Halberd — security grade SAFE, quality 68/100

Security audit verdict: SAFE · quality 68/100

Flagged: sudo usage. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →

by vectra-ai-research · Agent Tool · ★ 342

Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h

🔒 Is Halberd safe to install? View the security audit →

About Halberd

Halberd: Multi-Cloud Agentic Attack Tool Validate your multi-cloud defenses through realistic attack emulation 🚀 Get Started • 📖 About • ✨ Features • 📚 Guide • 📋 Docs • 🤝 Contributing 🚀 Quick Start Get Halberd running in under 60 seconds. Choose the installation method that works best for you. Option 1: Docker (Recommended) The fastest way to get started. One command and you're ready: Then open your browser to a

awsazureazureadblueteam-toolscloud-securityentra-idgcpm365microsoftmitre-attack

Quick Facts

Stars342
Forks35
LanguagePython
CategoryAgent Tool
LicenseGPL-3.0
Quality Score67.6357723180051/100
Open Issues7
Last Updated2026-04-08
Created2024-03-06
Platformsaws, python
Est. Tokens~220k

Compatible Skills

These tools work well together with Halberd for enhanced workflows:

  • mcp-security-hub — semantic(0.32)+complementary+rare_topics+same_lang+similar_pop+shared_platform (61%)
  • EVA — semantic(0.27)+complementary+rare_topics+same_lang+similar_pop+shared_platform (59%)
  • agentic-radar — semantic(0.26)+complementary+rare_topics+same_lang+similar_pop+shared_platform (59%)
  • communitytools — semantic(0.24)+complementary+rare_topics+same_lang+similar_pop+shared_platform (58%)
  • medusa — semantic(0.23)+complementary+rare_topics+same_lang+similar_pop+shared_platform (58%)

Halberd alternative? Top 6 similar tools

Looking for a Halberd alternative? If you're comparing Halberd with other agent tool tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.

  • Ivy-Framework by Ivy-Interactive · ⭐ 434

    The ultimate framework for building internal tools with LLM code generation by unifying front-end and back-end

  • terrashark by LukasNiessen · ⭐ 323

    Terraform Skill for Claude Code and Codex. LLMs hallucinate a lot with Terraform - TerraShark fixes this. It e

  • RedteamAgent by NeoTheCapt · ⭐ 112

    An AI red-team agent for authorized labs and web app pentesting workflows. Turns Claude Code / OpenCode / Code

  • LitterBox by BlackSnufkin · ⭐ 1.5k

    A self-hosted sandbox for red teams to test payloads against modern detection before deployment. MCP integrati

  • data-api-builder by Azure · ⭐ 1.5k

    Data API builder provides modern REST, GraphQL endpoints and MCP tools to your Azure Databases and on-prem sto

  • fuzzforge_ai by FuzzingLabs · ⭐ 769

    AI-powered workflow automation and AI Agents platform for AppSec, Fuzzing & Offensive Security. Automate vulne

More Agent Tool Tools

Explore other popular agent tool tools:

View all Agent Tool tools →

Popular Python Agent Tools

Frequently Asked Questions

What is Halberd?

Halberd is Halberd : Multi-Cloud Agentic Attack Tool. It is categorized as a Agent Tool with 342 GitHub stars.

What programming language is Halberd written in?

Halberd is primarily written in Python. It covers topics such as aws, azure, azuread.

How do I install or use Halberd?

You can find installation instructions and usage details in the Halberd GitHub repository at github.com/vectra-ai-research/Halberd. The project has 342 stars and 35 forks, indicating an active community.

What license does Halberd use?

Halberd is released under the GPL-3.0 license, making it free to use and modify according to the license terms.

What are the best alternatives to Halberd?

The top alternatives to Halberd on Agent Skills Hub include Ivy-Framework, terrashark, RedteamAgent. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.

How this security grade is produced

Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.

The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.

Sources & who's responsible:

View on GitHub → Browse Agent Tool tools