CveMate
Vulnerability intelligence, open to everyone. Vulnerabilities, threat metrics, KEV catalogs and exploit data from 30 sources, in one platform — no account needed.
30data sources
Star on GitHub, 8 starsLatest high-priority signals
Real-time tracking of critical vulnerabilities, ransomware campaigns, and CISA KEV additions.
CriticalCVE-2026-84999pie register · pie register premiumPie Register Premium <= 3.8.4.0 - Unauthenticated Arbitrary File Upload via piereg_file_upload AJAX HandlerThe Pie Register Premium plugin for WordPress is vulnerable to Unrestricted File Type Upload in all versions up to, and including, 3.8.4.0 via the piereg_file_upload function. This is due to the plugin performing file type validation via wp_check_filetype and wp_check_filetype_and_ext but never checking whether those validation calls produced errors before move_uploaded_file() is called, combined with no nonce or capability check on the AJAX handler. This makes it possible for unauthenticated attackers to upload files of dangerous types (e.g. PHP scripts) that may be executable, which makes remote code execution possible. The server returns the generated destination filename in the response, giving the attacker the exact path needed to trigger the uploaded file.CriticalCVE-2026-106013artbees · jupiter x coreJupiter X Core <= 4.61.0 - Unauthenticated Authentication Bypass via Twitter/X Social Login oauth_verifier ParameterThe Jupiter X Core plugin for WordPress is vulnerable to Authentication Bypass via the Twitter/X Social Login handler in all versions up to and including 4.61.0. The vulnerability exists within the verify_twitter_callback function because after exchanging the visitor's oauth_verifier for their access token, the handler never calls setOauthToken() to reassign the returned credentials to the TwitterOAuth client, causing the subsequent account/verify_credentials API call to remain signed with the site operator's stored Access Token and Access Token Secret rather than the authenticating visitor's token. This causes the X API to return the profile of the X application owner instead of the signing-in visitor, and the plugin then resolves and authenticates a WordPress account solely by matching that owner's email address via wp_set_auth_cookie(), allowing unauthenticated attackers to gain authenticated access as the WordPress user whose email corresponds to the X app owner, who is typically the site administrator.CriticalCVE-2026-110879debian · jackson-databind +1 morejackson-databind: PolymorphicTypeValidator bypass via generic type arguments and array element types allows arbitrary class instantiation (incomplete fix for CVE-2026-54512)jackson-databind validates the outer class of a class-name type ID but accepts attacker-selected classes inside its generic arguments, even when those classes violate the declared type or would be rejected as direct type IDs. This bypasses the generic-argument validation added for CVE-2026-54512 (GHSA-j3rv-43j4-c7qm), of which this is an incomplete fix. DatabindContext._resolveAndValidateGeneric() checks the outer class against the declared base type and then passes each argument from the type ID to _validateTypeParameter(), which validates the argument against the outer base type rather than enforcing its corresponding declared generic binding. It also invokes validators with arguments that have not passed the subtype check those validators expect. Under the default DefaultBaseTypeLimitingValidator this accepts an unrelated class inside a generic subtype even when the same class is rejected as a direct type ID. Where an allowIfBaseType(...) rule approves the outer type, Jackson substitutes LaissezFaireSubTypeValidator, so the configured validator no longer restricts the embedded arguments. The element type of a plain array type ID such as [Lpkg.Gadget; is likewise left unvalidated once the array's base type is approved. Affected paths include annotation-based class-name polymorphism with a default Jackson 3.x JsonMapper, configured default typing in both Jackson 2.x and 3.x, and integrations that expose that default typing. The default-validator case requires no application-authored validator: with @JsonTypeInfo(use = Id.CLASS), a direct gadget type ID is rejected as "Not a subtype", while a type ID naming a generic wrapper around the same gadget is accepted and deserialized, including attacker-supplied bean properties. Where class-name default typing permits Map, a property declared as Map<String,String> accepts HashMap<FileSystemXmlApplicationContext,String>; map-key discovery then passes the attacker-controlled JSON property name to the selected class's one-string constructor. With Spring on the classpath and attacker-controlled XML the application can load, that constructor loads the XML and invokes ProcessBuilder.start() under the application account, giving arbitrary command execution. This issue affects com.fasterxml.jackson.core:jackson-databind from 2.10.0 and tools.jackson.core:jackson-databind from 3.0.0. Releases before 2.10.0 are not affected, because PolymorphicTypeValidator was introduced in jackson-databind 2.10.0.CriticalCVE-2026-116692mlflow +1 moreMLflow: Remote code execution on unauthenticated MLflow servers via third-party scorer deserialization importing caller-placed artifact modulesMLflow vulnerable to remote code execution on unauthenticated servers via third-party scorer deserialization importing caller-placed artifact modules
## 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 `<artifact_location>/<run_id>/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`.
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.
## Technical details
### 1. The allow-list accepts attacker-created descendants
At release commit `86cd7f5a1dc25a1387ad07c87811bbc30f62951c`, `mlflow/genai/scorers/scorer_utils.py` defines these allowed roots:
```python
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:
```python
module_path == m or module_path.startswith(m + ".")
```
It then performs the following operations using request-controlled values:
```python
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:
```python
scorer = Scorer.model_validate_json(serialized_scorer)
```
Only after deserialization and construction does the handler call:
```python
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:
```text
<artifact_location>/<run_id>/artifacts/
```
The proof of concept selects this package directory in the official v3.14.0 image:
```text
file:///usr/local/lib/python3.10/site-packages/mlflow/genai/scorers/trulens
```
After creating a run and uploading `audit_<token>.py`, the resulting filesystem location is:
```text
/usr/local/lib/python3.10/site-packages/mlflow/genai/scorers/trulens/<run_id>/artifacts/audit_<token>.py
```
The corresponding import path is:
```text
mlflow.genai.scorers.trulens.<run_id>.artifacts.audit_<token>
```
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:
```bash
./run-official-image.sh
```
Expected output:
```text
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 security-relevant result is:
```text
PROOF_MATCH=1
RESULT=AFFECTED
```
This proves that the attacker-controlled constructor executed in the MLflow server process before the later scorer-processing failure.
## Reproduction environment
* **Release:** MLflow v3.14.0
* **Release commit:** `86cd7f5a1dc25a1387ad07c87811bbc30f62951c`
* **Official image:** `ghcr.io/mlflow/mlflow:v3.14.0`
* **Manifest digest:** `sha256:967de3ca34b6a5751718e380a0a383eb0ce1a6f5b484f363bd0da7559117a5f8`
* **Python version:** 3.10.18
* **Image-configured user:** empty
* **Observed runtime user:** uid 0
* **Server bind:** `127.0.0.1:5000`
* **Container network:** disabled
* **Backend store:** `sqlite:////work/mlflow.db`
* **Default artifact root:** `/work/artifacts`
* **Authentication:** disabled
* **Special job-execution option:** none
Server command inside the disposable container:
```bash
mlflow server \
--host 127.0.0.1 \
--port 5000 \
--backend-store-uri sqlite:////work/mlflow.db \
--default-artifact-root /work/artifacts
```
## Affected range and source lineage
The vulnerable third-party scorer deserialization behavior was introduced by commit:
```text
02eea26035b9d8e65af31e6e8f058ca195f62cdd
Enable third-party scorer registration in OSS MLflow (#22634)
```
The feature is absent from v3.11.1 and first appears in v3.12.0.
Runtime reproduction succeeds on v3.14.0. The relevant `mlflow/genai/scorers/base.py` content is unchanged at master commit:
```text
20858c9426a9644dfcf9545a92cbb534376779eb
```
No fixed release was identified as of 2026-07-12.
## Impact
On the reproduced stock no-auth image, any network caller who can reach the tracking server can execute Python code as the MLflow process without credentials or user interaction.
Successful exploitation provides confidentiality, integrity, and availability impact over all data, credentials, network access, and filesystem locations available to the MLflow server process.
The stock image's runtime uid 0 increases container-level impact. Host impact still depends on the container runtime configuration, mounted paths, granted capabilities, socket exposure, and other isolation controls.
In authentication-enabled deployments, exploitation requires an authenticated principal capable of creating an experiment and run and uploading artifacts. The exact chain also requires the server process to be able to write beneath one of the allowed scorer packages. This condition holds in the reproduced official image and may hold in virtual environments installed and run by the same operating-system user.
A read-only container filesystem or a scorer package directory owned by a different, non-writable user blocks this exact placement chain.
## Artifacts and attachments
### Remote proof-of-concept script
POC Script: [new_remote_poc.py](https://github.com/user-attachments/files/29936616/new_remote_poc.py)
The remote proof of concept was then executed as follows. The disposable machine’s public IP has been redacted:
```bash
python3 new_remote_poc.py \
--target http://<REDACTED_PUBLIC_IP> \
--package-dir /usr/local/lib/python3.10/site-packages/mlflow/genai/scorers/trulens \
--acknowledge-owned-target I_OWN_THIS_TARGET
```
The complete sanitized output was:
```text
TARGET=http://<REDACTED_PUBLIC_IP>
VERSION_STATUS=200
MLFLOW_VERSION=3.14.0
CREATE_EXPERIMENT_STATUS=200
EXPERIMENT_ID=2
CREATE_RUN_STATUS=200
RUN_ID=9a2d665eb38940b68983f0586f623115
UPLOAD_STATUS=200
IMPORT_PATH=mlflow.genai.scorers.trulens.9a2d665eb38940b68983f0586f623115.artifacts.audit_1783843760683972103
INVOKE_STATUS=500
INVOKE_RESPONSE_PREVIEW='<!doctype html>\n<html lang=en>\n<title>500 Internal Server Error</title>\n<h1>Internal Server Error</h1>\n<p>The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.</p>\n'
PROOF_DOWNLOAD_STATUS=200
EXPECTED_MARKER_SHA256=7a423b479df3524567fdf8cc3e7f6c4bfccc577fbcc4bac9519db8bd1cc61ed5
PROOF_MATCH=1
SERVER_CLEANUP_PATH=/usr/local/lib/python3.10/site-packages/mlflow/genai/scorers/trulens/9a2d665eb38940b68983f0586f623115
PROOF_OUTPUT_BEGIN
{
"cwd": "/",
"machine": "x86_64",
"marker": "MLFLOW_REMOTE_SAFE_MARKER_1783843760683972103",
"os_release": "PRETTY_NAME=\"Debian GNU/Linux 11 (bullseye)\"\nNAME=\"Debian GNU/Linux\"\nVERSION_ID=\"11\"\nVERSION=\"11 (bullseye)\"\nVERSION_CODENAME=bullseye\nID=debian\nHOME_URL=\"https://www.debian.org/\"\nSUPPORT_URL=\"https://www.debian.org/support\"\nBUG_REPORT_URL=\"https://bugs.debian.org/\"\n",
"platform": "Linux-7.0.0-1006-aws-x86_64-with-glibc2.31",
"python_implementation": "CPython",
"python_version": "3.10.18",
"server_datetime_utc": "2026-07-12T08:09:43.555055+00:00",
"whoami": "root",
"whoami_returncode": 0,
"whoami_stderr": ""
}
PROOF_OUTPUT_END
RESULT=AFFECTED
EXPERIMENT_METADATA_CLEANUP_STATUS=200
CLEANUP_NOTE=discard the temporary test instance or remove SERVER_CLEANUP_PATH on the owned host
```
The proof file was also verified directly inside the disposable container:
```bash
cat /usr/local/lib/python3.10/site-packages/mlflow/genai/scorers/trulens/9a2d665eb38940b68983f0586f623115/artifacts/execution-proof-1783843760683972103.json
```
The file contained:
```json
{
"cwd": "/",
"machine": "x86_64",
"marker": "MLFLOW_REMOTE_SAFE_MARKER_1783843760683972103",
"os_release": "PRETTY_NAME=\"Debian GNU/Linux 11 (bullseye)\"\nNAME=\"Debian GNU/Linux\"\nVERSION_ID=\"11\"\nVERSION=\"11 (bullseye)\"\nVERSION_CODENAME=bullseye\nID=debian\nHOME_URL=\"https://www.debian.org/\"\nSUPPORT_URL=\"https://www.debian.org/support\"\nBUG_REPORT_URL=\"https://bugs.debian.org/\"\n",
"platform": "Linux-7.0.0-1006-aws-x86_64-with-glibc2.31",
"python_implementation": "CPython",
"python_version": "3.10.18",
"server_datetime_utc": "2026-07-12T08:09:43.555055+00:00",
"whoami": "root",
"whoami_returncode": 0,
"whoami_stderr": ""
}
```
The results demonstrate that:
* the attacker-controlled module was uploaded beneath an allowed scorer package;
* MLflow imported the request-selected module and invoked its constructor;
* the constructor executed as the MLflow server process, which ran as `root` in the reproduced official image;
* the constructor created a server-side proof file containing a random marker and bounded runtime information;
* the proof file was retrieved successfully through MLflow’s artifact-download endpoint; and
* the later HTTP 500 occurred only after constructor execution.
The authoritative reproduction indicators are:
```text
UPLOAD_STATUS=200
PROOF_DOWNLOAD_STATUS=200
PROOF_MATCH=1
RESULT=AFFECTED
```
The disposable cloud machine and container were removed after testing.
Local Reproduction Scripts:
[run-official-image.sh](https://github.com/user-attachments/files/29936514/run-official-image.sh)
[poc_local.py](https://github.com/user-attachments/files/29936506/poc_local.py)CriticalCVE-2026-104353adobe · adobe campaign classicAdobe Campaign Classic (ACC) | Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection') (CWE-78)Adobe Campaign Classic (ACC) is affected by an Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection') vulnerability that could result in arbitrary code execution in the context of the current user. A low-privileged attacker could exploit this vulnerability to execute arbitrary code. Exploitation of this issue does not require user interaction. Scope is changed.CriticalCVE-2026-105780adobe · adobe campaign classicAdobe Campaign Classic (ACC) | Incorrect Authorization (CWE-863)Adobe Campaign Classic (ACC) is affected by an Incorrect Authorization vulnerability that could result in arbitrary code execution in the context of the current user. A low-privileged attacker could exploit this vulnerability to execute arbitrary code. Exploitation of this issue does not require user interaction. Scope is changed.CriticalCVE-2026-105489adobe · adobe campaign classicAdobe Campaign Classic (ACC) | Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection') (CWE-89)Adobe Campaign Classic (ACC) is affected by an Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection') vulnerability that could result in arbitrary code execution in the context of the current user. A low-privileged attacker could exploit this vulnerability to execute arbitrary code. Exploitation of this issue does not require user interaction. Scope is changed.CriticalCVE-2026-105781adobe · adobe campaign classicAdobe Campaign Classic (ACC) | Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection') (CWE-89)Adobe Campaign Classic (ACC) is affected by an Improper Neutralization of Special Elements used in an SQL Command ('SQL Injection') vulnerability that could result in arbitrary code execution in the context of the current user. An attacker could exploit this vulnerability to execute arbitrary SQL commands, potentially gaining elevated access or control over the victim's account or session. Exploitation of this issue does not require user interaction. Scope is changed.CriticalCVE-2026-104351adobe · adobe campaign classicAdobe Campaign Classic (ACC) | Improper Control of Generation of Code ('Code Injection') (CWE-94)Adobe Campaign Classic (ACC) is affected by an Improper Control of Generation of Code ('Code Injection') vulnerability that could result in arbitrary code execution in the context of the current user. A low-privileged attacker could exploit this vulnerability to execute arbitrary code. Exploitation of this issue does not require user interaction. Scope is changed.CriticalCVE-2026-114595the epics community · epics baseInteger overflow or wraparound in EPICS BaseIn EPICS Base release 7.0.10 and prior, an integer overflow in network request processing can cause memory to be allocated with an incorrect size. Subsequent handling of attacker supplied data may result in memory corruption, denial of service, or potential remote code execution.