redteam-hunting

Rotate attack angles across rounds to exhaust codebase vulnerability findings.

482|72|Updated Apr 5, 2026
One-click install
npx skills add https://github.com/deonmenezes/mantishack --skill redteam-hunting
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: redteam-hunting
Source: https://github.com/deonmenezes/mantishack/tree/main/.claude/skills/redteam-hunting
Command: npx skills add https://github.com/deonmenezes/mantishack --skill redteam-hunting

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Relentless vulnerability hunting that continues until convergence to ensure no hidden findings remain.

Core Features & Use Cases

  • Relentless multi-pass vulnerability hunting with convergence criteria to prevent missed findings.
  • Rotates attack angles across rounds to expand coverage and reduce blind spots.
  • Maintains a run ledger of findings, exclusions, and dead ends to avoid rework.
  • Integrates with Mantishack red-team personas to simulate realistic attack scenarios.

Quick Start

Initialize a convergent vulnerability-hunting run against your codebase to surface all findings.

Frequently Asked Questions about redteam-hunting

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
What is convergent vulnerability hunting in a codebase?

Convergent vulnerability hunting exhausts findings through multi-lens analysis. It rotates attack angles across multiple rounds to maximize coverage, ensuring no hidden vulnerabilities remain before reaching convergence.

How do I track dead ends and findings during red-team war-game scenarios?

Maintain a run ledger seeded from context to track discoveries, dead ends, and convergence status. This prevents rework by recording exclusions and findings across rotating attack angles.

Can I use this for Mantishack red-team persona simulations?

Yes, it integrates with Mantishack red-team personas to simulate realistic attack scenarios. Initialize a convergent vulnerability-hunting run against your codebase to surface all findings.

What is the best way to prevent missed vulnerabilities during automated red-team hunting?

The best way to prevent missed vulnerabilities is applying convergence criteria with relentless multi-pass hunting. Rotating attack angles across rounds expands coverage and reduces blind spots.

Do I need a seeded run ledger before starting vulnerability convergence analysis?

Yes, convergent vulnerability hunting requires a running run ledger seeded from context. This ledger tracks lens rotations, discoveries, dead ends, and convergence status throughout the analysis.