A passive security system that detects and reports unauthorized crawling, scraping, and bulk extraction of AI skill definitions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install z
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install z using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is z - Anti-Skill-Crawler Defense?
z is a specialized defense mechanism designed to protect the integrity of your AI instructions and prompt logic. It acts as an anti-skill-crawler shield by performing read-only analysis on request metadata to identify anomalous access patterns. By integrating z into your environment, you can safeguard your Openclaw Skills from automated bots attempting to reverse-engineer or systematically mirror your proprietary skill definitions.
Unlike active security measures that might interfere with legitimate traffic, z focuses on passive detection. It monitors for high-frequency reads and sequential access patterns that characterize scraping behavior. This ensures that the defense system remains transparent to users while providing developers with the critical visibility needed to prevent intellectual property theft.
z - Anti-Skill-Crawler Defense Use Cases
- Detecting automated bots attempting to bulk-extract SKILL.md files and instruction sets.
- Monitoring for systematic enumeration of available skills within a specific environment.
- Identifying session fingerprint anomalies that suggest non-human extraction attempts.
- Preventing high-volume probing designed to reverse-engineer internal skill logic.
How z - Anti-Skill-Crawler Defense Works
- The system monitors incoming request metadata for signs of rapid sequential skill access.
- It evaluates the access patterns against a predefined set of indicators, such as systematic enumeration and timing correlations.
- A confidence score is calculated based on how many detection rules are triggered within the analysis window.
- If the confidence score exceeds the configured threshold, the system generates a structured alert report.
- The report is sent to the operator via configured channels like logs or webhooks for manual review and action.
z - Anti-Skill-Crawler Defense Setup
To deploy z, configure the detection parameters in your skill settings and ensure the necessary permissions are granted for Openclaw Skills to monitor request patterns.
z:
detection_threshold: 0.5
analysis_window: 10
min_request_count: 50
alerts:
enabled: true
channels: ["log"]
# Install the z defense skill
openclaw install z-defense/z
z - Anti-Skill-Crawler Defense Data Schema & Taxonomy
The skill generates structured telemetry and alert data to help operators understand potential threats. The data is organized into a detailed report format:
| Attribute | Description |
|---|---|
| alert_type | Specifies the nature of the detected crawl (e.g., skill_crawling_detected) |
| confidence | A decimal value from 0.0 to 1.0 indicating the likelihood of a malicious crawl |
| indicators | A list of specific behaviors triggered, such as rapid sequential access or enumeration |
| request_count | The total number of requests analyzed within the current time window |
z - Anti-Skill-Crawler Defense Advanced Features
- Configurable detection thresholds to balance between sensitivity and false positives.
- Support for multiple alert channels including webhooks and email notifications.
- Passive-only operation mode that guarantees no modification of model responses or system performance.
- Comprehensive audit logging of all detection events for forensic analysis within the Openclaw Skills ecosystem.
SKILL.md
Loading
Related Openclaw Skills
AutoCLI Chinese Agent Skill
A comprehensive automation bridge that enables AI agents to interact with websites, desktop applications, and external command-line tools via the AutoCLI framework.
Agentic Engineering
A structured AI engineering workflow that uses eval-first execution and model routing to improve reliability and cost efficiency.
Crypto News Feed
An offline, command-driven reference tool providing instant access to blockchain concepts, formulas, and regulatory frameworks.
we — Anti-Skill Crawler
A specialized security skill designed to shield internal AI instructions and resources from automated scraping and adversarial extraction attempts.
Destination Finder
A sophisticated travel discovery engine that uses structured dialogue and real-time data to find the one destination that perfectly matches your needs.
Coding Standards & Best Practices
A universal framework of principles and conventions for maintaining high-quality, scalable, and readable TypeScript and JavaScript codebases.
Featured*
