that enabling this plugin with a TypeSafe API key allows selected chat content, tool-call parameters, and compaction snippets to be sent to TypeSafe AI for classification. Review the feature flags if you do not want default triage, tool guardrails, or compaction curation active.
SkillSpector was not run because this plugin release contains no bundled skills.
Static analysis
No suspicious patterns detected.