Skip to content
KitploitKITPLOIT
ToolsExploitsBlog
Log in
Submit
ToolsExploitsBlog
Submit

Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!

Kitploit is a directory of hacking, cybersecurity, and pentesting tools. Discover the latest project updates to find vulnerabilities, analyze systems, automate testing, and strengthen your security.

FeedsContactPrivacy© 2026 Kitploit

Tool Directory

Categories

View all categories
Loading categories
ai-agent-security-research — Open security research on AI coding agent infrastructure. Agent-executable remediation manifests for CVE-2026-22812 and CVE-2026-22813. | Kitploit
Tools/GitHubGitHub/hodgeluke/ai-agent-security-research
Cloud Infrastructure SecurityVulnerability AnalysisWeb SecurityAI Security
GitHubhodgeluke/ai-agent-security-research

ai-agent-security-research

Open security research on AI coding agent infrastructure. Agent-executable remediation manifests for CVE-2026-22812 and CVE-2026-22813.

View Repository
Share
1487 months agoNot yet reviewed

AI Agent Security Research

Open security research on AI coding agent infrastructure.

Published by Hodge Luke Digital Intelligence Agency


What This Is

This repository contains security research, vulnerability analysis, and remediation guidance for AI coding agent platforms. Every advisory includes an agent-executable remediation manifest — a structured document your AI coding agent can read and execute directly.

The Concept: Agent-Executable Security Advisories

Most security advisories tell you what's wrong and what to do. You still have to do it yourself.

Our advisories ship with a remediation manifest — a structured set of instructions with STOP_IF conditions, ASK_USER directives for credentials, and REPORT checkpoints. Copy the manifest. Paste it into your AI coding agent's chat. Say: "Do this."

The agent reads the manifest, executes each step in order, stops when a condition isn't met, and asks you for credentials only when it needs them. It does not guess. It does not skip.

This is how security advisories should work in 2026.


Current Advisories

CVE-2026-22812 + CVE-2026-22813: Unauthenticated RCE in AI Coding Agents

  • CVE-2026-22812 — CVSS 8.8 — Unauthenticated remote code execution via HTTP server
  • CVE-2026-22813 — CVSS 9.6 — XSS-to-RCE via unsanitized LLM output
  • 220,000+ instances exposed on the public internet
  • 15,200 confirmed vulnerable to direct RCE
  • Affects: OpenCode serve mode, OpenClaw instances, Mac Mini local deployments

Quick start: Open advisories/CVE-2026-22812-22813/remediation.md, copy the manifest, paste into your AI agent.


Repository Structure

advisories/
  CVE-2026-22812-22813/
    ADVISORY.md          # Technical summary
    remediation.md       # Agent-executable remediation manifest
    exposure-data.md     # Sourced exposure statistics
    SOURCES.md           # Full citation list
hardening/
  linux-vps.md           # Layer 4 hardening for Ubuntu/Debian
  macos.md               # Layer 4 hardening for macOS

Full Analysis

The complete vulnerability analysis, reproduction methodology, and 5-layer defense-in-depth architecture is published as a blog post:

Your AI Coding Agent Has Root Access to Your Machine. Does Anything Else?


About FORGE

FORGE implements the 5-layer defense architecture described in this research as a one-command deployment. $47 once. Your VPS. Your code. No recurring fees.

  • Product
  • Full Comparison (8 platforms)
  • Security Architecture

Contributing

See CONTRIBUTING.md for how to submit findings.

License

MIT — the research is free. Use it to protect your infrastructure.


Hodge Luke Digital Intelligence Agency forge.useacceda.com

Download Tool

Most Popular

View all →

Discover the most used tools by our community.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 The ultimate OSINT toolkit: investigation graph, confidence scoring, alias families, map, timeline and reports. 49 modules, 400+ sites, no API keys required.

    755
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    GhostLock One-Tap Execution App (CVE-2026-43499)

    510
  • 3

    ARTEX

    baarai/artex

    LLM multi-agent autonomous penetration testing platform with planner/worker agents, asset and exploration graphs, MITM traffic recording, and vulnerability retesting.

    415
  • 4

    redforge

    cybersecurity91/redforge

    Browser-side red team toolkit with 102 offline tools for recon, payload generation, credential attacks, crypto, encoding and engagement reporting.

    343
  • 5

    antifox

    antifox-ai/antifox

    AI multi-agent autonomous penetration testing platform that runs recon, attack-surface discovery, exploitation validation and evidence-backed report delivery.

    259
  • 6

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next.

    258

Explore all tools

Browse our collection of tools

View all tools →