CVE-2026-116692

validatedZero produced an exploit or a check.

GHSA-26p8-2jq9-3vq9 ↗EUVD-2026-111383 ↗

check availableZero validated a working check for it.not proven exploitableZero has a detection check, but hasn't produced a full working exploit yet.
cve.orgpublic pocExploit code has been published by someone. Source signals can also carry a PoC (Proof of Concept).−18dkevAttacks confirmed in the real world.
mlflow >= 3.12.0rc0, < 3.17.0
fixed in 3.17.0
CWE-94 Improper Control of Generation of Code ('Code Injection')
ETK-2026-2791Zero's internal tracking ID, independent of the CVE number., permanent identifier
pkg:pypi/mlflow
+44mTTV
14mAI analysisAnalysis: how long the AI spent reversing this signal.
Check if you are vulnerable
timeline
0 = cve.org publication
public poc−18dtrigger signal (cve.org) +0dcheck available+0.0dagent stopped
−15d
−10d
−5d

description

MLflow versions from 3.12.0rc0 through versions before 3.17.0 can load unintended modules when processing third-party scorers. An attacker able to place code in a server-writable allowed scorer package can cause code execution; this is unauthenticated when authentication is disabled, while basic-auth deployments require relevant artifact upload permissions.

mitigation

Upgrade MLflow to version 3.17.0. Until upgraded, restrict scorer invocation and ensure allowed third-party scorer package locations are not writable by the MLflow server process.

fixed in 3.17.0

check
available
exploit
no
automationDeterministic: a check that runs on its own, the same request on every target. Contextual: an agentic technique that needs to be applied per target, for bugs that depend on contextual flows.
deterministic
auth requiredWhether an attacker needs an account first, based on the CVSS vector's Privileges Required.
no
EPSSEstimate of the chance this CVE is exploited in the next 30 days. Refreshed daily.
–
agent run time
14m
cve.org record, t = 0The moment cve.org publishes the CVE in full.
2026-10-27 20:03Z
Δ to public pocExploit code has been published by someone. Source signals can also carry a PoC (Proof of Concept).
−18d
Δ to kevAttacks confirmed in the real world.
–

outcome

Check only, differential: the check fires on the vulnerable build and is silent on the patched one.

source signals

Public signals that triggered or updated the pipeline. These are inputs for the deltas above.

  1. 2026-10-09 08:20Z
  2. 2026-10-27 20:03Z
    cve_org: trigger signal, started the run

this entry in the feed

GET /api/v1/feed/CVE-2026-116692

{
  "etk_id": "ETK-2026-2791",
  "cve_id": "CVE-2026-116692",
  "display_id": "CVE-2026-116692",
  "ghsa_id": "GHSA-26p8-2jq9-3vq9",
  "euvd_id": "EUVD-2026-111383",
  "status": "published",
  "anchor_state": "anchored",
  "title": "Arbitrary code execution in MLflow third-party scorer loading",
  "summary": "MLflow versions from 3.12.0rc0 through versions before 3.17.0 can load unintended modules when processing third-party scorers. An attacker able to place code in a server-writable allowed scorer package can cause code execution; this is unauthenticated when authentication is disabled, while basic-auth deployments require relevant artifact upload permissions.",
  "mitigation": "Upgrade MLflow to version 3.17.0. Until upgraded, restrict scorer invocation and ensure allowed third-party scorer package locations are not writable by the MLflow server process.",
  "vendor": "mlflow",
  "product": "mlflow",
  "affected": [
    {
      "affected_notes": [],
      "affected_ranges": [
        ">= 3.12.0rc0, < 3.17.0"
      ],
      "ecosystem": "pip",
      "fixed_versions": [
        "3.17.0"
      ],
      "package_name": "mlflow",
      "purl": "pkg:pypi/mlflow"
    }
  ],
  "cwes": [
    94
  ],
  "cvss_score": 9.8,
  "cvss_src": "cna_vendor",
  "cvss_vector": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
  "epss_score": null,
  "epss_percentile": null,
  "epss_asof": null,
  "automation": "deterministic",
  "check": true,
  "check_confidence_basis": "differential",
  "exploit": false,
  "pipeline_minutes": 14.398878,
  "t0_at": "2026-10-27T20:03:49.661000+00:00",
  "signals": {
    "check": {
      "at": "2026-10-27T20:48:07.125607+00:00",
      "days": 0.030757692210648148,
      "is_lower_bound": false,
      "measured_at": "2026-10-27T20:03:49.661000+00:00",
      "url": null
    },
    "public_poc": {
      "at": "2026-10-09T08:20:37+00:00",
      "days": -18.488340983796295,
      "is_lower_bound": false,
      "measured_at": "2026-10-27T20:03:49.661000+00:00",
      "url": "https://github.com/mlflow/mlflow/security/advisories/GHSA-26p8-2jq9-3vq9"
    },
    "trigger_signal": {
      "at": "2026-10-27T20:03:49.661000+00:00",
      "days": 0,
      "is_lower_bound": false,
      "measured_at": "2026-10-27T20:03:49.661000+00:00",
      "url": "https://www.cve.org/CVERecord?id=CVE-2026-116692"
    }
  }
}

published 2026-10-27 20:33Z
last change 2026-10-27 21:33Z

Need the technical details behind an entry?

Every entry here shows what we attempted and when. The technical details, the generated exploit that fired it and the evidence behind the verdict are not public. Tell us about your use case and we’ll get back to you about extended API access.