ax-security
OfficialSecure your AI agent with robust controls and isolation.
Authorproject-ax
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This Skill ensures the security of your AI agent by implementing various controls and isolation measures to protect against security threats.
Core Features & Use Cases
- Sandbox Isolation: Sandboxes agents to prevent unauthorized actions and limit network access.
- Taint Budget: Tracks and manages the trustworthiness of content within sessions.
- Canary Tokens: Detects unauthorized access attempts by scanning for canary tokens in responses.
- Safe Path: Prevents path traversal attacks by sanitizing file paths.
- Provider Map Security: Ensures that only trusted providers are loaded and executed.
- Plugin Integrity: Verifies the integrity of plugins before loading them.
- Skill Screening: Analyzes skills for potential security risks before approval.
- Subagent Delegation Security: Limits the depth and concurrency of subagent delegations.
- HTTP Forward Proxy: Provides controlled outbound HTTP access with security controls.
- Install Validation: Validates the installation of sensitive actions.
- Common Tasks: Provides guidance on adding new sensitive actions and scanner patterns.
Quick Start
Use the ax-security skill to apply security measures to your AI agent.
Dependency Matrix
Required Modules
None requiredComponents
scriptsreferences
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: ax-security Download link: https://github.com/project-ax/ax-legacy/archive/main.zip#ax-security Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 620,000+ vetted skills library on demand.