🚨 CVE-2026-50881
Incorrect access control in the impworks Bonsai v6.0 allows authenticated attackers with Editor privileges to escalate privileges to Administrator and execute unauthorized account, password, and configuration changes.

🎖@cveNotify

🚨 CVE-2026-50882
An issue in the /api/v0/pastes endpoint of anna-is-cute paste v0.1.1 allows attackers to cause a Denial of Service (DoS) via a crafted POST request.

🎖@cveNotify

🚨 CVE-2026-50883
An HTML injection vulnerability in the /src/highlight.rs component of matze wastebin v3.4.1 allows attackers to execute arbitrary scripts via a crafted payload.

🎖@cveNotify

🚨 CVE-2026-50884
Incorrect access control in statping-ng v0.93.0 allows attackers to escalate privileges to Administrator and access sensitive components.

🎖@cveNotify

🚨 CVE-2026-50885
Incorrect access control in the share-based read endpoints of Sismics Docs (Teedy) v1.11 allow unauthorized attackers to access sensitive endpoints via a crafted request.

🎖@cveNotify

🚨 CVE-2026-50886
Incorrect access control in the webhook management component of Project Firefly III v6.5.9 allows attackers to scan internal resources via a crafted POST request.

🎖@cveNotify

🚨 CVE-2026-50887
A Server-Side Request Forgery (SSRF) in the automatic short URL title resolution component of shlink v5.0.1 allows attackers to scan internal resources via supplying a crafted longUrl.

🎖@cveNotify

🚨 CVE-2026-50888
An authenticated Server-Side Request Forgery (SSRF) in the custom scraper subsystem component of Benjamin Jonard Koillection v1.8.0 allows attackers to scan internal resources via supplying a crafted URL.

🎖@cveNotify

🚨 CVE-2026-50889
An input handling flaw in the HTTP refresh token process of LLDAP v0.6.2 allows attackers to cause a Denial of Service (DoS) via sending a crafted refresh-token header.

🎖@cveNotify

🚨 CVE-2026-50891
Incorrect access control in the /admin/api/config component of Filestash v0.4.0 allows attackers to escalate privileges via sending a crafted request.

🎖@cveNotify

🚨 CVE-2026-50892
Incorrect access control in the "Let's Encrypt" certificate download endpoint of Nginx Proxy Manager v2.14.0 allows authenticated attackers to obtain the TLS private key material via a crafted GET request.

🎖@cveNotify

🚨 CVE-2026-47261
Wasmtime is a runtime for WebAssembly. In versions prior to 24.0.9, 36.0.10, and 44.0.2, when a filesystem preopen is given DirPerms::all() and FilePerms::READ without FilePerms::WRITE, this access control mechanism can be bypassed via the wasip2 descriptor.open-at or wasip1 path_open interfaces by opening a file with only the OpenFlags::TRUNCATE oflag. The root cause is that the clause handling OpenFlags::TRUNCATE in crates/wasi/src/filesystem.rs (Dir::open_at, lines 967–969) did not set open_mode |= OpenMode::WRITE;, which is later used for the access control check against FilePerms to determine whether opening the file is permitted; the single-line fix adds that missing assignment, after which the affected calls correctly fail with error-code.not-permitted and ERRNO_PERM respectively. Only wasmtime-wasi embeddings that combine DirPerms::MUTATE with FilePerms::READ are affected by this bug. In particular, the Wasmtime project's wasmtime-cli's use of wasmtime-wasi is not affected, because it always sets FilePerms::all() for all preopens. This issue has been fixed in versions 24.0.9, 36.0.10 and44.0.2.

🎖@cveNotify

🚨 CVE-2026-48124
Cursor is a code editor built for programming with AI. In versions prior to 3.0.0, the Cursor Desktop could execute workspace-defined Claude hook commands from .claude/settings.local.json without dedicated user approval. A malicious workspace or agent-created file could configure hooks that run local commands in the user's context when an agent turn ends. This could allow sandbox escape, persistence across turns, local data access, or follow-on compromise. This issue has been fixed in version 3.0.0.

🎖@cveNotify

🚨 CVE-2026-48518
MultiJuicer is used to run separate Juice Shop instances on a central kubernetes cluster without the need for local instances. In versions 8.0.0 through 10.0.0, the team join endpoint (POST /multi-juicer/api/teams/{team}/join) accepted requests with any Content-Type, including text/plain. Because that content type does not trigger a CORS preflight, an attacker could host a cross-site HTML form that auto-submits to the endpoint and forces a victim's browser to log in as the attacker's team. A successful, undetected attacker can cause victims to unwittingly solve Juice Shop challenges under the attacker's team identity. In a CTF context this lets the attacker inflate their team's score using other players' activity, and any sensitive data the victim enters into "their" Juice Shop ends up in the attacker's instance. The vulnerability is exploitable without any prior authentication; the victim
only needs to visit a page the attacker controls while having network access to the MultiJuicer deployment. SameSite=Strict on the session cookie does not mitigate this, because the attack plants a new cookie rather than relying on an existing one. This issue was fixed in version 10.0.1.

🎖@cveNotify

🚨 CVE-2026-48708
OliveTin gives access to predefined shell commands from a web interface. In versions 3000.0.0 and prior, the template engine uses a single shared text/template.Template instance (tpl package-level variable in service/internal/tpl/templates.go) across all goroutines. Every action execution calls tpl.Parse(source) followed by t.Execute() on this shared instance with no synchronization. When two or more actions execute concurrently (which is the normal case — each ExecRequest spawns a goroutine), a race condition occurs: one goroutine's Parse overwrites the template tree while another goroutine is calling Execute, causing cross-user command contamination, Go runtime panic, and incorrect command execution. This issue has been resolved in version 3000.13.0.

🎖@cveNotify

🚨 CVE-2026-48709
OliveTin gives access to predefined shell commands from a web interface. In versions 3000.0.0 and prior, The ValidateArgumentType RPC endpoint in service/internal/api/api.go does not perform any authentication or authorization checks. Unlike all other data-returning API endpoints, it does not call auth.UserFromApiCall or checkDashboardAccess. When AuthRequireGuestsToLogin is enabled (the security-conscious configuration), this endpoint remains accessible to unauthenticated users and can be used as an oracle to enumerate valid action binding IDs and their argument configurations. This issue has been fixed in version 3000.13.0.

🎖@cveNotify

🚨 CVE-2026-48017
DbGate is cross-platform database manager. In versions 7.1.8 and prior, the POST /runners/load-reader endpoint in DbGate accepts a functionName parameter that is directly interpolated into a JavaScript code template without any sanitization or validation. An authenticated user (with basic access, no special permissions required) can inject arbitrary JavaScript code that executes on the server with full process privileges, bypassing the require=null sandbox restriction. An authenticated user with basic access (no admin role, no run-shell-script permission required) can: execute arbitrary OS commands on the DbGate server with the privileges of the Node.js process, read/write any file accessible to the process, pivot to connected databases by reading connection credentials from DbGate's storage, and compromise the host system - in Docker deployments, this typically means root access within the container.

🎖@cveNotify

🚨 CVE-2026-48157
Slim is a PHP micro framework that enables users to write simple web applications and APIs. In versions 4.4.0 through 4.15, if an application uses HttpException::setTitle() and/or setDescription() to include untrusted/request-derived data in the error title or description (e.g. "No products found matching '{$query}'."), an attacker could inject arbitrary HTML/JavaScript that executes in the victim's browser when they encounter an HTML error page generated by Slim. The vulnerability is present even with displayErrorDetails = false as the unescaped title and description are rendered on this error path. Built-in exceptions (HttpNotFoundException, HttpBadRequestException, etc.) ship plain-text defaults, so a vanilla Slim app with no user code is not exploitable. Only applications that feed untrusted data into setTitle() and/or setDescription() are affected. The issue has been fixed in 4.15.2. If developers are unable to immediately update their applications, they can work around this issue by avoiding passing untrusted/request-derived data into HttpException::setTitle() and setDescription() and using static, plain-text error copy instead.
They should also register a custom error renderer (an ErrorRendererInterface implementation, or a subclass of HtmlErrorRenderer that escapes the title and description) for the HTML media type.

🎖@cveNotify

🚨 CVE-2024-45636
IBM Security QRadar EDR 3.12 through 3.12.24 stores user credentials in plain text which can be read by a local privileged user.

🎖@cveNotify

🚨 CVE-2026-3341
IBM Langflow Desktop 1.0.0 through 1.9.2 IBM Langflow is vulnerable to server-side request forgery (SSRF). This may allow an authenticated attacker to send unauthorized requests from the system, potentially leading to network enumeration or facilitating other attacks.

🎖@cveNotify

🚨 CVE-2026-4096
IBM DevOps Plan 3.0.0 through 3.0.6 is vulnerable to HTTP header injection, caused by improper validation of input by the HOST headers. This could allow an attacker to conduct various attacks against the vulnerable system, including cross-site scripting, cache poisoning or session hijacking

🎖@cveNotify