containerd, Input Validation Error, CVE-2026-46680 (High)
The vulnerability CVE-2026-46680 arises from an input validation error in containerd when processing the `User` directive for container images. Specifically,
The vulnerability CVE-2026-46680 arises from an input validation error in containerd when processing the `User` directive for container images. Specifically,
How the mentioned CVE works (around 20 lines): `Compiler::string()` escapes characters “, $, \, NUL and TAB when generating PHP
The vulnerability, tracked as CVE-2026-46550, arises from the insecure implementation of the refresh-token cookie handling in NocoDB versions prior to
The vulnerability arises from a discrepancy between the tool discovery and execution layers. Three environment variables (ALLOW_ONLY_READONLY_TOOLS, ALLOW_ONLY_NON_DESTRUCTIVE_TOOLS, ALLOWED_TOOLS) are
The vulnerability exists because the confirmation dialog in the Umbraco backoffice fails to properly encode user-supplied input before rendering it.
The OAuth token strategy in NocoDB attaches `oauth_scope` and `oauth_granted_resources` to the request user, but the ACL middleware never consults
The vulnerability stems from how @hulumi/policies versions prior to 1.3.2 processed evidence for Cloudflare and deployment-governance guardrails. The system used
The vulnerability exists because `pkg/builder/builder.go` passes `Environment.spec.builder.command` directly to `exec.Command` after a simple `strings.Fields` split. No validation is performed on
How the CVE works: The vulnerable code attempts to shell-escape the binary path with `escapeshellarg(‘/usr/bin/wkhtmltopdf’)` before checking is_executable(). On POSIX,
The vulnerability arises when developers pass unsanitized user input into the `xsl-style-sheet` option of the `Knp\Snappy\Pdf` generator. The library uses
Fission runtime pods are created with ServiceAccountName: fission-fetcher, and the `fission-fetcher` ServiceAccount is granted namespace-wide `get` access to `secrets` and
The vulnerability in `@hulumi/policies` versions prior to 1.3.2 stems from an improper validation of “parent evidence” for the `SecureBucket` resource
The vulnerability CVE-2026-46552 is an authorization flaw in NocoDB’s shared-base functionality. When a user creates a shared-base link (e.g., a
The vulnerability arises because the upload-by-URL feature fails to enforce the `NC_ATTACHMENT_FIELD_SIZE` limit. This limitation is checked against neither the
How the CVE works (vulnerability mechanics): Fission router automatically registers internal route `/fission-function/` and `/fission-function//` for every Function object. This
How the CVE works: The vulnerability stems from unsafe growth handling in CryptoVec, a buffer type used in Russh SSH
The `uploadViaURL` path in the v1/v2 attachment API did not enforce `NC_ATTACHMENT_FIELD_SIZE` against the remote `content-length` or against the response
How GHSA‑gfp8‑mp24‑5vxg works The `@hulumi/baseline` package before 1.3.2 does not fully catch AWS CloudTrail event‑selector tampering. Attackers can call the
The vulnerability in @hulumi/policies versions before 1.3.2 stems from incomplete inspection of inline and attached IAM policy evidence for the
The `@hulumi/drift` package before version 1.3.2 contained a flaw in its orphan resource reconciliation logic. An “orphan reconciler” is a
How the vulnerability works NocoDB uses an in‑memory or Redis‑based authentication cache to speed up repeated API token lookups. When
How CVE-2026-46543 Works: The vulnerability resides in how the Nimiq blockchain node handles a peer-to-peer message of type RequestBatchSet. A
The vulnerability resides in the `BlockInclusionProof::is_block_proven` function. Under normal operation, this function performs a cryptographic verification of a block’s inclusion
How the mentioned CVE works: The SSRF mitigation added in commit 33c55da for GHSA-7gvf-3w72-p2pg was incomplete. The `parse_urls` API validates
How the CVE works: LMDeploy hardcodes `trust_remote_code=True` in Hugging Face `from_pretrained` calls, overriding Transformers’ default secure False. When a user
The vulnerability arises because the `Umbraco.Cms` package fails to validate certain redirect URLs, specifically within specific Surface Controllers (such as
The root cause of CVE-2026-46542 is an improper error handling pattern in the Ed25519 multisig delinearization code path. The `Ed25519PublicKey::delinearize()`
How CVE-2026-34926 Works This directory traversal vulnerability exists in the on-premise Apex One server due to improper validation of file
1. The vulnerability exists in the `MerkleRadixTrie::put_chunk` function. An unauthenticated remote attacker can send a malicious state-sync chunk. The chunk
The `fileID` field from `Manifest.db` (a SQLite database inside iOS backups) is used directly in filesystem path construction without validation.
The CVE-2026-8597 vulnerability affects the Amazon SageMaker Python SDK, specifically when deploying machine learning models with the Triton inference server
How the mentioned CVE works: The Amazon SageMaker Python SDK versions 2.199.0 through 2.257.1 and 3.0.0 through 3.7.1 contain a
How the mentioned CVE works The vulnerability stems from unsafe Markdown-to-HTML conversion. The application uses `markdown-it` with `html: true` in
How the CVE Works The `query.batch` function, introduced in SvelteKit 2.38.0, is a performance feature designed to solve the “N+1
The vulnerability arises because lmdeploy unconditionally sets `trust_remote_code=True` when calling HuggingFace Transformers APIs. An attacker who controls the `model_path` parameter
The vulnerability, identified as CVE-2025-47949, is a classic Signature Wrapping attack affecting the samlify Node.js library. While samlify correctly verifies
How the vulnerability works During acquisition, `getPathToLocalCopy()` builds local paths for downloaded APKs. It calls `extractFileName()` to get a filename
The vulnerability exists in OpenMetadata 1.12.1 when using SSO (e.g., Azure AD) and the database secrets manager (secretsManagerProvider: “db”). A
How the mentioned CVE works (around 20 lines) CVE-2026-41091 is a high-severity (CVSS 7.8) local privilege escalation vulnerability in Microsoft
How the mentioned CVE works: The vulnerability resides in the PDF parsing logic of Adobe Reader and Acrobat, specifically within
How the mentioned CVE works (technical details): CVE-2010-0249 is a use-after-free vulnerability in Internet Explorer’s handling of HTML objects. When
How the mentioned CVE works: The vulnerability resides in the Server service (srvsvc) of Microsoft Windows, specifically within the NetPathCanonicalize
The vulnerability exists in quartz.dll, the DirectShow QuickTime Movie Parser Filter, part of DirectX 7.0 through 9.0c. When processing a
CVE-2026-45498 is a Medium‑severity Denial‑of‑Service vulnerability in the Microsoft Defender Antimalware Platform. The root cause lies in insufficient input validation
The vulnerability exists in iepeers.dll, the Peer Objects component of Internet Explorer 6, 6 SP1, and 7. When IE processes
The vulnerability exists in the password reset API endpoint at /api/user/password/update, which can be triggered without any authentication. The code
The vulnerability arises from a default empty API client token combined with a flawed authentication check. During installation, phpMyFAQ seeds
The vulnerability exists in phpMyFAQ’s password reset API (UnauthorizedUserController.php, lines 56‑130). The `updatePassword()` method accepts a PUT request with only
The CVE works because the `/text-to-speech` generation endpoint hardcodes `Access-Control-Allow-Origin: ` and `Access-Control-Allow-Headers: Cache-Control` in the response, completely ignoring the
The vulnerability exists in the PUT /api/v1/user endpoint of Flowise. An authenticated user sends a request to update their profile.
How the CVE Works The vulnerability exists in the `overwritePassword()` method of phpMyFAQ’s Admin API. The endpoint is accessible at
How the mentioned CVE works (approximately 20 lines): The vulnerability exists in Composer’s github-oauth authentication validation logic. It affects GitHub
The vulnerability exists in Algernon’s SSE event server where `Access-Control-Allow-Origin` is hardcoded to . In `engine/config.go`, function `recwatch.EventServer` receives literal
The attack exploits the npm package registry’s trust model by publishing malicious minor/patch versions of three widely used CAP (Cloud
How the CVE works (technical details): RTK versions below 0.32.0 load `.rtk/filters.toml` automatically from the current working directory. This file
How the mentioned CVE works: The vulnerability exists in `shivammathur/setup-php` GitHub Action when it resolves the PHP version from repository-controlled
CVE-2026-33814 resides in the Go programming language’s HTTP/2 transport implementation (golang.org/x/net/http2). The flaw triggers an infinite loop when a client
How the mentioned CVE works The vulnerability stems from a platform-dependent default host value in engine/flags.go:39-46. On non-Windows systems (Linux/macOS),
All FHIRPathEngine implementations accept arbitrary FHIRPath expressions without validation. The matches(), matchesFull(), and `replaceMatches()` functions pass user‑controlled regex patterns directly
How CVE-2026-40310 Works The vulnerability resides within the JPEG 2000 (JP2) encoder of ImageMagick. The core issue is triggered when
Two FastAPI routes in NiceGUI—the per‑component resource route (added in v1.4.6) and the ESM module route (added in v3.0.0)—accept a
The vulnerability resides in OBI’s log enricher component, specifically within the `__fill_iov` function in `bpf/logenricher/logenricher.c` at line 50. For `ITER_IOVEC`
The CVE-2026-33902 vulnerability is a stack overflow issue located in ImageMagick’s FX expression parser. The parser fails to implement any
The vulnerability resides in the MIFF (Magick Image File Format) encoder when LZMA compression is selected. During the compression process,
How the CVE Works The vulnerability exists in the MNG (Multiple-image Network Graphics) coder of ImageMagick. The coder fails to
A heap-use-after-free vulnerability exists in ImageMagick when processing crafted MSL (Magick Scripting Language) scripts. The issue resides in the MSLStartElement
When an attacker provides specific arguments to ImageMagick’s polynomial distortion operation, the software incorrectly calculates the memory boundaries. This leads
How the mentioned CVE works (approx. 20 lines): The vulnerability exists in OBI’s memcached text protocol parser. When handling storage
How the mentioned CVE works (technical details): The vulnerability exists in NiceGUI’s `ui.restructured_text()` element, which renders user‑supplied reStructuredText (reST) server‑side
How the mentioned CVE works: The vulnerability resides in go.opentelemetry.io/obi versions v0.1.0 through v0.8.0. The MongoDB TCP parser processes raw
The attack leverages a compromised developer device to publish malicious npm package versions of @mistralai/mistralai, @mistralai/mistralai-azure, and @mistralai/mistralai-gcp. An automated
The Java TLS ioctl probe in OBI versions up to v0.0.0-rc.1+build reads a user‑controlled ioctl pointer using `bpf_probe_read` instead of
How the CVE works The vulnerability exists in view/update.php, which processes a POST request containing the parameter updateFile. This parameter
The CVE-2026-45681 vulnerability arises from a size mismatch in the per-CPU message-buffer fallback path within OpenTelemetry eBPF Instrumentation (OBI). The
The vulnerability exists in the `WebSocket.close()` method when a `TypedArray` (e.g., Float32Array, Uint8Array) is passed as the `reason` argument instead
The vulnerability is a supply chain attack on the `mistralai` PyPI package version 2.4.6. No legitimate tag, commit, or release
The vulnerability arises because OBI’s lightweight ELF parser (fastelf) trusts data from a potentially malicious executable without performing critical validation
How CVE-2026-42899 Works CVE-2026-42899 is a high‑severity denial‑of‑service vulnerability affecting .NET 8.0, 9.0, and 10.0. The flaw resides in ASP.NET
How this CVE Works The vulnerability exists in OBI’s Redis error-handling pipeline. In the `pkg/ebpf/common/redis_detect_transform.go` file, the `getRedisError` function trims
The vulnerability exists because the .NET `ZipFile.ExtractToDirectory` API does not validate path traversal sequences (e.g., `..\` or /) inside ZIP
This vulnerability in .NET 8.0, 9.0, and 10.0 on Windows stems from improper input validation, which leads to a heap-based
The vulnerability exists in TinyIce’s WebRTC source-ingest HTTP endpoint, POST /webrtc/source-offer?mount=. The handler `handleWebRTCSourceOffer` was registered at that path without
The vulnerability exists in Budibase V1 Views API endpoint POST /api/views. A parameter called “calculation” from the request body is
Technical details of how the vulnerability works: The multiparty Node.js library (versions ≤4.2.3) parses multipart/form-data requests. It builds field values
How the vulnerability works (technical details): The multiparty library parses `multipart/form-data` requests, including the `Content-Disposition` header’s `filename` parameter (used for
How the Vulnerability Works Formie is a popular Craft CMS plugin for building forms. The vulnerability arises from the way
How the CVE works (technical) When `docker cp` copies data into a container, the daemon builds a temporary filesystem view.
How the mentioned CVE works: The vulnerability exists in dynoxide’s MCP HTTP transport (enabled via `dynoxide mcp –http` or dynoxide
How the mentioned CVE works: The vulnerability exists in the `meta` encoder of ImageMagick due to an off-by-one error. When
The vulnerability exists in the `POST /api/global/users/onboard` endpoint, which is incorrectly protected by `workspaceBuilderOrAdmin` middleware instead of adminRoutes. On self-hosted
The vulnerability resides in the `/admin/api/users` endpoint. When an authenticated user makes a `GET` request to list or view a
The vulnerability exists in the `cloakserve` CDP (Chrome DevTools Protocol) multiplexer component of CloakBrowser. The `cloakserve` service binds to `0.0.0.0`
The `setValueAtPath()` function in `@tmlmobilidade/utils` fails to sanitize user-controlled path segments before assigning a value to an object property. This
When `ENABLE_MULTI_TENANT=true` is set, the HTTP transport documentation states that the target n8n instance is selected per-request using the `x-n8n-url`
How the vulnerability works: The row action trigger endpoint POST /api/tables/:sourceId/actions/:actionId/trigger fails to validate that the user-supplied rowId is within
How the mentioned CVE works (around 20 lines): The vulnerability resides in `Postgrex.Notifications.listen/3` and unlisten/3. These functions interpolate the `channel`
The vulnerability occurs in the `ConnectedComponentsImage()` function when processing a user-supplied invalid `connected-components:keep-top` define value. During connected components analysis, this
The vulnerability arises from improper sanitization of the `rancid_repo_url` configuration value within the RANCID Integration settings of LibreNMS. The application
How the CVE works The vulnerability exists in `Redirect30xInterceptor.java` when following a cross-origin redirect or an HTTPS-to‑HTTP downgrade. The client
How the mentioned CVE works: In default Dozzle deploy with no auth provider set, POST /api/notifications/test-webhook is exposed pre-auth. The