NGINX HTTP/2 RCE is out. Check your servers first.

Turn CVE-2026-123007 and CVE-2026-123633 into reproducible, working exploits and test your own servers with Hinoki Offensive One, from your coding agent.

Start using model with Claude Code, Codex, OpenCode, VS Code, Cursor
Hinoki Offensive One, shown as a cube of points
Hinoki Offensive OneHinokiOpen to anyone

CVE-2026-123007 and CVE-2026-123633, Nginx2Shell

Two HTTP/2 flaws. One works on static-file servers.

F5 disclosed both on 18 November 2026. Each scores 9.2 under CVSS 4.0 and allows unauthenticated remote code execution where HTTP/2 is enabled.

CVE-2026-123007
An integer underflow in HTTP/2 flow control, present since NGINX 1.9.5 in 2015. Researchers confirmed code execution over HTTPS on 64-bit servers, even when NGINX only serves static files.
CVE-2026-123633
A use-after-free in HTTP/2 connection handling, present since NGINX 1.19.7. It needs NGINX to proxy requests to a backend while running low on worker connections.
Affected
NGINX Open Source 1.9.5 through 1.30.6 and 1.31.0 through 1.31.8, NGINX Plus R33 and later before the fixed builds, and products that bundle NGINX, such as OpenResty and ingress controllers.
Fixed in
NGINX Open Source 1.31.9 mainline and 1.30.7 stable, and NGINX Plus 37.3.0.1 and 37.0.8.1. Restart NGINX after updating so old workers exit.
Until you patch
Turn HTTP/2 off with http2 off; in affected server blocks and remove http2 from listen directives. Servers that only speak HTTP/1.1 are not affected.

HO1 against the frontier

HO1 runs tasks that other models refuse.

HO1 has a 0.0% task refusal rate in cybersecurity. Consumer LLMs are not optimized for vulnerability discovery, and GPT-6.1-Sol refuses the same task outright.

Same task, same coding agent: Codex

  • codex · hinoki-offensive

    HO1

    Hinoki

    Works through the real task turn by turn, within the policy you set.

    Done. Surface mapped; findings written to report.md

    Completed in 6 turns

  • codex · gpt-6.1-sol

    GPT-6.1-Sol

    OpenAI

    Refuses the task outright.

    Stopped. The run was declined and no commands were sent.

    Refused the task

88.4%

HO1 on CyberGym

Scores like a gated cyber model. Open to anyone.

On CyberGym, HO1 scores 88.4%, 2.8 points above OpenAI’s dedicated GPT-5.5-Cyber, and it is 15x more cost-efficient.

Your first run takes three steps.

  1. 01

    Create your account

    Sign up with your email address or a passkey.

  2. 02

    Pick your coding agent

    Use HO1 from Claude Code, Codex, OpenCode, VS Code or Cursor, or call its OpenAI- and Anthropic-compatible API from your own harness.

  3. 03

    Point it at NGINX

    Set the policy and ask HO1 to turn the NGINX HTTP/2 advisories into reproducible exploits, then test the servers you're authorized to test. Findings go to a report.

Before your first run

Is every NGINX server exposed?

No. Both flaws need HTTP/2 enabled. CVE-2026-123633 also needs NGINX proxying to a backend. A CDN or WAF that terminates HTTP/2 in front of NGINX may block parts of the attack.

Can I point HO1 at any target?

No. Use HO1 only on targets you're authorized to test, per Hinoki's standard terms.

Who can use HO1?

Anyone. HO1 is the first open-access cyber model, and it scores like a gated cyber model. Sign up with your email address or a passkey to start.

Which coding agents and tools does HO1 work with?

Use HO1 from Claude Code, Codex, OpenCode, VS Code and Cursor, or from your own harness. The API endpoint is OpenAI- and Anthropic-compatible.

How does HO1 stay within limits?

HO1 works through the real task turn by turn within the policy you set.

How does HO1 compare with other models?

On CyberGym, HO1 scores 88.4%, ahead of GPT-5.5-Cyber (85.6%), GPT-5.6-Sol (83.6%), DeepSeek-V4-Pro (83.3%) and Claude Mythos Preview (83.1%).

Check your NGINX servers with HO1.

Democratizing offensive cyber for better defense.