Jev in production › Moderation and content filtering

Tomarigi

Watches Claude Code and Codex sessions and sends each message to Jev to flag abusive language toward the AI with a live anger mark.

Open on GitHub ↗

The source publishes no measured number.
Use
Moderation and content filtering
Industry
Developer tools
Form
Open-source tool
Stage
In production
Plugs into
Claude Code, Codex
Listed
2026-10-09
Found via
github
Repository
mohhh-ok/tomarigi-desktop
Stars
17
Forks
2
Last push
2026-10-09
Language
TypeScript
License
MIT
Tomarigi screenshot
docs/images/gallery-1-on-top.webp in the mohhh-ok/tomarigi-desktop README, MIT; shown from GitHub.

The README opens with

A macOS app that watches your AI coding agent sessions (Claude Code / Codex) as birds in an always-on-top window. Each session gets its own bird in the Garden, so you can tell at a glance which one is working, which one is waiting for your reply, and which one is done. Click a bird to jump to the Ghostty pane where that session is running.

[!IMPORTANT] The anger mark: a heads-up when you're harsh to the AI. With a TypeSafe key, every message you send is checked by Jev for abuse toward the AI, and the session's bird gets an anger mark when it looks abusive. Anthropic's 2026 Usage Policy update (effective November 12, 2026) prohibits sustained and needless abusive or cruel behavior toward Claude, and Claude can end conversations with persistently...

Badge

For the project's own README, linking back here:

Listed in Jev in production

Also used for moderation and content filtering