A minimal Claude Code clone with REPL interface and bash tool calling
Defense-in-depth protection for Pi coding agent. Blocks dangerous commands and protects sensitive files — rename of pi-damage-control.
Secure shell server for the Model Context Protocol. Integrates with Claude Desktop.
Interactive guardrails for dangerous bash commands and protected file edits in Pi.
System command safety blocker
GnamiAI command-line client. Talk to your hosted agent from your terminal; let it read files, run shell commands, and edit code on your machine.
Pi extension to handle permissions using a layered approach
Policy-controlled sandboxes for AI coding agents on HPC (Claude-first, Apptainer, SLURM).
Write a file to automatically create a directory function based on node - based FS
Pi coding agent extensions — security hardening, agent teams and more
Layered permission control extension for pi-coding-agent
Conditional execution gates for Pi. Hold, approve, or block tool calls based on rules. Safety layer for dangerous operations.
Claude Code PreToolUse hook for Interven. Scans every Claude Code tool call (Read/Write/Edit/Bash/Glob/Grep/MCP) through Interven before execution. Blocks .env reads, denies destructive shell, redacts secrets — no agent code change.
Security scanner for Claude Code settings files
Secure remote terminal bot with Claude Code integration
High level bindings over a WASM port of FatFS by ChaN
n8n nodes for PostgreSQL-backed event channel communication between workflows
Claude Code-style permission modes for pi. Controls approval for file writes, edits, and bash commands with four modes: default, acceptEdits, fullAuto, and bypassPermissions.
Production-grade Claude Code hooks. Fail closed by default, audit every run, never silently break.
Safety guardrails for Claude Code — blocks dangerous operations automatically
Structured advisor and command guardrail for Pi.
A production-ready, highly secure SSH MCP server featuring stateless connections, two-step operation confirmation, and comprehensive DevOps tool integration.
Security layer for LLM-driven code agents - blocks dangerous operations before they execute
🚀 Advanced terminal emulator built in Rust with Python integration - A comprehensive command-line experience