Enforces strict repository boundaries and monorepo safety protocols for AI-driven development.
Public Guardrails establishes a secure execution environment by restricting AI agent operations to the local repository tree, preventing accidental modifications to system files or external directories. Optimized specifically for pnpm workspace monorepos, it ensures that changes remain confined to individual package scopes, manages command execution contexts, and protects critical root configuration files from unauthorized edits. It is an essential skill for maintaining repository integrity and safety during automated coding sessions.
Key Features
01187 GitHub stars
02Strict repository boundary enforcement to prevent external file access
03Monorepo-aware isolation for pnpm workspace projects
04Context-aware command execution within specific package directories
05Automatic precedence handling for package-level skill overrides
06Protected root configuration and lockfile management policies
Use Cases
01Restricting development tasks to a specific package while preventing global config drifts
02Initializing a new agent session in a complex monorepo to ensure workspace safety
03Securing AI workflows by blocking access to system files and external git checkouts
