Multiple critical security vulnerabilities have been identified and fixed in the nginx web and proxy server as part of a Red Hat security advisory. These include heap-based buffer overflow, use-after-free conditions, flow control window overflow in HTTP/2, and denial of service via crafted TLS connections. The advisory covers Red Hat Enterprise Linux 9.6 Extended Update Support and related products. The update addresses seven distinct CVEs affecting various nginx modules and components.
CVE-2026-133696 is an information disclosure vulnerability in continuedev continue versions 2.0 and 2.1.0. It affects the DataLogger.logDevData function in the core/data/log.ts file of the Dev Data Logging component. The flaw allows remote attackers to cause information disclosure. The vendor was contacted but did not respond. The vulnerability has a medium severity rating with a CVSS 4.0 score of 5.3.
The dsh-blue-team-mode npm package versions 3.0.0-beta.17, 3.0.0-beta.18, and 3.0.0-beta.20 contain malicious code. Any computer with this package installed or running should be considered fully compromised. Secrets and keys stored on the affected system must be rotated immediately from a different, secure computer. Removing the package alone does not guarantee removal of all malicious software introduced by it.
The npm package 'tonty-tech-interviews' contains malicious code that compromises any computer on which it is installed or running. The compromise is severe enough that all secrets and keys stored on the affected system should be rotated immediately from a separate, trusted machine. Removal of the package alone may not fully eliminate the malicious presence, as the attacker may have established persistent control.
A vulnerability was detected in continuedev continue 2.0.0/2.1.0. Affected by this issue is the function DocsCrawler of the file core/indexing/docs/crawlers/DocsCrawler.ts of the component Docs Crawler. The manipulation of the argument startUrl/useLocalCrawling results in server-side request forgery. It is possible to launch the attack remotely. The attack requires a high level of complexity. The exploitation is known to be difficult. The vendor was contacted early about this disclosure but did not respond in any way.
CVE-2026-54026 is an improper access control vulnerability in LibreChat versions prior to 0.8.6. It allows an authenticated attacker to manipulate message storage by using a conversation ID in the URL that they own while specifying a victim's conversation ID in the request body. This causes attacker-controlled messages to be stored in the victim's conversation. The issue is fixed in LibreChat version 0.8.6.
CVE-2026-133694 is an OS command injection vulnerability in continuedev continue versions 2.0.0 and 2.1.0. The flaw exists in the getSidebarContent function of the file extensions/vscode/src/ContinueGUIWebviewViewProvider.ts within the Mermaid Diagram Renderer component. This vulnerability allows remote attackers to inject operating system commands. The vendor was contacted but did not respond to the disclosure. The CVSS 4.0 base score is 5.3, indicating a medium severity level.
A regular expression complexity vulnerability (CWE-1333) exists in LibreChat versions from 0.8.6-rc1 up to but not including 0.8.9. An authenticated user with people-picker permission can submit crafted regex patterns that cause catastrophic backtracking, blocking the Node.js event loop and making the service unresponsive. Invalid regex patterns may also cause server errors. This issue is fixed in version 0.8.9.
CVE-2026-54031 is a server-side request forgery (SSRF) vulnerability in LibreChat versions prior to 0.8.8. The issue arises because the OAuth fetch handler validates only the initial destination URL for certain OAuth requests but fails to validate subsequent redirect targets. This allows an attacker who can initiate OAuth requests against a controlled endpoint to cause the server to make requests to internal or private network addresses, bypassing intended SSRF protections. The vulnerability is fixed in LibreChat version 0.8.8.
CVE-2026-54044 is a vulnerability in LibreChat versions prior to 0.8.7 where insufficient protection of credentials allows a collaborator with EDIT permission to redirect an action to an attacker-controlled host. This causes the owner's sensitive credentials, including API keys and OAuth client secrets, to be decrypted and sent to the attacker. The issue is fixed in version 0.8.7.
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.7, unauthenticated POST /api/user/verify requests in api/server/routes/user.js call api/server/services/AuthService.js verifyEmail and return distinct responses for unknown, verified, and unverified email addresses and the route lacks request throttling. A remote attacker can submit candidate addresses with invalid tokens to enumerate registered accounts and verification state, supporting targeted phishing or credential attacks. This issue is fixed in version 0.8.7.
CVE-2026-54045 is a vulnerability in LibreChat versions prior to 0.8.7 involving a weak password recovery mechanism. An unauthenticated attacker who knows a target's email can repeatedly delete pending verification tokens, preventing account verification until the unverified account expires. This issue is fixed in version 0.8.7.
CVE-2026-54042 is a vulnerability in LibreChat versions prior to 0.8.7 where the two-factor authentication (2FA) verification endpoint lacks a proper limiter on repeated attempts. An attacker with valid primary credentials and a temporary token can repeatedly guess TOTP or backup codes without restriction, potentially gaining full session access if a correct code is guessed. This flaw is fixed in version 0.8.7.
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. From 0.8.5 until 0.8.6, GET /api/user in api/server/controllers/UserController.js getUserController serializes federatedTokens and tokenset to the authenticated browser because an earlier response-sanitization fix was not applied to this endpoint. Browser-context script can obtain OAuth access, ID, and refresh credentials that were intended to remain server-side. This issue is fixed in version 0.8.6.
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.7, packages/data-schemas/src/methods/file.ts updateFileUsage accepts chat attachment files[].file_id values without restricting the lookup to the authenticated owner. An authenticated attacker who obtains another user's file identifier, including through GET /api/share/{shareId} for a shared conversation, can attach that file to a chat, causing extracted document text to enter the model context. The same operation can remove expiration metadata from the victim file. This issue is fixed in version 0.8.7.
CVE-2026-54032 is a cross-site scripting (XSS) vulnerability in LibreChat versions prior to 0.8.6. The issue arises because certain user interface components render HTML content from remote or administrator-controlled sources without proper sanitization, allowing malicious scripts to execute in the browser context. This can lead to exposure of browser data or manipulation of displayed content. The vulnerability is fixed in version 0.8.6.
A vulnerability in LibreChat versions 0.7.0 through 0.8.5 allows an unauthenticated attacker to forge a provider value for a victim's email during registration when local registration and social login providers are enabled. This leads to the attacker and victim sharing the same user record, granting the attacker persistent read and write access to the victim's data. The issue is fixed in version 0.8.6.
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.6, the POST /api/auth/login local-login validation path in api/strategies/localStrategy.js logs req.body, including the submitted plaintext password, during a Zod validation failure. The standard browser form validation makes this path less likely through the normal interface, but a direct invalid login request can reach it, and the logging formatter does not redact the password stored in metadata. A principal with access to application stdout, error logs, or journal output can recover the submitted credential. This issue is fixed in version 0.8.6.
CVE-2026-133693 is a path traversal vulnerability in the continuedev continue product versions 2.0.0 and 2.1.0. It affects the validateSearchAndReplaceFilepath function in the Edit Tool component. Exploitation requires local access and could allow an attacker to manipulate file paths improperly. The vendor has not responded to disclosure requests, and no patch or fix information is available.
CVE-2026-114499 is a vulnerability in the codebase-memory-mcp product by DeusData affecting versions prior to 0.10.0. The indexer's .gitignore matcher used recursive backtracking without limits for patterns with wildcards, which could cause exponential growth in matching work. This could stall repository discovery indefinitely and impact server availability before worker isolation was implemented. The issue is fixed in version 0.10.0.