Skip to content
KitploitKITPLOIT
StrumentiExploitsBlog
Log in
Invia
StrumentiExploitsBlog
Invia

Strumenti di Hacking, PenTest e Cybersecurity per il tuo Arsenale di Sicurezza!

Kitploit è una directory di strumenti di hacking, cybersecurity e pentesting. Scopri gli ultimi aggiornamenti dei progetti per trovare vulnerabilità, analizzare sistemi, automatizzare i test e rafforzare la tua sicurezza.

FeedContattoPrivacy© 2026 Kitploit

Directory degli strumenti

Categorie

Vedi tutte le categorie
Loading categories
mlflow-scorer-scanner — Detection-only triage scanner for MLflow scorer RCE CVE-2026-116692: version check via the /version endpoint. No payloads, no scorer invocation. | Kitploit
Strumenti/GitHubGitHub/foxcornlab/mlflow-scorer-scanner
Defensive ToolsVulnerability ScannersWeb Vulnerability ScannersVulnerability AnalysisPenetration Testing
GitHubfoxcornlab/mlflow-scorer-scanner

mlflow-scorer-scanner

Detection-only triage scanner for MLflow scorer RCE CVE-2026-116692: version check via the /version endpoint. No payloads, no scorer invocation.

Vedi Repository
24 giorni faNon ancora revisionato
Condividi
Contenuto non disponibile nella lingua richiesta. Visualizzazione della versione inglese.

mlflow-scorer-scanner

Detection-only triage scanner for MLflow third-party scorer deserialization RCE, CVE-2026-116692. It asks a tracking server one safe question, its version, and grades it. It never exploits anything.

The flaw

DetailValue
CVECVE-2026-116692
AdvisoryGHSA-26p8-2jq9-3vq9
BugDeserialization of untrusted data (CWE-94), third-party scorer handling in mlflow/genai/scorers/base.py
ImpactUnauthenticated remote code execution on a default tracking server, root in the official image
Severity9.8 critical

MLflow 3.12.0 introduced registration and deserialization support for third-party scorers. The vulnerable code accepts a request-selected module path beneath one of four allowed package prefixes instead of requiring an exact allowed module, then imports that module and constructs the class during Scorer.model_validate_json(). An attacker who can place a Python module under a writable allowed package, via the normal artifact endpoints, and then submit it through /ajax-api/3.0/mlflow/scorer/invoke, gets code execution in the server process. This scanner never does any of that: no scorer request, no experiment, no run, no artifact upload.

Two source disagreements, and what this scanner follows. One write-up frames the range as 3.12.0 through 3.16.x; the advisory text itself, quoted in a second write-up, starts at 3.12.0rc0 and ends before 3.17.0. This scanner follows the advisory text, so 3.12.0rc0 grades VULNERABLE. On authentication, the advisory text says auth-disabled deployments permit the attack without credentials, while basic-auth deployments require a principal able to create experiments and runs and upload artifacts. A version in range on an auth-enabled server is therefore still a finding, but a weaker one than the same version exposed without auth.

Fixed versions

BranchStatus
MLflow < 3.12.0rc0NOT_AFFECTED, scorer support did not exist yet
MLflow 3.12.0rc0 to before 3.17.0VULNERABLE
MLflow >= 3.17.0FIXED

Updating to 3.17.0 is the fix. Until then, do not expose a tracking server without authentication, and make sure no scorer package path is writable by the server process.

What it checks

  • GET <base>/version, the endpoint MLflow itself documents for checking a tracking server version. That is the only request it makes.
  • The returned version string, graded against the range above. A JSON wrapper like {"version": "3.14.0"} is accepted; anything else unparseable grades UNKNOWN.

What it never does

No payloads. No scorer invocations. No experiment, run, or artifact creation. No upload of any file, malicious or otherwise. It reads one public version string and stops.

Grading

FindingSeverity
Version 3.12.0rc0 to before 3.17.0VULNERABLE
Version 3.17.0 or laterFIXED
Version before 3.12.0rc0NOT_AFFECTED
Version hidden behind auth, 404, or unparseableUNKNOWN, not clean
Target unreachable or not http(s)ERROR, not clean

Usage

# Triage one tracking server (default MLflow port is 5000)
python3 mlflow_scorer_scanner.py http://localhost:5000

# Machine-readable output
python3 mlflow_scorer_scanner.py https://mlflow.example.com --json

# Tests (mock servers only, no real MLflow needed)
python3 -m unittest discover -s tests -v

Exit codes: 0 = fixed, not affected, or unknown, 2 = vulnerable finding, 1 = scan error.

Honest caveats

  • A version in range is triage, not proof of exploitability. The full chain also needs a writable allowed scorer package, and on auth-enabled servers an authenticated principal with experiment, run, and artifact rights.
  • No public exploit code was indexed for this CVE at time of writing, and it was not in CISA KEV. That is a statement about what was public on 2026-10-28, not a safety claim.
  • No Nuclei template or widely-available public scanner specifically for CVE-2026-116692 was found at time of writing. Generic MLflow version and unauthenticated-access checks exist, but they do not grade this scorer range.
  • If /version is blocked or auth-gated, this scanner says UNKNOWN, not clean. Confirm the version another way before closing the finding.

References

  • GHSA-26p8-2jq9-3vq9 (MLflow advisory, affected range, fixed in 3.17.0)
  • CVE-2026-116692 record (published 2026-10-27, CVSS 9.8, CWE-94)
  • MLflow tracking server docs (the /version endpoint)

Thanks to

  • The MLflow maintainers for the advisory and the 3.17.0 fix
  • Everyone publishing MLflow tracking-server exposure research

Disclaimer

Built for defensive triage and authorized testing only. Scan only systems you own or are authorized to test. The authors are not responsible for misuse.

Research by Vineeth Kumar (foxcornlab). Found a bug or want a feature, open an issue.

Scarica lo strumento

Più Popolari

Vedi tutti →

Scopri gli strumenti più utilizzati dalla nostra community.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 Il toolkit OSINT definitivo: grafo di indagine, punteggio di affidabilità, famiglie di alias, mappa, timeline e report. 49 moduli, oltre 400 siti, nessuna chiave API richiesta.

    1.1k
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    # App di esecuzione con un tocco GhostLock (CVE-2026-43499)

    597
  • 3

    looking-glass

    anonghosty/looking-glass

    Piattaforma CSIRT che importa i report di Shadowserver in MongoDB e li riproduce come mappe interattive, analisi, risultati, avvisi e dashboard executive.

    396
  • 4

    ARTEX

    baarai/artex

    Piattaforma di penetration testing autonoma multi-agente basata su LLM con agenti planner/worker, grafi di asset e di esplorazione, registrazione del traffico MITM e retesting delle vulnerabilità.

    319
  • 5

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    Il sistema d'azione per la cybersecurity nativa AI—dove l'intento diventa esecuzione governata, le evidenze diventano memoria operativa, e ogni operazione migliora la successiva.

    274
  • 6

    Aliens_eye

    arxhr007/aliens_eye

    Motore OSINT basato sull'IA per scoprire e mappare oltre 7.700 identità sui social media in tutto il web

    260

Esplora tutti gli strumenti

Sfoglia la nostra collezione di strumenti

Vedi tutti gli strumenti →