| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| Cursor is a code editor built for programming with AI. Prior to Sep 27, 2024, if a user generated a terminal command via Cursor's Terminal Cmd-K/Ctrl-K feature and if the user explicitly imported a malicious web page into the Terminal Cmd-K prompt, an attacker with control over the referenced web page could have a significant chance of influencing a language model to output arbitrary commands for execution in the user's terminal. This scenario would require the user explicitly opt-in to including the contents of a compromised webpage, and it would require that the attacker display prompt injection text in the the contents of the compromised webpage.
A server-side patch to not stream back newlines or control characters was released on September 27, 2024, within two hours of the issue being reported. Additionally, Cursor 0.42 includes client-side mitigations to prevent any newline or control character from being streamed into the terminal directly. It also contains a new setting, `"cursor.terminal.usePreviewBox"`, which, if set to true, streams the response into a preview box whose contents then have to be manually accepted before being inserted into the terminal. This setting is useful if you're working in a shell environment where commands can be executed without pressing enter or any control character. The patch has been applied server-side, so no additional action is needed, even on older versions of Cursor. Separately, Cursor's maintainers also recommend, as best practice, to only include trusted pieces of context in prompts. |
| uploadsm in ChargePoint Home Flex 5.5.4.13 does not validate a user-controlled string for bz2 decompression, which allows command execution as the nobody user. |
| @jmondi/url-to-png is an open source URL to PNG utility featuring parallel rendering using Playwright for screenshots and with storage caching via Local, S3, or CouchDB. The package includes an `ALLOW_LIST` where the host can specify which services the user is permitted to capture screenshots of. By default, capturing screenshots of web services running on localhost, 127.0.0.1, or the [::] is allowed. If someone hosts this project on a server, users could then capture screenshots of other web services running locally. This issue has been addressed in version 2.1.1 with the addition of a blocklist. Users are advised to upgrade. There are no known workarounds for this vulnerability. |
| Improper input validation for some Intel(R) PROSet/Wireless WiFi software for Windows before version 23.60 may allow an unauthenticated user to potentially enable denial of service via network access. |
| matrix-js-sdk is the Matrix Client-Server SDK for JavaScript and TypeScript. In matrix-js-sdk versions versions 9.11.0 through 34.7.0, the method `MatrixClient.sendSharedHistoryKeys` is vulnerable to interception by malicious homeservers. The method was introduced by MSC3061) and is commonly used to share historical message keys with newly invited users, granting them access to past messages in the room. However, it unconditionally sends these "shared" keys to all of the invited user's devices, regardless of whether the user's cryptographic identity is verified or whether the user's devices are signed by that identity. This allows the attacker to potentially inject its own devices to receive sensitive historical keys without proper security checks. Note that this only affects clients running the SDK with the legacy crypto stack. Clients using the new Rust cryptography stack (i.e. those that call `MatrixClient.initRustCrypto()` instead of `MatrixClient.initCrypto()`) are unaffected by this vulnerability, because `MatrixClient.sendSharedHistoryKeys()` raises an exception in such environments. The vulnerability was fixed in matrix-js-sdk 34.8.0 by removing the vulnerable functionality. As a workaround, remove use of affected functionality from clients. |
| A flaw was found in github.com/go-viper/mapstructure/v2, in the field processing component using mapstructure.WeakDecode. This vulnerability allows information disclosure through detailed error messages that may leak sensitive input values via malformed user-supplied data processed in security-critical contexts. |
| Improper Input Validation vulnerability in pbkdf2 allows Signature Spoofing by Improper Validation. This vulnerability is associated with program files lib/to-buffer.Js.
This issue affects pbkdf2: from 3.0.10 through 3.1.2. |
| Padding oracle attack vulnerability in Oberon microsystem AG’s ocrypto library in all versions since 3.1.0 and prior to 3.9.2 allows an attacker to recover plaintexts via timing measurements of AES-CBC PKCS#7 decrypt operations. |
| web-auth/webauthn-lib is an open source set of PHP libraries and a Symfony bundle to allow developers to integrate that authentication mechanism into their web applications. The ProfileBasedRequestOptionsBuilder method returns allowedCredentials without any credentials if no username was found. When WebAuthn is used as the first or only authentication method, an attacker can enumerate usernames based on the absence of the `allowedCredentials` property in the assertion options response. This allows enumeration of valid or invalid usernames. By knowing which usernames are valid, attackers can focus their efforts on a smaller set of potential targets, increasing the efficiency and likelihood of successful attacks. This issue has been addressed in version 4.9.0 and all users are advised to upgrade. There are no known workarounds for this vulnerability. |
| An information disclosure vulnerability in ISPmanager v6.98.0 allows attackers to access sensitive details of the root user's session via an arbitrary command (ISP6-1779). |
| A flaw was found in the XFIXES extension. The XFixesSetClientDisconnectMode handler does not validate the request length, allowing a client to read unintended memory from previous requests. |
| E-WEBInformationCo. FS-EZViewer(Web) exposes sensitive information in the service. A remote attacker can obtain the database configuration file path through the webpage source code without login. Accessing this path allows attacker to obtain the database credential with the highest privilege and database host IP address. With this information, attackers can connect to the database and perform actions such as adding, modifying, or deleting database contents. |
| The Restrict – membership, site, content and user access restrictions for WordPress plugin for WordPress is vulnerable to Sensitive Information Exposure in all versions up to, and including, 2.2.8 via the WordPress core search feature. This makes it possible for unauthenticated attackers to extract sensitive data from posts that have been restricted to higher-level roles such as administrator. |
| A Local File Inclusion vulnerability in Vegam Solutions Vegam 4i versions 6.3.47.0 and earlier allows a remote attacker to obtain sensitive information through the print label function. Specifically, the filePathList parameter is susceptible to LFI, enabling a malicious user to include files from the web server, such as web.config or /etc/host, leading to the disclosure of sensitive information. |
| The Janssen Project is an open-source identity and access management (IAM) platform. Prior to version 1.8.0, the Config API returns results without scope verification. This has a large internal surface attack area that exposes all sorts of information from the IDP including clients, users, scripts ..etc. This issue has been patched in version 1.8.0. A workaround for this vulnerability involves users forking and building the config api, patching it in their system following commit 92eea4d. |
| A data exfiltration vulnerability exists in Anthropic’s deprecated Slack Model Context Protocol (MCP) Server via automatic link unfurling. When an AI agent using the Slack MCP Server processes untrusted data, it can be manipulated to generate messages containing attacker-crafted hyperlinks embedding sensitive data. Slack’s link preview bots (e.g., Slack-LinkExpanding, Slackbot, Slack-ImgProxy) will then issue outbound requests to the attacker-controlled URL, resulting in zero-click exfiltration of private data. |
| HCL iNotes is susceptible to a Reflected Cross-site Scripting (XSS) vulnerability caused by improper validation of user-supplied input. A remote, unauthenticated attacker can specially craft a URL to execute script in a victim's Web browser within the security context of the hosting Web site and/or steal the victim's cookie-based authentication credentials. |
| The AI Engine plugin for WordPress is vulnerable to Sensitive Information Exposure in all versions up to, and including, 3.1.3 via the /mcp/v1/ REST API endpoint that exposes the 'Bearer Token' value when 'No-Auth URL' is enabled. This makes it possible for unauthenticated attackers to extract the bearer token, which can be used to gain access to a valid session and perform many actions like creating a new administrator account, leading to privilege escalation. |
| Improper input validation in some Intel(R) Neural Compressor software before version v3.0 may allow an authenticated user to potentially enable escalation of privilege via adjacent access. |
| Saleor is an e-commerce platform. Starting in version 3.21.0 and prior to version 3.21.16, requesting certain fields in the response of `accountRegister` may result in errors that could unintentionally reveal whether a user with the provided email already exists in Saleor. Version 3.21.16 fixes the issue. As a workaround, rate-limit the mutation to reduce the impact. |
