NVD や CVE、ほか複数の脅威フィードを束ね、RCE など高リスクな事象を深く追える一覧です。CVSS と EPSS を組み合わせ、Exploit 参照や PoC の有無から悪用しやすさを追跡します。ベンダー修正や緩和策の文脈とあわせて優先度を決め、対応サイクルを短く保ちつつ重要資産を守る支援をします。
Assigner（CNA／発行元）：[email protected] この条件を外す
| CVE | 説明 | CVSS 最大値 | EPSS（%） | 公開 | 更新 |
|---|---|---|---|---|---|
| CVE-2026-4372 | A critical remote code execution vulnerability exists in all versions of the HuggingFace transformers library prior to version 5.3.0. The vulnerability allows an attacker to craft a malicious `config.json` file containing the `_attn_implementation_internal` field set to an attacker-controlled HuggingFace Hub repository ID. When a victim loads this model using the standard `AutoModelForCausalLM.from_pretrained()` API, the library downloads and executes arbitrary Python code from the attacker's re | 7.8 | 0.48% | 2026-05-24 | 2026-06-17 |
| CVE-2026-3515 | A vulnerability in the `GitHubRepository` block of the `prefect-github` integration in Prefect version 3.6.18 allows an attacker to inject arbitrary git command-line options via the `reference` field. The `reference` field is concatenated directly into a `git clone` command string without proper sanitization, and then parsed by `shlex.split()`. This enables injection of options such as `-c`, leading to potential Server-Side Request Forgery (SSRF), credential theft, or remote code execution (RCE) | 8.5 | 0.30% | 2026-05-24 | 2026-06-17 |
| CVE-2026-2734 | In mlflow/mlflow versions up to 3.9.0, the `SearchModelVersions` REST API endpoint and the `mlflowSearchModelVersions` GraphQL query lack proper per-model authorization checks when basic authentication is enabled. This allows any authenticated user to enumerate all model versions across all registered models, regardless of their permission level. The issue arises due to the absence of `SearchModelVersions` in the `BEFORE_REQUEST_VALIDATORS` and `AFTER_REQUEST_HANDLERS` for the REST API, and its | 6.5 | 0.40% | 2026-05-21 | 2026-06-17 |
| CVE-2026-2611 | In MLflow version 3.9.0, the MLflow Assistant feature introduced improper origin validation in its /ajax-api endpoints. This vulnerability allows a remote attacker to exploit cross-origin requests from a malicious webpage to interact with the MLflow Assistant running on a victim's local machine. By bypassing the loopback-only restriction, the attacker can modify the Assistant's configuration to enable full access, which in turn allows the execution of arbitrary commands via the Claude Code sub-a | 9.6 | 0.37% | 2026-05-19 | 2026-06-27 |
| CVE-2026-4137 | In mlflow/mlflow versions prior to 3.11.0, the `get_or_create_nfs_tmp_dir()` function in `mlflow/utils/file_utils.py` creates temporary directories with world-writable permissions (0o777), and the `_create_model_downloading_tmp_dir()` function in `mlflow/pyfunc/__init__.py` creates directories with group-writable permissions (0o770). These insecure permissions allow local attackers to tamper with model artifacts, such as cloudpickle-serialized Python objects, and achieve arbitrary code execution | 7.8 | 0.19% | 2026-05-18 | 2026-06-17 |
| CVE-2026-2652 | A vulnerability in mlflow/mlflow versions 3.9.0 and earlier allows unauthenticated access to certain FastAPI routes when the server is started with authentication enabled (`--app-name basic-auth`) and served via uvicorn (ASGI). The FastAPI permission middleware only enforces authentication on `/gateway/` routes, leaving other routes such as the Job API (`/ajax-api/3.0/jobs/*`) and the OpenTelemetry trace ingestion API (`/v1/traces`) unprotected. This allows unauthenticated remote attackers to su | 8.6 | 1.50% | 2026-05-14 | 2026-06-17 |
| CVE-2026-2614 | A vulnerability in the `_create_model_version()` handler of `mlflow/server/handlers.py` in mlflow/mlflow versions 3.9.0 and earlier allows an unauthenticated remote attacker to read arbitrary files from the server's filesystem. The issue arises when a `CreateModelVersion` request includes the tag `mlflow.prompt.is_prompt`, which bypasses source path validation. This enables an attacker to store an arbitrary local filesystem path as the model version source. The `get_model_version_artifact_handle | 7.5 | 0.74% | 2026-05-11 | 2026-07-02 |
| CVE-2026-2393 | A Server-Side Request Forgery (SSRF) vulnerability exists in MLflow versions prior to 3.9.0. The `_create_webhook()` function in `mlflow/server/handlers.py` accepts a user-controlled `url` parameter without validation, and the `_send_webhook_request()` function in `mlflow/webhooks/delivery.py` sends HTTP POST requests to this attacker-controlled URL. This allows an authenticated attacker to force the MLflow backend to send HTTP requests to internal services, cloud metadata endpoints, or arbitrar | 7.1 | 0.29% | 2026-05-11 | 2026-06-17 |
| CVE-2026-3960 | 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 vulnerability arises due to insufficient security controls in the parameter blacklist mechanism, which only targets MySQL JDBC driver-specific dangerous parameters. An attacker can bypass these controls by switching the JDBC URL protocol to jdbc:postgresql: and exploiting PostgreSQL JDBC driver-specific parameters such as socketFactory and soc | 9.8 | 0.94% | 2026-04-23 | 2026-06-17 |
| CVE-2026-1462 | A vulnerability in the `TFSMLayer` class of the `keras` package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of `.keras` models, even when `safe_mode=True`. This bypasses the security guarantees of `safe_mode` and enables arbitrary attacker-controlled code execution during model inference under the victim's privileges. The issue arises due to the unconditional loading of external SavedModels, serialization of attacker-controlled file path | 7.8 | 0.33% | 2026-04-13 | 2026-06-29 |
| CVE-2026-1116 | A Cross-site Scripting (XSS) vulnerability was identified in the `from_dict` method of the `AppLollmsMessage` class in parisneo/lollms prior to version 2.2.0. The vulnerability arises from the lack of sanitization or HTML encoding of the `content` field when deserializing user-provided data. This allows an attacker to inject malicious HTML or JavaScript payloads, which can be executed in the context of another user's browser. Exploitation of this vulnerability can lead to account takeover, sessi | 6.1 | 0.26% | 2026-04-11 | 2026-06-17 |
| CVE-2026-1115 | A Stored Cross-Site Scripting (XSS) vulnerability was identified in the social feature of parisneo/lollms, affecting the latest version prior to 2.2.0. The vulnerability exists in the `create_post` function within `backend/routers/social/__init__.py`, where user-provided content is directly assigned to the `DBPost` model without sanitization. This allows attackers to inject and store malicious JavaScript, which is executed in the browsers of users viewing the Home Feed, including administrators. | 9.6 | 0.40% | 2026-04-10 | 2026-06-17 |
| CVE-2026-1163 | An insufficient session expiration vulnerability exists in the latest version of parisneo/lollms. The application fails to invalidate active sessions after a password reset, allowing an attacker to continue using an old session token. This issue arises due to the absence of logic to reject requests after a period of inactivity and the excessively long default session duration of 31 days. The vulnerability enables an attacker to maintain persistent access to a compromised account, even after the | 4.1 | 0.21% | 2026-04-07 | 2026-06-17 |
| CVE-2026-5627 | A path traversal vulnerability exists in mintplex-labs/anything-llm versions up to and including 1.9.1, within the `AgentFlows` component. The vulnerability arises from improper handling of user input in the `loadFlow` and `deleteFlow` methods in `server/utils/agentFlows/index.js`. Specifically, the combination of `path.join` and `normalizePath` allows attackers to bypass directory restrictions and access or delete arbitrary `.json` files on the server. This can lead to information disclosure, s | 7.2 | 0.81% | 2026-04-07 | 2026-06-17 |
| CVE-2026-1114 | In parisneo/lollms version 2.1.0, the application's session management is vulnerable to improper access control due to the use of a weak secret key for signing JSON Web Tokens (JWT). This vulnerability allows an attacker to perform an offline brute-force attack to recover the secret key. Once the secret key is obtained, the attacker can forge administrative tokens by modifying the JWT payload and resigning it with the cracked secret. This enables unauthorized users to escalate privileges, impers | 9.8 | 0.54% | 2026-04-07 | 2026-06-17 |
| CVE-2026-1839 | A vulnerability in the HuggingFace Transformers library, specifically in the `Trainer` class, allows for arbitrary code execution. The `_load_rng_state()` method in `src/transformers/trainer.py` at line 3059 calls `torch.load()` without the `weights_only=True` parameter. This issue affects all versions of the library supporting `torch>=2.2` when used with PyTorch versions below 2.6, as the `safe_globals()` context manager provides no protection in these versions. An attacker can exploit this vul | 7.8 | 0.30% | 2026-04-07 | 2026-06-17 |
| CVE-2026-0545 | In mlflow/mlflow, the FastAPI job endpoints under `/ajax-api/3.0/jobs/*` are not protected by authentication or authorization when the `basic-auth` app is enabled. This vulnerability affects the latest version of the repository. If job execution is enabled (`MLFLOW_SERVER_ENABLE_JOB_EXECUTION=true`) and any job function is allowlisted, any network client can submit, read, search, and cancel jobs without credentials, bypassing basic-auth entirely. This can lead to unauthenticated remote code exec | 9.8 | 4.39% | 2026-04-03 | 2026-06-29 |
| CVE-2026-0596 | A command injection vulnerability exists in mlflow/mlflow when serving a model with `enable_mlserver=True`. The `model_uri` is embedded directly into a shell command executed via `bash -c` without proper sanitization. If the `model_uri` contains shell metacharacters, such as `$()` or backticks, it allows for command substitution and execution of attacker-controlled commands. This vulnerability affects the latest version of mlflow/mlflow and can lead to privilege escalation if a higher-privileged | 7.8 | 1.33% | 2026-03-31 | 2026-06-17 |
| CVE-2025-15379 | A command injection vulnerability exists in MLflow's model serving container initialization code, specifically in the `_install_model_dependencies_to_env()` function. When deploying a model with `env_manager=LOCAL`, MLflow reads dependency specifications from the model artifact's `python_env.yaml` file and directly interpolates them into a shell command without sanitization. This allows an attacker to supply a malicious model artifact and achieve arbitrary command execution on systems that deplo | 9.8 | 1.99% | 2026-03-30 | 2026-06-29 |
| CVE-2025-15036 | A path traversal vulnerability exists in the `extract_archive_to_dir` function within the `mlflow/pyfunc/dbconnect_artifact_cache.py` file of the mlflow/mlflow repository. This vulnerability, present in versions before v3.7.0, arises due to the lack of validation of tar member paths during extraction. An attacker with control over the tar.gz file can exploit this issue to overwrite arbitrary files or gain elevated privileges, potentially escaping the sandbox directory in multi-tenant or shared c | 10.0 | 0.59% | 2026-03-29 | 2026-06-29 |