mirage-ogpp

Community

Single-tool gateway for safe agent actions.

Authorponyyuhan
Version1.0.0
Installs0

System Documentation

What problem does it solve?

The MIRAGE-OG++ gateway centralizes control of untrusted agent runtimes by exposing a single, vetted tool for executing high-level intents, preventing direct side-effect access and opaque data exposure.

Core Features & Use Cases

  • Enforces use of the mirage_act tool as the sole side-effect interface for agent runtimes.
  • Routes high-level intents (e.g., ReadFile, ReadSecret, SendMessage) through a trusted gateway with governance and auditability.
  • Provides strong guardrails: sealed-handle model, 2-server PIR-based policy checks, dual-authorization executor enforcement, and MCP-style exposure.

Quick Start

Invoke the mirage_act tool to issue a high-level intent via the gateway and observe the resulting JSON.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: mirage-ogpp
Download link: https://github.com/ponyyuhan/privacy_agent/archive/main.zip#mirage-ogpp

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.