Security audit
TypeSafe AI
Security checks for vulnerabilities and agentic risk
Overview
This plugin coherently adds a typed decision provider and discloses that selected evidence is sent either to TypeSafe AI or to a configured local server.
Before installing, be comfortable that evidence supplied to decision_evaluate may be sent to TypeSafe AI and may incur usage charges. Use a protected SecretRef for hosted Jev, or configure only a loopback baseUrl for local Kev if you do not want hosted credential use.
SkillSpector was not run because this plugin release contains no bundled skills.
Static analysis
No suspicious patterns detected.