CVE-20XX-?
agent?

CVEs of 2025.

Every CVE published in 2025 that we hold, ranked by exploit odds.

584 CVEs · updated 2026-11-20 16:32 UTC

Reset all
  1. CVE-2024-10986

    GPT Academic version 3.83 is vulnerable to a Local File Read (LFI) vulnerability through its HotReload function. This function can download and extract tar.gz files from arxiv.org. Despite implementing protections against path traversal, the application overlooks the Tarslip triggered by symlinks. This oversight allows attackers to read arbitrary local files from the victim server.

    Disclosure Server binary-husky Gpt Academic

    8.8 HIGH 0.8% exploit odds No fix listed
  2. CVE-2020-36851

    Rob--W cors-anywhere instances configured as an open proxy allow unauthenticated external users to induce the server to make HTTP requests to arbitrary targets (SSRF). Because the proxy forwards requests and headers, an attacker can reach internal-only endpoints and link-local metadata services, retrieve instance role credentials or other sensitive metadata, and interact with internal APIs and services that are not intended to be internet-facing. The vulnerability is exploitable by sending crafted requests to the proxy with the target resource encoded in the URL; many cors-anywhere deployments forward arbitrary methods and headers (including PUT), which can permit exploitation of IMDSv2 workflows as well as access to internal management APIs. Successful exploitation can result in theft of cloud credentials, unauthorized access to internal services, remote code execution or privilege escalation (depending on reachable backends), data exfiltration, and full compromise of cloud resources. Mitigation includes: restricting the proxy to trusted origins or authentication, whitelisting allowed target hosts, preventing access to link-local and internal IP ranges, removing support for unsafe HTTP methods/headers, enabling cloud provider mitigations, and deploying network-level protections.

    Disclosure Server

    9.5 CRITICAL 0.8% exploit odds Fix listed
  3. CVE-2025-55282

    aiven-db-migrate is an Aiven database migration tool. Prior to 1.0.7, there is a privilege escalation vulnerability that allows a user to elevate to superuser inside PostgreSQL databases during a migration from an untrusted source server. By exploiting a lack of search_path restriction, an attacker can override pg_catalog and execute untrusted operators as a superuser. This vulnerability is fixed in 1.0.7.

    Disclosure Server Aiven aiven-db-migrate

    9.1 CRITICAL 0.8% exploit odds Fix listed
  4. CVE-2025-1086

    A vulnerability has been found in Safetytest Cloud-Master Server up to 1.1.1 and classified as critical. This vulnerability affects unknown code of the file /static/. The manipulation leads to path traversal: '../filedir'. The attack can be initiated remotely. The exploit has been disclosed to the public and may be used. The vendor was contacted early about this disclosure but did not respond in any way.

    Disclosure Server

    6.9 MEDIUM 0.8% exploit odds No fix listed
  5. CVE-2025-6167

    A vulnerability classified as critical has been found in themanojdesai python-a2a up to 0.5.5. Affected is the function create_workflow of the file python_a2a/agent_flow/server/api.py. The manipulation leads to path traversal. Upgrading to version 0.5.6 is able to address this issue. It is recommended to upgrade the affected component.

    Disclosure Server Themanojdesai Python a2a

    5.1 MEDIUM 0.8% exploit odds Fix listed
  6. CVE-2025-32633

    Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') vulnerability in neoslab Database Toolset database-toolset allows Path Traversal.This issue affects Database Toolset: from n/a through <= 1.8.4.

    Disclosure Server

    8.6 HIGH 0.8% exploit odds No fix listed
  7. CVE-2025-65897

    zdh_web is a data collection, processing, monitoring, scheduling, and management platform. In zdh_web thru 5.6.17, insufficient validation of file upload paths in the application allows an authenticated user to write arbitrary files to the server file system, potentially overwriting existing files and leading to privilege escalation or remote code execution.

    Disclosure Server Zhaoyachao Zdh Web

    8.8 HIGH 0.8% exploit odds Fix listed
  8. CVE-2024-56324

    GoCD is a continuous deliver server. GoCD versions prior to 24.4.0 can allow GoCD "group admins" to abuse ability to edit the raw XML configuration for groups they administer to trigger XML External Entity (XXE) injection on the GoCD server. Theoretically, the XXE vulnerability can result in additional attacks such as SSRF, information disclosure from the GoCD server, and directory traversal, although these additional attacks have not been explicitly demonstrated as exploitable. This issue is fixed in GoCD 24.5.0. Some workarounds are available. One may temporarily block access to `/go/*/pipelines/snippet` routes from an external reverse proxy or WAF if one's "group admin" users do not need the functionality to edit the XML of pipelines directly (rather than using the UI, or using a configuration repository). One may also prevent external access from one's GoCD server to arbitrary locations using some kind of environment egress control.

    Disclosure Server Thoughtworks Gocd

    2.1 LOW 0.8% exploit odds Fix listed
  9. CVE-2025-27787

    Applio is a voice conversion tool. Versions 3.2.8-bugfix and prior are vulnerable to denial of service (DoS) in restart.py. `model_name` in train.py takes user input, and passes it to the `stop_train` function in restart.py, which uses it construct a path to a folder with `config.json`. That `config.json` is opened and the list of values under "process_pids" are read. Next all the process IDs listed in the JSON are killed. Using one of the arbitrary file writes, one can write to `logs/foobar` a `config.json` file, which contains a list of process IDs. Then one can access this endpoint to kill these processes. Since an attacker can't know what process is running on which process ID, they can send a list of hundreds of process IDs, which can kill the process that applio is using to run, as well as other, potentially important processes, which leads to DoS. Note that constructing a path with user input also enables path traversal. For example, by supplying "../../" in `model_name` one can access `config.json` freom locations two folders down on the server. As of time of publication, no known patches are available.

    Disclosure Server Applio

    7.8 HIGH 0.8% exploit odds Fix listed
  10. CVE-2024-43779

    An information disclosure vulnerability exists in the Vault API functionality of ClearML Enterprise Server 3.22.5-1533. A specially crafted HTTP request can lead to reading vaults that have been previously disabled, possibly leaking sensitive credentials. An attacker can send a series of HTTP requests to trigger this vulnerability.

    Disclosure Server Clear Clearml Enterprise Server

    7.7 HIGH 0.8% exploit odds No fix listed
  11. CVE-2025-5740

    CWE-22: Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') vulnerability exists that could cause arbitrary file writes when an authenticated user on the web server manipulates file path.

    Disclosure Server

    8.6 HIGH 0.8% exploit odds No fix listed
  12. CVE-2024-6854

    In h2oai/h2o-3 version 3.46.0, the endpoint for exporting models does not restrict the export location, allowing an attacker to export a model to any file in the server's file structure, thereby overwriting it. This vulnerability can be exploited to overwrite any file on the target server with a trained model file, although the content of the overwrite is not controllable by the attacker.

    Disclosure Server h2o

    7.1 HIGH 0.8% exploit odds No fix listed
  13. CVE-2025-4377

    Improper Limitation of a Pathname caused a Path Traversal vulnerability in Sparx Systems Pro Cloud Server. This vulnerability is present in logview.php and it allows reading arbitrary files on the filesystem.  Logview is accessible on Pro Cloud Server Configuration interface. This issue affects Pro Cloud Server: earlier than 6.0.165.

    Disclosure Server

    8.3 HIGH 0.8% exploit odds No fix listed
  14. CVE-2024-4982

    A directory traversal vulnerability was discovered in Pagure server. If a malicious user submits a specially cratfted git repository they could discover secrets on the server.

    Disclosure Server Redhat Pagure

    7.6 HIGH 0.8% exploit odds Fix listed
  15. CVE-2024-8438

    A path traversal vulnerability exists in modelscope/agentscope version v.0.0.4. The API endpoint `/api/file` does not properly sanitize the `path` parameter, allowing an attacker to read arbitrary files on the server.

    Disclosure Server Modelscope Agentscope

    7.5 HIGH 0.8% exploit odds No fix listed
  16. CVE-2025-56869

    Directory traversal vulnerability in Sync In server thru 1.1.1 allowing authenticated attackers to gain read and write access to the system via FilesManager.saveMultipart function in backend/src/applications/files/services/files-manager.service.ts, and FilesManager.compress function in backend/src/applications/files/services/files-manager.service.ts.

    Disclosure Server sync-in Server

    5.3 MEDIUM 0.8% exploit odds Fix listed
  17. CVE-2025-55149

    Tiny-Scientist is a lightweight framework for automating the entire lifecycle of scientific research—from ideation to implementation, writing, and review. In versions 0.1.1 and below, a critical path traversal vulnerability has been identified in the review_paper function in backend/app.py. The vulnerability allows malicious users to access arbitrary PDF files on the server by providing crafted file paths that bypass the intended security restrictions. This vulnerability allows attackers to: read any PDF file accessible to the server process, potentially access sensitive documents outside the intended directory and perform reconnaissance on the server's file system structure. This issue does not currently have a fix.

    Disclosure Server

    6.7 MEDIUM 0.8% exploit odds Fix listed
  18. CVE-2025-54347

    A Directory Traversal vulnerability was found in the Application Server of Desktop Alert PingAlert version 6.1.0.11 to 6.1.1.2 which allows an attacker to write arbitrary files under certain conditions.

    Disclosure Server Desktopalert Pingalert Application Server

    9.9 CRITICAL 0.8% exploit odds Fix listed
  19. CVE-2025-59503

    Server-side request forgery (ssrf) in Azure Compute Gallery allows an unauthorized attacker to elevate privileges over a network.

    Disclosure Server Microsoft Azure Compute Resource Provider

    10.0 CRITICAL 0.8% exploit odds Fix listed
  20. CVE-2024-12882

    comfyanonymous/comfyui version v0.2.4 suffers from a non-blind Server-Side Request Forgery (SSRF) vulnerability. This vulnerability can be exploited by combining the REST APIs `POST /internal/models/download` and `GET /view`, allowing attackers to abuse the victim server's credentials to access unauthorized web resources.

    Disclosure Server Comfy Comfyui

    7.5 HIGH 0.8% exploit odds No fix listed
  21. CVE-2025-26357

    A CWE-35 "Path Traversal" in maxtime/api/database/database.lua in Q-Free MaxTime less than or equal to version 2.11.0 allows an authenticated remote attacker to read sensitive files via crafted HTTP requests.

    Disclosure Server q-free Maxtime

    4.9 MEDIUM 0.8% exploit odds No fix listed
  22. CVE-2024-12766

    parisneo/lollms-webui version V13 (feather) suffers from a Server-Side Request Forgery (SSRF) vulnerability in the `POST /api/proxy` REST API. Attackers can exploit this vulnerability to abuse the victim server's credentials to access unauthorized web resources by specifying the JSON parameter `{"url":"http://steal.target"}`. Existing security mechanisms such as `forbid_remote_access(lollmsElfServer)`, `lollmsElfServer.config.headless_server_mode`, and `check_access(lollmsElfServer, request.client_id)` do not protect against this vulnerability.

    Disclosure Server Lollms Web Ui

    7.5 HIGH 0.8% exploit odds No fix listed
  23. CVE-2024-12376

    A Server-Side Request Forgery (SSRF) vulnerability was identified in the lm-sys/fastchat web server, specifically in the affected version git 2c68a13. This vulnerability allows an attacker to access internal server resources and data that are otherwise inaccessible, such as AWS metadata credentials.

    Disclosure Server lm-sys Fastchat

    7.5 HIGH 0.8% exploit odds No fix listed
  24. CVE-2025-26864

    Exposure of Sensitive Information to an Unauthorized Actor, Insertion of Sensitive Information into Log File vulnerability in the OpenIdAuthorizer of Apache IoTDB. This issue affects Apache IoTDB: from 0.10.0 through 1.3.3, from 2.0.1-beta before 2.0.2. Users are recommended to upgrade to version 1.3.4 and 2.0.2, which fix the issue.

    Disclosure Server Apache Iotdb

    7.5 HIGH 0.8% exploit odds Fix listed
  25. CVE-2025-26795

    Exposure of Sensitive Information to an Unauthorized Actor, Insertion of Sensitive Information into Log File vulnerability in Apache IoTDB JDBC driver. This issue affects iotdb-jdbc: from 0.10.0 through 1.3.3, from 2.0.1-beta before 2.0.2. Users are recommended to upgrade to version 2.0.2 and 1.3.4, which fix the issue.

    Disclosure Server Apache Iotdb

    7.5 HIGH 0.8% exploit odds Fix listed
  26. CVE-2024-56514

    Karmada is a Kubernetes management system that allows users to run cloud-native applications across multiple Kubernetes clusters and clouds. Prior to version 1.12.0, both in karmadactl and karmada-operator, it is possible to supply a filesystem path, or an HTTP(s) URL to retrieve the custom resource definitions(CRDs) needed by Karmada. The CRDs are downloaded as a gzipped tarfile and are vulnerable to a TarSlip vulnerability. An attacker able to supply a malicious CRD file into a Karmada initialization could write arbitrary files in arbitrary paths of the filesystem. From Karmada version 1.12.0, when processing custom CRDs files, CRDs archive verification is utilized to enhance file system robustness. A workaround is available. Someone who needs to set flag `--crd` to customize the CRD files required for Karmada initialization when using `karmadactl init` to set up Karmada can manually inspect the CRD files to check whether they contain sequences such as `../` that would alter file paths, to determine if they potentially include malicious files. When using karmada-operator to set up Karmada, one must upgrade one's karmada-operator to one of the fixed versions.

    Disclosure Server

    5.3 MEDIUM 0.8% exploit odds Fix listed
  27. CVE-2025-65878

    The warehouse management system version 1.2 contains an arbitrary file read vulnerability. The endpoint `/file/showImageByPath` does not sanitize user-controlled path parameters. An attacker could exploit directory traversal to read arbitrary files on the server's file system. This could lead to the leakage of sensitive system information.

    Disclosure Server Yeqifu Warehouse Management System

    7.5 HIGH 0.7% exploit odds No fix listed
  28. CVE-2025-30677

    Apache Pulsar contains multiple connectors for integrating with Apache Kafka. The Pulsar IO Apache Kafka Source Connector, Sink Connector, and Kafka Connect Adaptor Sink Connector log sensitive configuration properties in plain text in application logs. This vulnerability can lead to unintended exposure of credentials in log files, potentially allowing attackers with access to these logs to obtain Apache Kafka credentials. The vulnerability's impact is limited by the fact that an attacker would need access to the application logs to exploit this issue. This issue affects Apache Pulsar IO's Apache Kafka connectors in all versions before 3.0.11, 3.3.6, and 4.0.4. 3.0.x version users should upgrade to at least 3.0.11. 3.3.x version users should upgrade to at least 3.3.6. 4.0.x version users should upgrade to at least 4.0.4. Users operating versions prior to those listed above should upgrade to the aforementioned patched versions or newer versions.

    Disclosure Server Apache Pulsar

    6.3 MEDIUM 0.7% exploit odds Fix listed
  29. CVE-2024-10707

    gaizhenbiao/chuanhuchatgpt version git d4ec6a3 is affected by a local file inclusion vulnerability due to the use of the gradio component gr.JSON, which has a known issue (CVE-2024-4941). This vulnerability allows unauthenticated users to access arbitrary files on the server by uploading a specially crafted JSON file and exploiting the improper input validation in the handle_dataset_selection function.

    Disclosure Server Gaizhenbiao Chuanhuchatgpt

    6.5 MEDIUM 0.7% exploit odds No fix listed
  30. CVE-2024-8955

    A Server-Side Request Forgery (SSRF) vulnerability exists in composiohq/composio version v0.4.4. This vulnerability allows an attacker to read the contents of any file in the system by exploiting the BROWSERTOOL_GOTO_PAGE and BROWSERTOOL_GET_PAGE_DETAILS actions.

    Disclosure Server Composio

    7.5 HIGH 0.7% exploit odds No fix listed
  31. CVE-2025-34154

    UnForm Server Manager versions prior to 10.1.12 expose an unauthenticated file read vulnerability via its log file analysis interface. The flaw resides in the arc endpoint, which accepts a fl parameter to specify the log file to be opened. Due to insufficient input validation and lack of path sanitization, attackers can supply relative paths to access arbitrary files on the host system — including sensitive OS-level files — without authentication.

    Disclosure Server

    9.2 CRITICAL 0.7% exploit odds No fix listed
  32. CVE-2025-54446

    Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') vulnerability in Samsung Electronics MagicINFO 9 Server allows Upload a Web Shell to a Web Server.This issue affects MagicINFO 9 Server: less than 21.1080.0

    Disclosure Server Samsung Magicinfo 9 Server

    9.8 CRITICAL 0.7% exploit odds Fix listed
  33. CVE-2025-54438

    Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') vulnerability in Samsung Electronics MagicINFO 9 Server allows Upload a Web Shell to a Web Server.This issue affects MagicINFO 9 Server: less than 21.1080.0

    Disclosure Server Samsung Magicinfo 9 Server

    9.8 CRITICAL 0.7% exploit odds Fix listed
  34. CVE-2025-51591

    A Server-Side Request Forgery (SSRF) in JGM Pandoc v3.6.4 allows attackers to gain access to and compromise the whole infrastructure via injecting a crafted iframe. Note: Some users have stated that Pandoc by default can retrieve and parse untrusted HTML content which can enable SSRF vulnerabilities. Using the ‘--sandbox’ option or ‘pandoc-server’ can mitigate such vulnerabilities. Using pandoc with an external ‘--pdf-engine’ can also enable SSRF vulnerabilities, such as CVE-2022-35583 in wkhtmltopdf.

    Disclosure Server

    3.7 LOW 0.7% exploit odds Fix listed
  35. CVE-2024-8952

    A Server-Side Request Forgery (SSRF) vulnerability exists in composiohq/composio version v0.4.2, specifically in the /api/actions/execute/WEBTOOL_SCRAPE_WEBSITE_CONTENT endpoint. This vulnerability allows an attacker to read files, access AWS metadata, and interact with local services on the system.

    Disclosure Server Composio

    7.5 HIGH 0.7% exploit odds No fix listed
  36. CVE-2024-10830

    A Path Traversal vulnerability exists in the eosphoros-ai/db-gpt version 0.6.0 at the API endpoint `/v1/resource/file/delete`. This vulnerability allows an attacker to delete any file on the server by manipulating the `file_key` parameter. The `file_key` parameter is not properly sanitized, enabling an attacker to specify arbitrary file paths. If the specified file exists, the application will delete it.

    Disclosure Server Dbgpt db-gpt

    8.2 HIGH 0.7% exploit odds No fix listed
  37. CVE-2024-10648

    A path traversal vulnerability exists in the Gradio Audio component of gradio-app/gradio, as of version git 98cbcae. This vulnerability allows an attacker to control the format of the audio file, leading to arbitrary file content deletion. By manipulating the output format, an attacker can reset any file to an empty file, causing a denial of service (DOS) on the server.

    Disclosure Server Gradio Project Gradio

    8.2 HIGH 0.7% exploit odds No fix listed
  38. CVE-2025-66626

    Argo Workflows is an open source container-native workflow engine for orchestrating parallel jobs on Kubernetes. Versions 3.6.13 and below and versions 3.7.0 through 3.7.4, contain unsafe untar code that handles symbolic links in archives. Concretely, the computation of a link's target and the subsequent check are flawed. An attacker can overwrite the file /var/run/argo/argoexec with a script of their choice, which would be executed at the pod's start. The patch deployed against CVE-2025-62156 is ineffective against malicious archives containing symbolic links. This issue is fixed in versions 3.6.14 and 3.7.5.

    Disclosure Server Argoproj Argo Workflows

    8.1 HIGH 0.7% exploit odds Fix listed
  39. CVE-2025-34490

    GFI MailEssentials prior to version 21.8 is vulnerable to an XML External Entity (XXE) issue. An authenticated and remote attacker can send crafted HTTP requests to read arbitrary system files.

    Disclosure Server Gfi Mailessentials

    6.5 MEDIUM 0.7% exploit odds No fix listed
  40. CVE-2025-10951

    A vulnerability was identified in geyang ml-logger up to acf255bade5be6ad88d90735c8367b28cbe3a743. Affected by this vulnerability is the function log_handler of the file ml_logger/server.py. Such manipulation of the argument File leads to path traversal. It is possible to launch the attack remotely. The exploit is publicly available and might be used. This product takes the approach of rolling releases to provide continious delivery. Therefore, version details for affected and updated releases are not available.

    Disclosure Server

    5.5 MEDIUM 0.7% exploit odds No fix listed
  41. CVE-2025-30675

    In Apache CloudStack, a flaw in access control affects the listTemplates and listIsos APIs. A malicious Domain Admin or Resource Admin can exploit this issue by intentionally specifying the 'domainid' parameter along with the 'filter=self' or 'filter=selfexecutable' values. This allows the attacker to gain unauthorized visibility into templates and ISOs under the ROOT domain. A malicious admin can enumerate and extract metadata of templates and ISOs that belong to unrelated domains, violating isolation boundaries and potentially exposing sensitive or internal configuration details.  This vulnerability has been fixed by ensuring the domain resolution strictly adheres to the caller's scope rather than defaulting to the ROOT domain. Affected users are recommended to upgrade to Apache CloudStack 4.19.3.0 or 4.20.1.0.

    Disclosure Server Apache Cloudstack

    4.7 MEDIUM 0.7% exploit odds Fix listed
  42. CVE-2025-30086

    CNCF Harbor 2.13.x before 2.13.1 and 2.12.x before 2.12.4 allows information disclosure by administrators who can exploit an ORM Leak present in the /api/v2.0/users endpoint to leak users' password hash and salt values. The q URL parameter allows a user to filter users by any column, and filter password=~ could be abused to leak out a user's password hash character by character. An attacker with administrator access could exploit this to leak highly sensitive information stored in the Harbor database. All endpoints that support the q URL parameter are vulnerable to this ORM leak attack.

    Disclosure Server

    4.9 MEDIUM 0.7% exploit odds Fix listed
  43. CVE-2025-26521

    When an Apache CloudStack user-account creates a CKS-based Kubernetes cluster in a project, the API key and the secret key of the 'kubeadmin' user of the caller account are used to create the secret config in the CKS-based Kubernetes cluster. A member of the project who can access the CKS-based Kubernetes cluster, can also access the API key and secret key of the 'kubeadmin' user of the CKS cluster's creator's account. An attacker who's a member of the project can exploit this to impersonate and perform privileged actions that can result in complete compromise of the confidentiality, integrity, and availability of resources owned by the creator's account. CKS users are recommended to upgrade to version 4.19.3.0 or 4.20.1.0, which fixes this issue.Updating Existing Kubernetes Clusters in ProjectsA service account should be created for each project to provide limited access specifically for Kubernetes cluster providers and autoscaling. Follow the steps below to create a new service account, update the secret inside the cluster, and regenerate existing API and service keys:1. Create a New Service AccountCreate a new account using the role "Project Kubernetes Service Role" with the following details: Account Name kubeadmin-<FIRST_EIGHT_CHARACTERS_OF_PROJECT_ID> First Name Kubernetes Last Name Service User Account Type 0 (Normal User) Role ID <ID_OF_SERVICE_ROLE> 2. Add the Service Account to the ProjectAdd this account to the project where the Kubernetes cluster(s) are hosted. 3. Generate API and Secret KeysGenerate API Key and Secret Key for the default user of this account. 4. Update the CloudStack Secret in the Kubernetes ClusterCreate a temporary file `/tmp/cloud-config` with the following data:    api-url = <API_URL>     # For example: <MS_URL>/client/api   api-key = <SERVICE_USER_API_KEY>   secret-key = <SERVICE_USER_SECRET_KEY>   project-id = <PROJECT_ID> Delete the existing secret using kubectl and Kubernetes cluster config:    ./kubectl --kubeconfig kube.conf -n kube-system delete secret cloudstack-secret Create a new secret using kubectl and Kubernetes cluster config:     ./kubectl --kubeconfig kube.conf -n kube-system create secret generic cloudstack-secret --from-file=/tmp/cloud-config Remove the temporary file:     rm /tmp/cloud-config5. Regenerate API and Secret KeysRegenerate the API and secret keys for the original user account that was used to create the Kubernetes cluster.

    Disclosure Server Apache Cloudstack

    8.1 HIGH 0.7% exploit odds Fix listed
  44. CVE-2025-54450

    Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal') vulnerability in Samsung Electronics MagicINFO 9 Server allows Code Injection.This issue affects MagicINFO 9 Server: less than 21.1080.0.

    Disclosure Server Samsung Magicinfo 9 Server

    7.2 HIGH 0.7% exploit odds Fix listed
  45. CVE-2025-41714

    The upload endpoint insufficiently validates the 'Upload-Key' request header. By supplying path traversal sequences, an authenticated attacker can cause the server to create upload-related artifacts outside the intended storage location. In certain configurations this enables arbitrary file write and may be leveraged to achieve remote code execution.

    Disclosure Server

    8.8 HIGH 0.7% exploit odds No fix listed
  46. CVE-2025-58045

    Dataease is an open source data analytics and visualization platform. In Dataease versions up to 2.10.12, the patch introduced to mitigate DB2 JDBC deserialization remote code execution attacks only blacklisted the rmi parameter. The ldap parameter in the DB2 JDBC connection string was not filtered, allowing attackers to exploit the DB2 JDBC connection string to trigger server-side request forgery (SSRF). In higher versions of Java, ldap deserialization (autoDeserialize) is disabled by default, preventing remote code execution, but SSRF remains exploitable. Versions up to 2.10.12 are affected. The issue is fixed in version 2.10.13. Updating to 2.10.13 or later is recommended. No known workarounds are documented aside from upgrading.

    Disclosure Server Dataease

    7.1 HIGH 0.7% exploit odds Fix listed
  47. CVE-2024-12068

    A Server-Side Request Forgery (SSRF) vulnerability was discovered in haotian-liu/llava, affecting version git c121f04. This vulnerability allows an attacker to make the server perform HTTP requests to arbitrary URLs, potentially accessing sensitive data that is only accessible from the server, such as AWS metadata credentials.

    Disclosure Server Hliu Llava

    7.5 HIGH 0.7% exploit odds No fix listed
  48. CVE-2024-11449

    A vulnerability in haotian-liu/llava version 1.2.0 (LLaVA-1.6) allows for Server-Side Request Forgery (SSRF) through the /run/predict endpoint. An attacker can gain unauthorized access to internal networks or the AWS metadata endpoint by sending crafted requests that exploit insufficient validation of the path parameter. This flaw can lead to unauthorized network access, sensitive data exposure, and further exploitation within the network.

    Disclosure Server Hliu Large Language And Vision Assistant

    7.5 HIGH 0.7% exploit odds No fix listed
  49. CVE-2024-11603

    A Server-Side Request Forgery (SSRF) vulnerability exists in lm-sys/fastchat version 0.2.36. The vulnerability is present in the `/queue/join?` endpoint, where insufficient validation of the path parameter allows an attacker to send crafted requests. This can lead to unauthorized access to internal networks or the AWS metadata endpoint, potentially exposing sensitive data and compromising internal servers.

    Disclosure Server lm-sys Fastchat

    7.5 HIGH 0.7% exploit odds No fix listed
  50. CVE-2025-22130

    Soft Serve is a self-hostable Git server for the command line. Prior to 0.8.2 , a path traversal attack allows existing non-admin users to access and take over other user's repositories. A malicious user then can modify, delete, and arbitrarily repositories as if they were an admin user without explicitly giving them permissions. This is patched in v0.8.2.

    Disclosure Server Charm Soft Serve

    5.3 MEDIUM 0.7% exploit odds Fix listed

CVE records published in 2025 (ids starting CVE-2025-), sorted by how likely they are to be exploited.