GuardrlyGuardrly

Documentation

Everything you need to install, configure, and get the most out of Guardrly.

📖

Overview

What is Guardrly?

A non-invasive AI Agent monitoring layer that intercepts, records, and alerts on every API call.

How MCP Interception Works

How Guardrly proxies outbound HTTP requests through the MCP protocol without modifying Agent behavior.

🚀

Getting Started

Install Guardrly MCP Server for Claude and Cursor

Install Guardrly in 3 minutes with one command, then monitor AI agent API calls from Claude Desktop, Cursor, or any MCP-compatible tool.

🔌

Platforms

Shopify API Monitoring for AI Agents

Monitor AI agents using the Shopify Admin API with rules for product deletes, webhooks, shop settings, rate limits, and audit logs.

Meta Ads API Monitoring for AI Agents

Monitor AI agents using the Meta Marketing API with risk rules for campaign deletes, budget changes, audiences, 403s, and 429s.

🛡️

Features

8 AI Agent Alert Rules for Risky API Calls

Configure 8 Guardrly alert rules for DELETE storms, 403s, 429s, write bursts, large payloads, and off-hours AI agent actions.

🔒

Security

PII Scrubbing for AI Agent API Logs

See how Guardrly redacts API keys, tokens, emails, phone numbers, and cards locally before AI agent logs leave your machine.