KEVWire

Security intelligence, with context.

Response snapshot Oct 27, 2026, 23:56 UTC

Back to the wire
Permalink
VulnerabilityCVE-2026-116692

MLflow vulnerable to remote code execution on unauthenticated servers via third-party scorer deserialization importing caller-placed artifact modules

MLflow 3.12.0 introduced registration and deserialization support for third-party scorers in OSS MLflow.

Published Updated

At a glance

Exploited
Not known
Severity
9.8 Critical (CVSS 3.1)
Chance of exploitation
Not estimated
Sources
2 independent sources

What to do

GitHub lists first patched versions: mlflow: 3.17.0. Confirm the affected range and upgrade guidance in the exact advisory.

Read GitHub-reviewed advisory GHSA-26p8-2jq9-3vq9 (opens a new tab)

What happened

Summary

MLflow 3.12.0 introduced registration and deserialization support for third-party scorers in OSS MLflow. During scorer deserialization, the server accepts a request-controlled Python module path beneath one of four allowed package prefixes, imports that module, obtains a request-controlled class, and instantiates it synchronously.

An HTTP caller can combine this behavior with MLflow's built-in experiment, run, and artifact-upload endpoints:

1. Create an experiment whose local artifact location points to a writable allowed scorer package. 2. Create a run, causing MLflow to allocate / /artifacts/. 3. Upload a Python module into that run artifact directory. 4. Submit a serialized third-party scorer whose module path resolves to the uploaded file. 5. Trigger server-side import and class construction through /ajax-api/3.0/mlflow/scorer/invoke.

Continue reading

On the default no-auth server using the official MLflow v3.14.0 image, this produces unauthenticated arbitrary Python code execution with the privileges of the MLflow server process. The reproduced image runs the process as uid 0 and permits writes to its installed Python package tree.

The attached proof of concept is intentionally harmless. The uploaded constructor performs one action only: it writes a random marker beside its own uploaded run artifact. The marker is then downloaded through MLflow's run-artifact endpoint and checked for an exact match.

1. The allow-list accepts attacker-created descendants

At release commit 86cd7f5a1dc25a1387ad07c87811bbc30f62951c, mlflow/genai/scorers/scorer_utils.py defines these allowed roots:

THIRD_PARTY_SCORER_ALLOWED_MODULES = frozenset({
 "mlflow.genai.scorers.ragas",
 "mlflow.genai.scorers.deepeval",
 "mlflow.genai.scorers.trulens",
 "mlflow.genai.scorers.phoenix",
})

In mlflow/genai/scorers/base.py, the deserializer accepts either an exact root or any descendant:

module_path == m or module_path.startswith(m + ".")

It then performs the following operations using request-controlled values:

module = importlib.import_module(module_path)
scorer_class = getattr(module, class_name, None)
scorer_instance = scorer_class(**init_kwargs)

A package-name prefix is not a provenance boundary. If the server process can write below an allowed package, an attacker-created descendant can satisfy the string-prefix check and become importable Python code.

Relevant source locations at the v3.14.0 release commit:

* mlflow/genai/scorers/scorer_utils.py, allowed third-party scorer modules * mlflow/genai/scorers/base.py, third-party scorer deserialization and construction

2. Deserialization occurs before trace processing

In mlflow/server/handlers.py, _invoke_scorer_handler() parses the supplied serialized scorer and calls:

scorer = Scorer.model_validate_json(serialized_scorer)

Only after deserialization and construction does the handler call:

batches = get_trace_batches_for_scorer(trace_ids, scorer, tracking_store)

The attacker-controlled constructor therefore runs before the server validates or resolves the supplied trace IDs.

The minimal proof class is deliberately not a complete MLflow scorer. After its constructor writes the marker, later trace-batching logic accesses scorer attributes that the proof class does not implement. The resulting HTTP 500 occurs after code execution and does not invalidate the proof.

No background job execution or special job-execution server option is required to reach the constructor.

3. MLflow's artifact layout places the module under the trusted prefix

The experiment creation endpoint accepts an artifact_location. For file-backed experiments, MLflow derives each run artifact directory as:

 / /artifacts/

The proof of concept selects this package directory in the official v3.14.0 image:

file:///usr/local/lib/python3.10/site-packages/mlflow/genai/scorers/trulens

After creating a run and uploading audit_ .py, the resulting filesystem location is:

/usr/local/lib/python3.10/site-packages/mlflow/genai/scorers/trulens/ /artifacts/audit_ .py

The corresponding import path is:

mlflow.genai.scorers.trulens. .artifacts.audit_

The attack does not overwrite an existing package module. It creates a new run-scoped module in the normal file-backed artifact layout. The run ID is returned to the caller.

4. Authentication and authorization behavior

With authentication disabled, as in the default reproduced server configuration, no credentials are required.

When MLflow basic authentication is enabled, authentication still applies. However, the scorer-invocation AJAX endpoint is assigned to validate_gateway_proxy, which returns True without performing an experiment- or scorer-level resource permission check.

Practical exploitation in an authenticated deployment still requires the caller to create an experiment and run and upload artifacts to them. The key authorization issue is that scorer deserialization itself is not gated by a resource-specific permission check for the supplied experiment before the import occurs.

Proof of concept

The attached reproduction is intentionally isolated:

* exact official MLflow v3.14.0 image pinned by manifest digest; * Docker network disabled; * MLflow bound to container loopback; * no shell command or external callback; * no external request; * no host filesystem write; * marker created beside the uploaded run artifact; * marker retrieved through the run-artifact download endpoint; and * disposable container removed after completion.

Run:

./run-official-image.sh

Expected output:

MLFLOW_VERSION=3.14.0
PACKAGE_DIR=/usr/local/lib/python3.10/site-packages/mlflow/genai/scorers/trulens
CREATE_EXPERIMENT_STATUS=200
CREATE_RUN_STATUS=200
UPLOAD_STATUS=200
INVOKE_STATUS=500
PROOF_DOWNLOAD_STATUS=200
PROOF_MATCH=1
RESULT=AFFECTED

The secur

Affected package ranges reported by GitHub: mlflow: >= 3.12.0rc0, < 3.17.0

Scores and how they were measured

CVSS base severity

9.8

Version 3.1As of

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

Measurement source (opens a new tab)

How bad it is if exploited. Not a prediction of exploitation or proof you are exposed.

EPSS probability

Not available

A model's estimate of exploitation in the next 30 days, not certainty or your own risk.

Related technology

Sources

  1. Primary source

    GitHub-reviewed advisory GHSA-26p8-2jq9-3vq9 (opens a new tab)

    The public advisory supplies the description, affected package/version ranges, and any listed CVSS metric. GitHub review is not local review.

    Published Checked

  2. Primary source

    GitHub public advisory API page (opens a new tab)

    The API page supplied GHSA-26p8-2jq9-3vq9, its actual identifiers, source dates, and structured metrics.

    Published Date not suppliedChecked

  3. Primary source

    NVD CVE-2026-116692 (opens a new tab)

    NVD record: description, CVSS (the CNA score is preferred), affected products, and publication date. Not evidence of exploitation.

    Published Checked

  4. Primary source

    NVD CVE API 2.0 (opens a new tab)

    The NVD API page that supplied this CVE record.

    Published Date not suppliedChecked

  5. Primary source

    NVD CVE API 2.0 (opens a new tab)

    The NVD API page that supplied this CVE record.

    Published Date not suppliedChecked

  6. Primary source

    NVD CVE API 2.0 (opens a new tab)

    The NVD API page that supplied this CVE record.

    Published Date not suppliedChecked

Cite this record

For factual claims, cite the primary sources above; cite KEVWire for its compilation, dates, and classifications.

APA

KEVWire. (2026, October 27). MLflow vulnerable to remote code execution on unauthenticated servers via third-party scorer deserialization importing caller-placed artifact modules (CVE-2026-116692). https://kevwire.com/record/CVE-2026-116692

MLA

"MLflow vulnerable to remote code execution on unauthenticated servers via third-party scorer deserialization importing caller-placed artifact modules (CVE-2026-116692)." KEVWire, 27 Oct. 2026, kevwire.com/record/CVE-2026-116692.

Chicago

KEVWire. "MLflow vulnerable to remote code execution on unauthenticated servers via third-party scorer deserialization importing caller-placed artifact modules (CVE-2026-116692)." Last modified October 27, 2026. https://kevwire.com/record/CVE-2026-116692.

Evidence summary (plain text)

MLflow vulnerable to remote code execution on unauthenticated servers via third-party scorer deserialization importing caller-placed artifact modules (Oct 27, 2026).
MLflow 3.12.0 introduced registration and deserialization support for third-party scorers in OSS MLflow. During scorer deserialization, the server accepts a request-controlled Python module path beneath one of four allowed package prefixes, imports that module, obtains a request-controlled class, and instantiates it synchronously.

An HTTP caller can combine this behavior with MLflow's built-in experiment, run, and artifact-upload endpoints:

1. Create an experiment whose local artifact location points to a writable allowed scorer package.
2.
CVSS 9.8 (version 3.1; evidence as of Oct 27, 2026; https://github.com/advisories/GHSA-26p8-2jq9-3vq9).
Sources:
GitHub-reviewed advisory GHSA-26p8-2jq9-3vq9: https://github.com/advisories/GHSA-26p8-2jq9-3vq9
GitHub public advisory API page: https://api.github.com/advisories?type=reviewed&per_page=100&sort=updated&direction=desc
NVD CVE-2026-116692: https://nvd.nist.gov/vuln/detail/CVE-2026-116692
NVD CVE API 2.0: https://services.nvd.nist.gov/rest/json/cves/2.0?lastModStartDate=2026-10-27T21:10:59.661Z&lastModEndDate=2026-10-27T22:40:56.314Z&resultsPerPage=500&startIndex=0
NVD CVE API 2.0: https://services.nvd.nist.gov/rest/json/cves/2.0?lastModStartDate=2026-10-27T20:35:58.697Z&lastModEndDate=2026-10-27T22:10:57.840Z&resultsPerPage=500&startIndex=0
NVD CVE API 2.0: https://services.nvd.nist.gov/rest/json/cves/2.0?lastModStartDate=2026-10-27T20:06:01.291Z&lastModEndDate=2026-10-27T21:35:56.840Z&resultsPerPage=500&startIndex=0
Record ID: CVE-2026-116692.
Record origin: Ingested record.
AI evidence: Not assessed.
Known exploitation: Not assessed.
Zero-day status: Not assessed. KEV membership alone does not establish zero-day status.
About this data

This record was ingested from a data source. Retrieved: Oct 27, 2026, 23:40 UTC. Ingestion is not an independent editorial investigation.

A CISA KEV listing documents known exploitation, not necessarily a zero-day; absence from the catalog does not prove a vulnerability has never been exploited.

Technology tags indicate possible relevance, not verified exposure. Confirm the affected product, version, and configuration against your own inventory; KEVWire does not scan your systems.

A personal reading checklist on this device, not verified remediation.

Record JSON

Your account

Email alerts

Email alerts are not available yet. To follow your stack today, use its RSS feed, which honors your filters, or connect an AI agent through the API.

Suggest an improvement

Ideas, missing sources, corrections, and bugs are all welcome. No account needed. Suggestions are reviewed by a person before anything changes.

Record details

Record detail