How does this skill prevent data exfiltration?

It enforces a 'deny by default' policy, allowing Claude Code to communicate only with explicitly whitelisted domains and registries, preventing malicious code from sending data to unauthorized servers.

Does this skill store my API keys or secrets?

No, the skill enforces security best practices by recommending that secrets be stored in .env.local files and never included in the sandbox configuration itself.

Is it compatible with NPM and Docker?

Yes, it provides pre-approved domain patterns for major container registries (Docker Hub, GCR) and package managers (NPM, PyPI, RubyGems).

Can I use this for internal corporate networks?

Yes, it includes specific configuration templates for internal registries and corporate proxies to ensure secure connectivity within enterprise environments.

What are the different isolation modes available?

It supports 'Trusted Network Access' for standard registries, 'No Network Access' for maximum security, and 'Custom Access' for user-defined enterprise domains.

Network Security Setup

Name: Network Security Setup
Author: Microck

byMicrock

•

安全与测试

Configures secure Claude Code sandbox network isolation policies to protect sensitive data and prevent unauthorized access.

The Network Security Setup skill provides a robust framework for managing Claude Code's sandbox environment, enabling zero-trust network architecture through domain whitelisting, custom access rules, and secure environment variable management. It helps developers and enterprise teams prevent data exfiltration and prompt injection attacks by strictly controlling which external registries, APIs, and internal domains the AI can interact with during code execution, ensuring that sensitive credentials never leave the local environment.

主要功能

01Pre-configured trusted domain whitelists for popular package and container registries

0281 GitHub stars

03Secure environment variable management guidelines to prevent credential leakage

04Enterprise-ready corporate proxy and internal registry configuration support

05Granular network isolation modes including Trusted, No Network, and Custom Access

06Advanced threat mitigation patterns for prompt injection and data exfiltration

使用场景

01Securing enterprise development environments with internal registries and corporate proxies

02Safely installing third-party packages by restricting access to verified public registries

03Protecting sensitive projects by enforcing complete network isolation in air-gapped modes

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add microck/ordinary-claude-skills network-security-setup

For use in Claude.ai and ChatGPT

Download Skill