AI · Agents · Blockchain · Crypto

Submit a project Sign in

Menu

Directory category

Agent Security & Sandboxing

Compare isolation, permissions and sandboxing tools for agent execution. Compare official resources, requirements and limitations.

38 projects

Directory results

Agent Security & Sandboxing projects

A Agent Opfor Agent Opfor is an open-source developer tool that emulates adversarial attacks against AI agents, LLM applications and MCP servers, then uses an LLM to judge target responses.

D Destructive Command Guard Destructive Command Guard (dcg) is a Rust command-line tool that hooks into AI coding agents to inspect and block destructive Git and shell commands before execution.

P Prompt Hardener Prompt Hardener is an open-source Python CLI that analyzes prompt-injection risks in LLM agents and applications from an agent_spec.yaml description, recommends mitigations, supports adversarial attack simulation and exports security results as Markdown, HTML or JSON.

C Clawk Clawk is a command-line tool that runs coding agents and shells in disposable Linux virtual machines, with a project directory mounted from the host.

A Adrian Adrian is an open-source runtime security monitoring and control tool for AI agents.

V Vestige Vestige is a local runtime security project for AI agents.

C Cc Safety Net Cc Safety Net is a TypeScript command-line guard for AI coding agents that checks tool calls before execution and blocks destructive Git or filesystem commands and common attempts to access sensitive files.

E Emilia Protocol Emilia Protocol is authority-control infrastructure for autonomous work.