What it does
It also scans tool results for prompt injection and checks installed skills and plugins across several popular coding agents.
Agent tooling
jev-guard
Jev Guard checks coding-agent tool calls and returns an allow, ask, or deny decision based on risk and session context.
Bookmark: jev-guard
Keep this in your collection.
Leave a noteWhat would you try with this?
: jev-guard
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
How you can use it
Try this as a second opinion before your coding AI changes files or deletes data. Check how your assistant handles a request for approval; some only show a warning. A separate online AI reads the planned action and what the assistant has read. That can include private files. Keep your normal access limits, because a missed warning or a lost connection can let work continue.