AgentLeash – Monitor and control AI agent access in your codebase

Summary: AgentLeash provides real-time monitoring and control over AI agents accessing your codebase by logging file operations and enforcing access rules defined in YAML. It supports agents like Claude Code, Cursor, and Copilot, enabling visibility and alerts on unauthorized file access.

What it does

AgentLeash tracks AI agent file reads and writes in real time, allowing users to define access boundaries via YAML path rules. It blocks restricted file operations and sends notifications on violations.

Who it's for

Developers and teams using AI coding agents who need to monitor and restrict agent access to sensitive files in their projects.

Why it matters

It addresses the lack of visibility and control over AI agents reading entire codebases, protecting sensitive files from unauthorized access.