Skip to content

GeminiZ

GeminiZ is an open-source privacy governance layer for AI agents, built as a fork of Gemini CLI.

Code Video Cerebral Valley

Core trick

PIMS.md declares which PII each tool may access, undeclared fields are blocked, local PII detection runs on-device, and every access is logged.

Tags

securitydev tools

Description

GeminiZ is an open-source privacy governance layer for AI agents, built as a fork of Gemini CLI. It uses a declarative file called PIMS.md (like .gitignore but for your personal data) to control what PII each tool and skill can access. PII detection runs locally using open-source models (AI4Privacy, Gemma) so your data never leaves your machine. Tools must declare what personal information they need. No declaration, no data. Every access is logged with a full audit trail, giving developers compliance with GDPR and the EU AI Act out of the box. With 40,000 exposed OpenClaw instances and 400 malicious plugins in the wild, nobody had built the privacy layer for AI agents. GeminiZ is that layer.
MikeDoes/geminiz★ 10 forksTypeScriptApache-2.0Created Mar 14, 2026 · last push Jun 15, 2026

README excerpt

# GeminiZ

![GeminiZ Logo](/docs/assets/GeminiZ.png)

> **🛡️ Your personal data, your rules, auditable by default.**

GeminiZ is a fork of [Gemini CLI](https://github.com/google-gemini/gemini-cli)
that adds a **privacy governance layer** for AI agents. It uses a declarative
file called `PIMS.md` — think `.gitignore` but for your personal data — to
control what PII each tool can access, with full audit trails.

![GeminiZ in action](/docs/assets/Global%20Screenshot.png)

## 🔒 The Problem

Everyone is using AI CLI tools — Claude Code, Gemini CLI, Cursor — sending
personal data to every tool, every skill, every MCP server. Most of these are
unverified. **40,000 exposed OpenClaw instances, 400 malicious plugins, 10,000
compromised developers** from supply chain attacks. Your PII is one bad skill
away from being exfiltrated.

Nobody has built the privacy layer for AI agents. **Until now.**

## 🛡️ How GeminiZ Works

1. **📋 PIMS.md** — A declarative policy file that defines what personal data
   exists, sensitivity levels, and which tools can access what
2. **🤖 Local PII Detection** — Uses open-source models like
   [AI4Privacy](https://huggingface.co/ai4privacy/llama-ai4privacy-multilingual-categorical-anonymiser-openpii)
   or Gemma, running locally — your data never leaves your machine
3. **✅🚫 Access Control** — Tools must declare what PII they need. No
   declaration, no data
4. **📝 Audit Trail** — Every access is logged. Not just to protect you — but to
   prove it to your clien

Team