Agent Guardrails

Agent Guardrails is a tool designed for teams and developers who use AI agents to make code changes in software projects. It acts as a repo-level safeguard, automatically checking and enforcing boundaries for AI-generated code changes before they are merged. It verifies scope, validation, consistency, and risk, creating repeatable, trustworthy handoff processes and surfacing clear review outputs for human reviewers. It works alongside popular coding agents like Claude Code, Codex, Cursor, Gemini CLI, and OpenCode and provides MCP (Model Context Protocol) integration for advanced agent orchestration.