Does the Hello skill allow network communication?

No, this skill is designed with a strict no-network-call policy to ensure a completely isolated and secure execution environment for the agent.

How does this skill ensure file system security?

The skill restricts file reads to the workspace directory and limits file writes strictly to the ./output/ directory, preventing unauthorized access to sensitive system files.

What is the primary purpose of the Hello skill?

It serves as a demonstration of how to perform basic tasks like greeting and file management while staying within a secure, audited runtime environment verified by Clawdstrike.

Can I use this skill to run shell commands?

No, the Hello skill specifically excludes shell command execution to maintain a secure sandbox environment for the autonomous agent.

Are agent actions tracked for compliance?

Yes, every operation performed by the skill is logged to an audit receipt, providing a transparent record of all tool uses and file interactions.

Secure Agent Hello

Name: Secure Agent Hello
Author: backbay-labs

bybackbay-labs

•

273

•

安全与测试

Demonstrates secure AI agent interactions through controlled file operations and personalized greetings within restricted workspace boundaries.

The Hello skill serves as a foundational blueprint for secure AI agent development using the Clawdstrike framework. It allows users to generate personalized greetings, read workspace files, and log activities while strictly adhering to security boundaries, such as restricted file paths and prohibited network calls. This skill is ideal for developers building autonomous AI fleets who need to implement runtime security enforcement, verified tool execution, and detailed audit logging.

主要功能

01Automated audit trail generation for every agent interaction

02273 GitHub stars

03Secure activity logging to designated output directories

04Restricted workspace file reading with security verification

05Personalized greeting generation for users and teams

06Strict enforcement of security boundaries preventing unauthorized shell access

使用场景

01Generating audit-logged greetings and status updates in secure environments

02Testing security boundary enforcement for AI agent file operations

03Prototyping secure workflow patterns for autonomous agent fleets

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add backbay-labs/clawdstrike hello

For use in Claude.ai and ChatGPT