Lucene search
+L

4082 matches found

Huntr
Huntr
added 2026/02/18 8:11 a.m.12 views

Remote Code Execution via Flow Studio Node Definitions

Description LOLLMS Flow Studio contains multiple code execution vulnerabilities via unsafe use of Python's exec function. Two distinct code paths allow arbitrary Python code execution on the server: 1. Direct Code Execution via/api/flows/testcode Admin endpoint File: backend/routers/flowstudio.py...

6.3AI score
SaveExploits0
Huntr
Huntr
added 2026/02/17 8:26 p.m.15 views

Incomplete Fix for CVE-2025-10279: get_or_create_nfs_tmp_dir() Still Creates World-Writable (0o777) Directories Enabling Local Code Execution

Description Description CVE-2025-10279 huntr bounty 01d3b81e identified that MLflow's getorcreatetmpdir created temporary directories with world-writable permissions 0o777, enabling local attackers to tamper with model artifacts and achieve arbitrary code execution. The fix PR 17544, commit...

7.8CVSS7.4AI score0.00215EPSS
SaveExploits2
Huntr
Huntr
added 2026/02/17 6:0 a.m.11 views

Missing Authorization Validation on MLflow MPU Endpoints Leads to Cross-Resource Artifact Overwrite, Model Poisoning, and Cross-Boundary Command Execution on Model Load

Analyzed version: 5af88dc08a54d40dddfc019da9e7f0fd0fcf34e2 git describe: nightly-2300-g5af88dc08, local mlflow.version: 3.10.1.dev0 In --serve-artifacts mode, MLflow exposes MPU endpoints for large-file multipart uploads. However, its authorization logic only covers the /mlflow-artifacts/artifact...

9CVSS6.1AI score0.00345EPSS
SaveExploits1
Huntr
Huntr
added 2026/02/14 2:13 a.m.13 views

Authentication Bypass on FastAPI Routes (Job API, OTel API) When Basic Auth Enabled

Summary When MLflow is started with authentication enabled --app-name basic-auth and served via uvicorn ASGI, the FastAPI permission middleware only enforces authentication on /gateway/ routes. All other FastAPI routes -- including the Job API /ajax-api/3.0/jobs/ and the OpenTelemetry trace...

8.6CVSS6AI score0.18861EPSS
SaveExploits1
Huntr
Huntr
added 2026/02/13 3:49 a.m.19 views

Authorization Bypass in SearchModelVersions Allows Any Authenticated User to Enumerate All Model Versions Regardless of Permissions

Summary MLflow's SearchModelVersions REST API endpoint GET /api/2.0/mlflow/model-versions/search and GraphQL query mlflowSearchModelVersions lack per-model authorization checks when basic auth is enabled. Any authenticated user can enumerate ALL model versions across ALL registered models,...

6.5CVSS5.8AI score0.00441EPSS
SaveExploits1
Huntr
Huntr
added 2026/02/11 9:16 a.m.9 views

Improper Origin Validation in MLflow Assistant /ajax-api Endpoints Enables Browser-Mediated Local Command Execution

Description Analyzed project version: MLflow 3.9.0 /version, commit 6e61043b0ff5d845bea479d7e7ea24dcd4b2c629. In MLflow 3.9.0, a new feature called MLflow Assistant was introduced, intended only for local development and designed to integrate with Claude Code accepting requests only from loopback...

9.6CVSS7.9AI score0.00384EPSS
SaveExploits1
Huntr
Huntr
added 2026/02/10 7:2 p.m.40 views

Arbitrary File Read via Prompt Tag Source Validation Bypass in CreateModelVersion

The createmodelversion handler in mlflow/server/handlers.py uses a client-controlled tag to decide whether to skip source path validation. When a CreateModelVersion request includes the tag mlflow.prompt.isprompt, the helper ispromptrequest returns True, and the entire source validation block...

7.5CVSS7.3AI score0.00712EPSS
SaveExploits1
Huntr
Huntr
added 2026/02/10 4:29 p.m.13 views

SSRF in MLflow via user-controlled webhook URL parameter

Description A Server-Side Request Forgery SSRF vulnerability exists in the webhook creation functionality of MLflow. The createwebhook handler accepts a user-controlled url parameter and stores it without any validation. When webhooks are tested or triggered, the sendwebhookrequest function sends...

7.1CVSS7.3AI score0.0037EPSS
SaveExploits1
Huntr
Huntr
added 2026/02/02 5:36 a.m.20 views

Zip Slip path traversal in keras.utils.get_file(..., extract=True) archive extraction

Summary Keras' download helper keras.utils.getfile..., extract=True via keras/src/utils/fileutils.py extracts zip/tar archives and attempts to filter unsafe member paths. However, the filter computes its base directory as the process CWD resolvepath"." rather than the extraction target directory...

6.2AI score
SaveExploits0
Huntr
Huntr
added 2026/01/28 12:48 p.m.19 views

Infinite Loop Denial of Service via Circular Dependencies in Functional Model Deserialization

Description A vulnerability in keras.src.models.functional.functionalfromconfig allows a Denial of Service DoS attack via an infinite loop. When reconstructing a Functional model from a configuration e.g., via keras.models.loadmodel, the deserialization logic fails to detect or break out of...

5.8AI score
SaveExploits0
Huntr
Huntr
added 2026/01/16 8:47 a.m.13 views

H2O-3 PostgreSQL Driver RCE - Bypassing CVE-2025-6544 Mitigation

Description A critical remote code execution vulnerability exists in the unauthenticated REST API endpoint /99/ImportSQLTable in H2O-3 version 3.46.0.9 and prior. The current security mitigation implemented in H2O-3 relies on a parameter blacklist mechanism that exclusively targets MySQL JDBC...

9.8CVSS6.9AI score0.00938EPSS
SaveExploits2
Huntr
Huntr
added 2026/01/14 7:2 p.m.10 views

Uncontrolled Recursion in NLTK StupidBackoff Language Model Allows Denial of Service

This report is not public...

5.3AI score
SaveExploits0
Huntr
Huntr
added 2026/01/13 3:32 a.m.17 views

Integer Overflow lead to DOS in handling Accept-Encoding header in API /v2/models/<model-name>/generate

This report is not public...

5.8AI score
SaveExploits0
Huntr
Huntr
added 2026/01/12 2:22 a.m.7 views

XSS in Chat Message Leads to Account Tackover

Description The vulnerability resides in the data persistence layer of the application. The fromdict method in the AppLollmsMessage class acts as a "sink" for raw data. It retrieves the content value from an input dictionary and assigns it directly to the object without any form of sanitization o...

8.2CVSS6AI score0.00258EPSS
SaveExploits1
Huntr
Huntr
added 2026/01/11 8:12 p.m.5 views

Arbitrary File Read via Log Symlink following in FileTaskHandler

This report is not public...

5.3AI score
SaveExploits0
Huntr
Huntr
added 2026/01/10 6:22 a.m.9 views

Stored XSS in Home Feed via Post Content Lead to Account Takeover

Description A Stored Cross-Site Scripting XSS vulnerability was identified in the social feature of the application. The backend fails to sanitize user-provided content in the post creation endpoint. This allows an attacker to inject and store malicious JavaScript, which is then executed in the...

9.6CVSS7.5AI score0.00405EPSS
SaveExploits1
Huntr
Huntr
added 2026/01/08 2:20 a.m.6 views

Session is not expiring after password resetting

This report is not public...

4.1CVSS5.9AI score0.0021EPSS
SaveExploits0
Huntr
Huntr
added 2026/01/07 1:6 p.m.6 views

Path Traversal in Agent Flows via `uuid` (Arbitrary .json File Read/Delete)

Description : Summary I discovered a Path Traversal vulnerability in the AgentFlows component that allows reading and deleting arbitrary .json files on the server. The issue stems from the improper usage of path.join combined with normalizePath. The application resolves the file path using user...

9.1CVSS7.1AI score0.00809EPSS
SaveExploits1
Huntr
Huntr
added 2026/01/07 6:18 a.m.8 views

Improper Access Control via Weak JWT Token Leads to Admin Takeover and Privilege Escalation

Description The application's session management is vulnerable to Authorization Bypass and Vertical Privilege Escalation. During dynamic analysis of the application's authentication flow, I discovered that the JSON Web Tokens JWT are signed with a weak secret key. This allowed me to perform an...

9.8CVSS5.9AI score0.0054EPSS
SaveExploits1
Huntr
Huntr
added 2026/01/07 5:21 a.m.32 views

Arbitrary Code Execution via Unsafe torch.load() in Trainer Checkpoint Loading

Summary A critical arbitrary code execution vulnerability exists in HuggingFace Transformers' Trainer class. The loadrngstate method at src/transformers/trainer.py:3059 calls torch.load without the weightsonly=True parameter. While a safeglobals context manager wraps this call, it provides no...

7.8CVSS6.6AI score0.00349EPSS
SaveExploits1
Huntr
Huntr
added 2026/01/06 2:55 p.m.6 views

TFSMLayer bypasses `safe_mode=True`, allowing attacker-controlled code execution during model inference

Summary TFSMLayer allows loading attacker-controlled TensorFlow SavedModels when deserializing a .keras model, even when safemode=True the default. While TensorFlow does not execute SavedModel functions during load, the attacker-controlled graph is registered during deserialization and executes...

8.8CVSS7.8AI score0.00405EPSS
SaveExploits1
Huntr
Huntr
added 2025/12/31 2:25 p.m.53 views

Command Injection through bash -c

This report is not public...

9.6CVSS5.8AI score0.01328EPSS
SaveExploits2
Huntr
Huntr
added 2025/12/30 10:50 a.m.5 views

Unbounded Classification Output Sorting Leads to Remote Denial-of-Service in Triton Inference Server

This report is not public...

5.3AI score
SaveExploits0
Huntr
Huntr
added 2025/12/29 5:53 p.m.7 views

Unauthenticated File Upload in LollMS

Executive Summary A critical security vulnerability has been identified in LollMS that allows unauthenticated users to upload and process files through the /api/files/extract-text endpoint. This endpoint lacks authentication requirements, contradicting the application's documented "Secure...

9.8CVSS5.8AI score0.0043EPSS
SaveExploits1
Huntr
Huntr
added 2025/12/29 5:51 p.m.6 views

Server-Side Request Forgery (SSRF) in LollMS Export Content

Executive Summary A security vulnerability has been identified in LollMS that allows Server-Side Request Forgery SSRF attacks through the /api/files/export-content endpoint. The downloadimagetotemp function downloads images from arbitrary user-controlled URLs without validation, allowing attacker...

7.5CVSS6AI score0.01765EPSS
SaveExploits2
Huntr
Huntr
added 2025/12/29 5:49 p.m.7 views

Insecure Direct Object Reference (IDOR) in LollMS Friend Request Response

Executive Summary A critical security vulnerability has been identified in LollMS that allows any authenticated user to accept or reject friend requests belonging to other users. The respondrequest function lacks authorization checks, enabling Insecure Direct Object Reference IDOR attacks. Affect...

8.3CVSS5.8AI score0.00268EPSS
SaveExploits1
Huntr
Huntr
added 2025/12/29 5:46 p.m.7 views

Content-Type Spoofing in LollMS Image Upload

Executive Summary A security vulnerability has been identified in LollMS that allows authenticated users to bypass file type validation by spoofing the Content-Type header. The /api/upload/chatimage endpoint only validates the HTTP header, not the actual file content, allowing malicious files to ...

6AI score
SaveExploits0
Huntr
Huntr
added 2025/12/27 5:2 p.m.32 views

Job API exposed without authorization

This report is not public...

9.8CVSS5.9AI score0.04392EPSS
SaveExploits1
Huntr
Huntr
added 2025/12/27 4:17 p.m.11 views

Tracing + Assessments Access

This report is not public...

8.1CVSS7.1AI score0.00331EPSS
SaveExploits1
Huntr
Huntr
added 2025/12/26 3:34 p.m.10 views

Airflow externalLogUrl Permission Bypass

1. Summary The externalLogUrl endpoint in Airflow’s FastAPI enforces only the weaker Task Instance access permission TASKINSTANCE instead of the intended Task Logs permission TASKLOGS. As a result, low-privileged users who are not authorized to view task logs can still obtain external log access...

6.1AI score
SaveExploits0
Huntr
Huntr
added 2025/12/23 7:16 a.m.18 views

Command Injection via Malicious Model Artifacts

A command injection vulnerability exists in MLflow's model serving container initialization code. When deploying a model with envmanager=LOCAL, MLflow reads dependency specifications from the model artifact's pythonenv.yaml file and directly interpolates them into a shell command without...

10CVSS6.3AI score0.01994EPSS
SaveExploits1
Huntr
Huntr
added 2025/12/23 1:15 a.m.12 views

MLflow Tarfile Path traversal in mlflow/mlflow

Description Vulnerability Report: Unsafe Tar Extraction Path Traversal Due to the lack of path traversal verification in the tar decompression part, it may lead to the possibility of overwriting any file or gaining elevated privileges. This is a non-expected vulnerability. Location File:...

10CVSS7.2AI score0.00579EPSS
SaveExploits1
Huntr
Huntr
added 2025/12/20 9:56 p.m.7 views

Apache Arrow IPC cached prebuffer path triggers signed integer overflow UB in read-range coalescing

Description Apache Arrow C++ commit d89c14b5d5203bc403fb62060fdf1ef2c0a49339 contains a signed integer overflow undefined behavior in the IO range coalescing logic, specifically in arrow/cpp/src/arrow/io/interfaces.cc:475 arrow::io::internal::CoalesceReadRanges. The overflow is reachable from...

6AI score
SaveExploits0
Huntr
Huntr
added 2025/12/14 3:18 a.m.6 views

Unsafe Pickle Deserialization in apache-airflow-providers-http leading to RCE via HttpOperator

A High severity Unsafe Deserialization vulnerability exists in the airflow.providers.http package. The HttpOperator uses pickle.loads to deserialize untrusted data received from the Triggerer service via the database in the executecomplete method. This allows an attacker who has gained write acce...

6.4AI score
SaveExploits0
Huntr
Huntr
added 2025/12/11 10:6 p.m.13 views

Arbitrary file write via tar traversal

Summary A crafted tar.gz passed to MLflow pyfunc extraction is unpacked with tarfile.extractall without path validation. Archive entries containing .. or absolute paths can escape the destination directory and write arbitrary files on the host. This is reachable when users supply prebuiltenvuri o...

9.1CVSS7.5AI score0.00852EPSS
SaveExploits1
Huntr
Huntr
added 2025/12/09 7:18 p.m.11 views

Arbitrary File Read via Absolute Path Input in nltk.util.filestring() enabling Local & Remote File Disclosure

This report is not public...

8.6CVSS5.8AI score0.00428EPSS
SaveExploits1
Huntr
Huntr
added 2025/12/08 9:27 a.m.8 views

MLflow SageMaker Command Injection Vulnerability

Description The vulnerability exists in /mlflow/sagemaker/init.py at lines 161-167, where user-supplied container image names are directly interpolated into shell commands without proper sanitization before being passed to os.system. Vulnerable Code Path : CLI Input --container parameter ↓...

8.8CVSS7.5AI score0.01456EPSS
SaveExploits1
Huntr
Huntr
added 2025/12/05 8:47 p.m.16 views

Arbitrary Code Execution in NLTK StanfordSegmenter via untrusted JAR loading

This report is not public...

10CVSS5.9AI score0.00809EPSS
SaveExploits3
Huntr
Huntr
added 2025/12/04 6:25 p.m.21 views

NLTK – Multiple CorpusReader classes allow Arbitrary File Read via Path Traversal

This report is not public...

8.6CVSS5.9AI score0.00899EPSS
SaveExploits3
Huntr
Huntr
added 2025/12/04 4:26 p.m.8 views

Arbitrary File Read via FileSystemPathPointer + PlaintextCorpusReader (bypass even if nltk.data.find() is patched

This report is not public...

5.3AI score
SaveExploits0
Huntr
Huntr
added 2025/12/04 3:39 p.m.7 views

Command Injection in example_xcom.py via XCom race condition

This report is not public...

5.8AI score
SaveExploits0
Huntr
Huntr
added 2025/11/13 5:44 p.m.17 views

Zip Slip Vulnerability in NLTK Downloader Leading to Remote Code Execution

This report is not public...

10CVSS5.5AI score0.0079EPSS
SaveExploits1
Huntr
Huntr
added 2025/11/11 9:51 a.m.12 views

Unlimited-memory decompression leads to DoS bypassing `--http-max-input-size`

This report is not public...

5.4AI score
SaveExploits0
Huntr
Huntr
added 2025/11/09 10:42 p.m.6 views

Persistent Temp-File incomplete cleanup / resource exhaustion in `transformers` Serve

Description The transformers OpenAI-compatible server leaks every base64 image it decodes to disk. Because the temporary files are never cleaned up, an attacker can exhaust disk space by repeatedly calling /v1/chat/completions with base64 imageurl entries. Vulnerable Code In...

5.8AI score
SaveExploits0
Huntr
Huntr
added 2025/11/03 5:47 a.m.4 views

Remote code execution via transformers_utils/get_config

This report is not public...

5.8AI score
SaveExploits0
Huntr
Huntr
added 2025/10/19 4:53 a.m.8 views

Path traversal vulnerability via `FileSystemPathPointer.join()` method allows unauthorized file access

Description A critical path traversal vulnerability exists in the FileSystemPathPointer.join method within the nltk library. The vulnerability allows attackers to bypass directory restrictions and access files outside the intended directory structure by using path traversal sequences such as ../ ...

5.5AI score
SaveExploits0
Huntr
Huntr
added 2025/10/07 8:15 a.m.9 views

MLFlow server is exposed to data exfiltration and destruction due to lack of Origin validation

The MLFlow REST server is vulnerable to DNS rebinding attacks, allowing malicious websites to bypass Same-Origin Policy protections and execute unauthorized calls against REST endpoints. Once rebinding is successful, the attacker can: Query for experiments via the 2.0/mlflow/experiments/search...

8.1CVSS6AI score0.00193EPSS
SaveExploits1
Huntr
Huntr
added 2025/10/03 6:25 p.m.12 views

text-generation-inference: Unbounded external image fetch in validation leads to resource-exhaustion DoS

Description Text Generation Inference Router DoS via pre-validation image fetch in VLM mode. Affected: Router workspace version 3.3.6 the latest repo, when deployed with a vision/VLM model e.g., Idefics/Mllama/Idefics2/Idefics3/Gemma3/Llama4/Paligemma/LlavaNext/Qwen2VL/Qwen25VL. Pure text LLMs do...

7.5CVSS7.2AI score0.22494EPSS
SaveExploits0
Huntr
Huntr
added 2025/10/02 4:18 a.m.7 views

Integer Overflow lead to DOS in API `v2/models/<model-name>/infer`

This report is not public...

6.9AI score
SaveExploits0
Huntr
Huntr
added 2025/09/26 7:3 a.m.7 views

Arbitrary code execution during YAML config parsing in Kubernetes materializer

Summary The Kubernetes materializer entry point feast/sdk/python/feast/infra/computeengines/kubernetes/main.py deserializes /var/feast/featurestore.yaml and /var/feast/materializationconfig.yaml using yaml.load..., Loader=yaml.Loader. Because yaml.Loader eagerly instantiates arbitrary Python...

7.8CVSS6.8AI score0.00264EPSS
SaveExploits0
Total number of security vulnerabilities4082