Is a HuggingFace token required?

Yes, you must configure an HF_TOKEN with write permissions in your .env file to manage repositories, push data, or access gated datasets.

What are E1 and E2 repositories in this context?

E1 repositories contain network-log datasets, while E2 repositories focus on configuration-verification data. Both are core components of the Security Verifiers reinforcement learning environment.

Can I verify my dataset schema before uploading?

Absolutely. The skill includes validation commands that check your data splits and features locally to ensure compatibility with HuggingFace Hub before any upload occurs.

How does the skill handle data privacy?

It implements a dual-repo strategy: public repositories host only non-sensitive metadata for discovery, while full datasets are pushed to private repositories with manual gated access enabled.

What happens during a canonical force push?

A canonical push with the force flag will delete and recreate the remote repository. This is intended for use when major schema changes are required that are incompatible with the existing Hub version.

Security Verifiers HF Manager

Name: Security Verifiers HF Manager
Author: intertwine

byintertwine

•

数据科学与机器学习

Manages HuggingFace datasets for security-focused reinforcement learning, streamlining data validation, metadata exports, and gated repository management.

The sv-hf skill provides a specialized interface for handling cybersecurity datasets within the Security Verifiers framework. It automates the complex lifecycle of E1 (network logs) and E2 (configuration verification) datasets by facilitating local builds, schema validation, and synchronization with the HuggingFace Hub. By separating public metadata from private canonical data through gated access, it allows researchers to share dataset previews without exposing sensitive training information, ensuring secure and reproducible AI security research.

主要功能

01Automated metadata extraction and push for public Dataset Viewer compatibility

02Support for forced canonical updates to synchronize schema changes

03Environment-aware dataset loading for local, hub, or synthetic data modes

042 GitHub stars

05Pre-push validation scripts for E1 and E2 dataset splits and schemas

06Gated repository configuration for private canonical dataset management

使用场景

01Generating public-facing metadata catalogs for cybersecurity benchmarks

02Configuring custom organizational repositories for Security Verifiers deployment

03Uploading sensitive network log datasets to private, gated HuggingFace repositories

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add intertwine/security-verifiers sv-hf

For use in Claude.ai and ChatGPT

主要功能

01Automated metadata extraction and push for public Dataset Viewer compatibility

02Support for forced canonical updates to synchronize schema changes

03Environment-aware dataset loading for local, hub, or synthetic data modes

042 GitHub stars

05Pre-push validation scripts for E1 and E2 dataset splits and schemas

06Gated repository configuration for private canonical dataset management

使用场景

01Generating public-facing metadata catalogs for cybersecurity benchmarks

02Configuring custom organizational repositories for Security Verifiers deployment

03Uploading sensitive network log datasets to private, gated HuggingFace repositories