# Uniplex

> The trust layer for AI agents

Uniplex is an open protocol that provides verifiable identity and scoped authorization for AI agents.

## What it does

- Passports: Signed credentials that prove what an agent is and what it's allowed to do
- Gates: Local verification at tool boundaries (no network round-trip)
- Attestations: Signed proof of authorization decisions for audit trails

## Install

pip install uniplex

## Quick example

from uniplex import Agent

agent = Agent.create("my-agent", permissions="mcp:*")
decision = agent.authorize("mcp:search")

## Links

- Documentation: https://uniplex.io
- GitHub: https://github.com/uniplexprotocol/uniplex
- PyPI: https://pypi.org/project/uniplex/
- Specification: https://github.com/uniplexprotocol/uniplex/blob/main/SPECIFICATION.md

## Key concepts

- Works with MCP (Model Context Protocol)
- L1/L2/L3 trust profiles
- Ed25519 signatures
- No central server required for verification
