| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| PraisonAI is a multi-agent teams system. Prior to 4.6.59, the CODE_TOOLS wrappers keep _workspace_root as None and pass workspace=None to read_file, search_replace, and apply_diff helpers that enforce path containment only for a truthy workspace. An application that exposes code_read_file, code_search_replace, or code_apply_diff before set_workspace can therefore let prompt-influenced calls read and modify files outside the intended project directory, while explicitly configured workspaces remain effective. This vulnerability is fixed in 4.6.59. |
| dbt-mcp is a Model Context Protocol server for interacting with dbt. Prior to 1.20.0, the local OAuth helper in src/dbt_mcp/oauth/fastapi_app.py exposes GET /dbt_platform_context without authentication or Host validation after a user completes the dbt Platform OAuth flow. The endpoint returns the full DbtPlatformContext, including access_token and refresh_token values persisted by the context manager, to any process that can reach 127.0.0.1:6785. The absence of TrustedHostMiddleware allows a remote attacker to use DNS rebinding against a victim's browser because the helper accepts arbitrary Host headers, while a co-located process can request the endpoint directly. The stolen tokens provide immediate dbt Platform API access as the victim and persistent access through the refresh token, allowing access to or modification of projects, jobs, environment secrets, and related account data. This issue is fixed in version 1.20.0. |
| langgraph-api implements the LangGraph API for rapid development and testing. Prior to 0.10.0, the langgraph-api run-creation path authorizes the assistant attached to a run by dispatching assistants.search with an incomplete value instead of the assistants.read event used by direct reads and cron creation. In deployments with custom resource handlers that register only assistants.read, omit an assistants.search handler, and have no global fallback handler, no applicable handler supplies an owner filter, allowing a low-privileged user to reference another user's private assistant through POST /runs or POST /threads/{thread_id}/runs. The run-creation response can disclose the private assistant's metadata, config, and context, and the run can execute using that assistant's configuration. Deployments without custom authorization handlers, or with an equivalent owner filter applied through a global handler or across all assistant events, are not affected. This issue is fixed in version 0.10.0. |
| WeasyPrint helps web developers to create PDF documents. Prior to 70.0, server-side applications that configure a restrictive url_fetcher and pass attacker-influenced values to HTML.write_pdf() can have the restriction bypassed through the xmp_metadata or stylesheets options. In weasyprint/pdf/init.py, xmp_metadata calls select_source() without the document url_fetcher, allowing an accessible local file to be read and embedded verbatim in the output PDF. In weasyprint/document.py, stylesheets constructs CSS() without the document url_fetcher, allowing local or internal resource loading and propagating the permissive fetcher through nested CSS imports and url() references. The stylesheets channel applies fetched resources but does not by itself disclose stylesheet comments verbatim. This issue is fixed in version 70.0. |
| SQLAdmin is a flexible Admin interface for SQLAlchemy models. Prior to 0.27.1, ModelView.sort_query in sqladmin/models.py accepts the attacker-controlled sortBy list-view query parameter without enforcing the configured column_sortable_list server-side allow-list in self._sort_fields. The value is resolved with getattr and passed to relationship joins and order_by, allowing requests to sort by columns hidden from column_list and by related-model columns through dotted paths. The resulting row order forms an information-exposure oracle for unexposed values, and reversing ascending and descending order confirms their relative ordering. Pairing sortBy with searchable or filterable columns and pagination can narrow the oracle toward specific values, but exact recovery depends on the application's available fields and data. This issue is fixed in version 0.27.1. |
| safeurl is a server-side request forgery protection library. Prior to 0.2.4, the privateNetworks list in ip.go omits the IPv6 ranges 64:ff9b:1::/48, 5f00::/16, 3fff::/20, and 100:0:0:1::/64. When an application enables IPv6 with EnableIPv6(true), an attacker-controlled destination in one of these ranges is not recognized as non-public and can pass the SSRF destination check, potentially allowing access to resources hosted within the omitted ranges. IPv6 is disabled by default, and configurations that retain EnableIPv6(false) are not exposed to this bypass. This issue is fixed in version 0.2.4. |
| node-opcua is an OPC UA implementation for TypeScript and Node.js. Prior to 2.166.0, the process-global g_alreadyUsedNonce cache used by nonceAlreadyBeenUsed in packages/node-opcua-secure-channel/source/server/server_secure_channel_layer.ts records nonces from OpenSecureChannelRequest and CreateSession without expiration or a size limit. An unauthenticated remote attacker can repeatedly create sessions with unique nonces, causing entries to persist after session expiry and accumulate across connection cycles even when maxSessions=10 limits concurrent sessions. The resulting unbounded heap growth can exhaust the default Node.js heap and crash the node-opcua server process. This issue is fixed in version 2.166.0. |
| node-opcua is an OPC UA implementation for TypeScript and Node.js. Prior to 2.166.0, the UserNameIdentityToken authentication handler in packages/node-opcua-server/source/opcua_server.ts decrypts an RSA-OAEP password blob but does not verify that the trailing bytes match the current session serverNonce. An unauthenticated remote attacker can obtain the server public key through GetEndpoints and forge a blob whose little-endian length produces an empty password passed to isValidUser, compromising accounts that accept an empty password. Missing nonce binding also allows a captured UserNameIdentityToken ciphertext to be replayed in another session, and SecurityMode=None removes the separate client-signature safeguard. This issue is fixed in version 2.166.0. |
| ContextForge is an AI gateway, registry, and proxy that provides centralized discovery, guardrails, and management for MCP, A2A, and REST or gRPC APIs. Prior to 1.0.3, the /admin/gateways/test call site in mcpgateway/admin.py calls validate_gateway_test_url() in mcpgateway/common/validators.py to resolve and reject private, loopback, link-local, and cloud-metadata addresses, but ResilientHttpClient later resolves the original hostname again without binding the validated address. When MCPGATEWAY_ADMIN_API_ENABLED is enabled, an attacker with a database-backed role containing explicit gateways.read permission can use DNS rebinding to return a public address during validation and a private or metadata address during connection, bypassing ssrf_blocked_networks and ssrf_dns_fail_closed because those controls apply only to the validation-time result. The endpoint's allow_admin_bypass=False setting means a bootstrap-only virtual platform-admin identity without a database role is not sufficient. Successful exploitation can reach internal services and cloud metadata, expose cloud credentials, access internal APIs, or probe internal network ports. This issue is fixed in version 1.0.3. |
| SQL Injection vulnerability in Woltlab WCF v.6.2.4 and before allows a remote attacker to updateUserOptions in UserEditor.class.php and the update action in UserAction.class.php |
| FFmpeg before 9.0 has an out-of-bounds read because there is insufficiently padded extradata in the MOV parsing path in mov_read_iacb in libavformat/mov.c. |
| The OpenTelemetry Operator is a Kubernetes Operator for the OpenTelemetry Collector. Prior to 0.152.0, cmd/otel-allocator TargetAllocator instances with targetAllocator.prometheusCR.enabled set to true preserve a selected ServiceMonitor endpoint's bearerTokenFile value as HTTPClientConfig.Authorization.CredentialsFile. A tenant who can create or update a ServiceMonitor matched by serviceMonitorSelector and serviceMonitorNamespaceSelector can point bearerTokenFile at a file in the Collector pod, including /var/run/secrets/kubernetes.io/serviceaccount/token, and direct scraping to a tenant-controlled endpoint. The Collector reads that file at scrape time and sends its contents as bearer authorization on every scrape interval. Exploitation also requires the Collector service-account token or another sensitive file to be mounted and the Collector to reach the chosen target. The DenyFSAccessThroughSMs control was absent, allowing disclosure of the Collector's service-account JWT or other mounted files, and resulting Kubernetes API impact is limited by the Collector service account's permissions. This issue is fixed in version 0.152.0. |
| In Oraios AI Serena before 1.0.0, the listen address of the MCP server in HTTP mode is 0.0.0.0. NOTE: the Supplier observed that 0.0.0.0 was a "potential security hazard" but the Serena documentation, at the time of the issue report proposing 127.0.0.1 instead of 0.0.0.0, recommended "use a sandboxed environment for running Serena." |
| CrewAI before fb2323b offers a Python blocklist approach that operates at the wrong level of abstraction, a different vulnerability than CVE-2026-2275. Import-time blocking of module names does not address the availability of Python's complete object graph. For example, calling ctypes.CDLL(None) loads the C library without relying in any import statements. In other words, a within-process sandbox cannot merely account for the import system and instead must account for the complete runtime of the Python interpreter. |
| An issue was discovered in camera in Samsung Mobile Processor Exynos 1330, 1380, 1480, 2400, 1580, 2500, 2600, and 1680. In the camera driver, an out-of-bounds array access vulnerability in the error-handling path leads to memory corruption. |
| An issue was discovered in camera in Samsung Mobile Processor Exynos 1580 and 2500. An untrusted pointer dereference occurs when a malformed message is sent to the camera driver, causing limited information disclosure or denial of service. |
| An issue was discovered in Samsung Mobile Processor and Wearable Processor Exynos 1330, 1380, 1480, 1580, 1680, W920, W930, and W1000. . A malformed ioctl command to the Wi-Fi interface device can lead to improper buffer size allocation, resulting in an out-of-bounds write and causing a denial of service (DoS). |
| CyberPanel before 2.4.4 omits a "return 0" that is required by the business logic. |
| In Mbed TLS 3.6.x before 3.6.7 and 4.1.x before 4.1.2, the TLS 1.3 client accepts HelloRetryRequest selecting an unadvertised group. |
| An issue was discovered in NR RRC in Samsung Mobile Processor and Modem Exynos 1080, 2100, 1280, 2200, 1330, 1380, 1480, 2400, 1580, 2500, W1000, Modem 5300, Modem 5400, and Modem 5410. Incorrect handling of unauthenticated downlink RRC Setup messages can cause the baseband to crash. |
