Export limit exceeded: 374096 CVEs match your query. Please refine your search to export 10,000 CVEs or fewer.
Search
Search Results (374096 CVEs found)
| CVE | Vendors | Products | Updated | CVSS v3.1 |
|---|---|---|---|---|
| CVE-2026-70640 | 2026-08-06 | 7 High | ||
| llama.cpp builds b1886 through b7445 contain a race condition use-after-free vulnerability in the LLaMA-Android JNI wrapper where bench_1model() and free_1context() lack synchronization, allowing Thread A to operate on freed memory while Thread B concurrently frees the llama_context. Attackers can exploit this by performing heap spray with attacker-controlled data containing a fake vtable to hijack the vtable pointer at offset +0x30, causing llama_batch_allocr::clear() to dereference arbitrary memory and achieve remote code execution. | ||||
| CVE-2026-70639 | 2026-08-06 | 5.5 Medium | ||
| llama.cpp builds b1886 through b7445 contain a null pointer dereference vulnerability in the LLaMA-Android JNI wrapper where the bench_1model() function fails to validate the model context pointer before dereferencing it. Attackers can supply a malicious, corrupt, or truncated model file to trigger a null context condition, causing a SIGSEGV crash that terminates the Android application process and results in denial of service. | ||||
| CVE-2026-18276 | 2026-08-06 | 4.3 Medium | ||
| Missing authorization in the websocket consumer in Scripta eScriptorium through 26.04.1 allows a remote authenticated user to subscribe to any document's event stream and observe another user's segmentation, transcription, import, export and training activity via the object_cls and object_pk values of a join-room message, which are passed to group_add without an access check | ||||
| CVE-2026-70638 | 2026-08-06 | 7.8 High | ||
| llama.cpp builds b1886 through b7445 contain an integer overflow vulnerability in the LLaMA-Android JNI wrapper where the new_1batch() function multiplies sizeof(llama_seq_id) by an attacker-controlled n_seq_max parameter without overflow validation, causing heap buffer allocation to wrap and allocate insufficient memory. Attackers can exploit this by providing a crafted n_seq_max value through a malicious model file or JNI call to trigger heap corruption and achieve denial of service or arbitrary code execution on Android applications using the LLaMA-Android binding. | ||||
| CVE-2026-18275 | 2026-08-06 | 6.5 Medium | ||
| Authorization bypass in the process and annotation taxonomy serializers in Scripta eScriptorium through 26.04.1 allows a remote authenticated user to run segmentation and transcription against other users' document parts, overwriting their content, via part primary keys supplied to a many=True related field whose queryset restriction was applied to the ManyRelatedField instead of its child_relation and therefore had no effect | ||||
| CVE-2026-43628 | 2026-08-06 | 7.8 High | ||
| llama.cpp builds b3978 through b9058 contain an integer underflow and out-of-bounds read vulnerability in the DRY sampler that allows unauthenticated attackers to trigger a heap buffer underflow by sending a crafted HTTP request with dry_allowed_length set to INT32_MIN to the /v1/completions or /v1/chat/completions endpoints. Attackers can exploit this vulnerability to crash the server with SIGSEGV causing denial of service for all connected users, or corrupt token sampling probabilities by reading garbage values from memory before the allocated buffer. | ||||
| CVE-2026-18258 | 2026-08-06 | 8.8 High | ||
| Authorization bypass in the Line, LineTranscription, VirtualCollection, tag and process API endpoints in Scripta/eScriptorium through 26.04.1 allows a remote authenticated user to read, modify and delete other users' transcription content via primary keys supplied in the request body, which are queried against the global model manager instead of the request-scoped queryset | ||||
| CVE-2026-5423 | 1 Neo4j | 1 Graphql | 2026-08-06 | N/A |
| @neo4j/graphql library versions prior to 7.5.6 fail to verify the authenticity of a client-supplied, pre-decoded JWT object passed through GraphQL subscription connectionParams. As a result, any unauthenticated remote client that can open a GraphQL-over-WebSocket connection can forge arbitrary JWT claims (e.g. sub, roles) in connectionParams.jwt and have them accepted as authenticated identity for the purposes of @authentication and @subscriptionsAuthorization directive evaluation. This allows a fully unauthenticated attacker to receive subscription events that should be restricted to specific authenticated roles/users. Upgrade the library to versions 7.5.6+ or 5.12.14+. v6 is end-of-life and will not receive a fix. | ||||
| CVE-2026-19058 | 1 Foundationagents | 1 Metagpt | 2026-08-06 | 5.3 Medium |
| A vulnerability was found in FoundationAgents MetaGPT up to 0.8.2. The impacted element is the function DataInterpreter of the file metagpt/roles/di/data_interpreter.py. The manipulation results in code injection. The attack must be initiated from a local position. The exploit has been made public and could be used. The vendor was contacted early about this disclosure but did not respond in any way. | ||||
| CVE-2026-19044 | 1 Leesinliang | 1 Godot-mcp | 2026-08-06 | 5.3 Medium |
| A flaw has been found in LeeSinLiang godot-mcp 0.1.0. Affected by this vulnerability is the function executeOperation of the file src/index.ts of the component create_scene/add_node. This manipulation of the argument projectPath causes command injection. The attack needs to be launched locally. The project was informed of the problem early through an issue report but has not responded yet. | ||||
| CVE-2026-65576 | 2026-08-06 | 9.8 Critical | ||
| Unauthenticated PHP Object Injection in Adrena <= 1.2.14 versions. | ||||
| CVE-2026-61964 | 2 Wordpress, Wpmanageninja | 2 Wordpress, Ninja Tables | 2026-08-06 | 7.1 High |
| Unauthenticated Cross Site Scripting (XSS) in Ninja Tables <= 5.2.9 versions. | ||||
| CVE-2026-65502 | 2 Bdthemes, Wordpress | 2 Element Pack Elementor Addons, Wordpress | 2026-08-06 | 5.3 Medium |
| Unauthenticated Bypass Vulnerability in Element Pack Elementor Addons <= 8.7.13 versions. | ||||
| CVE-2026-18427 | 2026-08-06 | 7.5 High | ||
| @fastify/static before version 10.1.3 contains an incomplete fix for a previous route guard bypass. The static file handler rejected only parent directory segments, but it did not canonicalize dot segments, duplicate slashes, encoded dots, or backslashes before route matching and before delegating to the send layer. As a result, an unauthenticated attacker could request a file protected by a route based guard using a non canonical path form that misses the guarded route yet resolves back onto the protected file, disclosing its contents. Applications that protect a subtree of the static root with a route based guard are affected, while applications relying on the allowedPath option are not. This is fixed in @fastify/static 10.1.3, which canonicalizes the pathname, including rejecting backslashes, on the path used for routing and serving. | ||||
| CVE-2026-68749 | 1 Rrrene | 1 Html Sanitize Ex | 2026-08-06 | N/A |
| Inefficient Regular Expression Complexity vulnerability in the CSS scrubber in rrrene html_sanitize_ex allows an unauthenticated remote attacker to exhaust server CPU via a long CSS declaration in sanitized HTML. The declaration regex in HtmlSanitizeEx.Scrubber.CSS.scrub/1 matches the property name with an unbounded greedy [-\w]+ followed by a mandatory :, so a long run of word characters not followed by a colon makes the engine give back one character at a time and retry the colon at every start offset. The work is quadratic in the length of the run, and no length cap is applied to the CSS handed to the scrubber. An 80 KB <style> body costs roughly 2.4 seconds of scheduler time, so a few concurrent requests saturate the BEAM scheduler pool and make the application unresponsive. The impact is CPU exhaustion only. Nothing is read, modified or disclosed. This issue affects html_sanitize_ex: from 0.3.1 before 1.5.3. | ||||
| CVE-2026-68747 | 1 Rrrene | 1 Html Sanitize Ex | 2026-08-06 | N/A |
| Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection') vulnerability in the CSS scrubber in rrrene html_sanitize_ex allows an unauthenticated remote attacker to inject CSS at-rules, including an import of a remote stylesheet, into a page served to other users. HtmlSanitizeEx.Scrubber.CSS.scrub/1 applies its property and value allowlist through a Regex.replace over substrings matching a property: value declaration pattern, so input that does not match that pattern is never inspected and is copied to the output unchanged. @import url(//attacker.example/style.css); survives, while the same URL inside a background: url(...) declaration is removed. Element boundaries are resolved before the scrubber runs, so injected content does not escape the <style> element and no script executes. This issue affects html_sanitize_ex: from 0.3.1 before 1.5.4. | ||||
| CVE-2026-43627 | 2026-08-06 | 7.8 High | ||
| llama.cpp builds b1283 through b9058 contain an integer overflow vulnerability in the llama_batch_init() function where unchecked multiplications in malloc() calls can wrap past INT32_MAX when computing allocation sizes. Attackers can pass specially crafted parameters to trigger integer overflow, causing heap corruption and potentially achieving arbitrary code execution through subsequent batch operations that write past allocated buffer boundaries. | ||||
| CVE-2026-53983 | 2026-08-06 | 8.6 High | ||
| Ground Station prior to 0.6.0 contains an unauthenticated blind server-side request forgery vulnerability in the orbital-source configuration path that allows any unauthenticated Socket.IO client to cause the ground-station process to issue outbound HTTP requests to attacker-chosen destinations. Attackers can connect to the Socket.IO server on port 7000 without credentials due to disabled authentication enforcement and a wildcard CORS policy, then submit a data_submission event with submit-orbital-sources action to persist an attacker-supplied URL in the database, then trigger an orbital sync via the equally unauthenticated background_task:start event. The URL is stored with no scheme allowlist, no host validation, and no rejection of loopback, RFC1918, or link-local (cloud instance metadata at 169.254.169.254) addresses, and is passed directly to requests.get in _fetch_http_3le and _fetch_http_omm in backend/tlesync/source_adapters.py. HTTP status codes and error messages from the outbound request are emitted in the orbital_sync_state Socket.IO event to all connected clients, providing a serviceable oracle for interpreting internal-service and cloud-metadata responses even though the raw response body is not directly leaked. Because the malicious source persists in the database across restarts and re-fires every 24 hours on the scheduled sync cycle, the primitive gives durable long-term SSRF without the attacker needing to remain connected. | ||||
| CVE-2026-67551 | 1 Apache | 1 Qpid Proton Dotnet | 2026-08-06 | 7.5 High |
| pre-authentication attacker could leverage type size/count handling to cause excessive allocation leading to potential denial of service. This issue affects Apache Qpid Proton-Dotnet: through 1.0.0. Users are recommended to upgrade to version 1.1.0, which fixes the issue. | ||||
| CVE-2026-15991 | 2026-08-06 | 8.8 High | ||
| The File Manager plugin for WordPress is vulnerable to arbitrary file deletion due to insufficient file path validation in the connector function in all versions from 6.0 - 6.9. This makes it possible for authenticated attackers, with subscriber-level access and above, to read and delete arbitrary files on the server, which can lead to remote code execution when the right file is deleted (such as wp-config.php). The bypass is triggered by passing cmd=rm or cmf=file in the URL query string of a POST request: elFinder's bind registration reads the command exclusively from $_POST and therefore never registers the rm.pre permission handler, while the dispatcher reads from the merged $_GET+$_POST superglobal and executes the rm or file command unchecked against a volume that defaults to ABSPATH. | ||||