Signalements récents de vulnérabilités issus des annonces des éditeurs et de l’actualité de la sécurité. Ces alertes ne modifient pas la note de sécurité.
Selon OpenNet, Node.js présente une vulnérabilité d’exécution de code à distance (CVE-2026-111497). Les versions exactes affectées ne sont pas précisées. Le message ne précise pas de version corrigée.
Que faire maintenant : Vérifiez si une mise à jour de sécurité de Node.js contenant le correctif est disponible et installez-la.
Selon GitHub Advisory Database, Node.js présente une vulnérabilité de déni de service (CVE-2026-109932). Les versions exactes affectées ne sont pas précisées. Le message ne précise pas de version corrigée.
Que faire maintenant : Vérifiez si une mise à jour de sécurité de Node.js contenant le correctif est disponible et installez-la.
HFS user adding a "web link" in HFS is vulnerable to "target=_blank" exploit
### Summary
When adding a "web link" to the HFS virtual filesystem, the frontend opens it with `target="_blank"` but without the `rel="noopener noreferrer"` attribute. This allows the opened page to use the `window.opener` property to change the location of the original HFS tab.
### Details
While most modern browsers have fixes already implemented for this `target="_blank"` exploit at the browser level, users on outdated browsers remain vulnerable. This means that if an admin of the HFS instance adds a link to an external third-party service (that they believe is safe at the time) and that service they added later becomes compromised, the ma
AuthKit React Router: Sensitive auth data rendered in HTML
The AuthKit library for React Router 7+ provides helpers for authentication and session management using WorkOS & AuthKit with React Router. In versions 0.6.1 and below, @workos-inc/authkit-react-router exposed sensitive authentication artifacts — specifically sealedSession and accessToken by returning them from the authkitLoader. This caused them to be rendered into the browser HTML. This issue is fixed in version 0.7.0.
The AuthKit library for Remix provides convenient helpers for authentication and session management using WorkOS & AuthKit with Remix. In versions 0.14.1 and below, @workos-inc/authkit-remix exposed sensitive authentication artifacts — specifically sealedSession and accessToken — by returning them from the authkitLoader. This caused them to be rendered into the browser HTML.
@fedify/fedify: Improper Authentication and Incorrect Authorization
Fedify is a TypeScript library for building federated server apps powered by ActivityPub. In versions below 1.3.20, 1.4.0-dev.585 through 1.4.12, 1.5.0-dev.636 through 1.5.4, 1.6.0-dev.754 through 1.6.7, 1.7.0-pr.251.885 through 1.7.8 and 1.8.0-dev.909 through 1.8.4, an authentication bypass vulnerability allows any unauthenticated attacker to impersonate any ActivityPub actor by sending forged activities signed with their own keys. Activities are processed before verifying the signing key belongs to the claimed actor, enabling complete actor impersonation across all Fedify instances. This is fixed in versions 1.3.20, 1.4.13, 1.5.5, 1.6.8, 1.
oak: ReDoS in x-forwarded-proto and x-forwarded-for headers
oak is a middleware framework for Deno's native HTTP server, Deno Deploy, Node.js 16.5 and later, Cloudflare Workers and Bun. In versions 17.1.5 and below, it's possible to significantly slow down an oak server with specially crafted values of the x-forwarded-proto or x-forwarded-for headers.
Astro: Duplicate trailing slash feature can lead to Open Redirects
Astro is a web framework for content-driven websites. In versions 5.2.0 through 5.12.7, there is an Open Redirect vulnerability in the trailing slash redirection logic when handling paths with double slashes. This allows an attacker to redirect users to arbitrary external domains by crafting URLs such as https://mydomain.com//malicious-site.com/. This increases the risk of phishing and other social engineering attacks. This affects sites that use on-demand rendering (SSR) with the Node or Cloudflare adapters. It does not affect static sites, or sites deployed to Netlify or Vercel. This issue is fixed in version 5.12.8. To work around this iss
tmp does not restrict arbitrary temporary file / directory write via symbolic link `dir` parameter
tmp is a temporary file and directory creator for node.js. In versions 0.2.3 and below, tmp is vulnerable to an arbitrary temporary file / directory write via symbolic link dir parameter. This is fixed in version 0.2.4.
Thinbus generates insufficient entropy: 252 bits vs minimum 256 bits
Thinbus Javascript Secure Remote Password is a browser SRP6a implementation for zero-knowledge password authentication. In versions 2.0.0 and below, a protocol compliance bug causes the client to generate a fixed 252 bits of entropy instead of the intended bit length of the safe prime (defaulted to 2048 bits). The client public value is being generated from a private value that is 4 bits below the specification. This reduces the protocol's designed security margin it is now practically exploitable. The servers full sized 2048 bit random number is used to create the shared session key and password proof. This is fixed in version 2.0.1.
IPX is Vulnerable to Path Traversal via Prefix Matching Bypass
IPX is an image optimizer powered by sharp and svgo. In versions 1.3.1 and below, 2.0.0-0 through 2.1.0, and 3.0.0 through 3.1.0, the approach used to check whether a path is within allowed directories is vulnerable to path prefix bypass when the allowed directories do not end with a path separator. This occurs because the check relies on a raw string prefix comparison. This is fixed in versions 1.3.2, 2.1.1 and 3.1.1.
Claude Code Research Preview has a Path Restriction Bypass which could allow unauthorized file access
Claude Code is an agentic coding tool. In versions below 0.2.111, a path validation flaw using prefix matching instead of canonical path comparison, makes it possible to bypass directory restrictions and access files outside the CWD. Successful exploitation depends on the presence of (or ability to create) a directory with the same prefix as the CWD and the ability to add untrusted content into a Claude Code context window. This is fixed in version 0.2.111.
Claude Code echo command allowed bypass of user approval prompt for command execution
Claude Code is an agentic coding tool. In versions below 1.0.20, an error in command parsing makes it possible to bypass the Claude Code confirmation prompt to trigger execution of an untrusted command. Reliably exploiting this requires the ability to add untrusted content into a Claude Code context window. This is fixed in version 1.0.20.
js-toml is a TOML parser for JavaScript, fully compliant with the TOML 1.0.0 Spec. In versions below 1.0.2, a prototype pollution vulnerability in js-toml allows a remote attacker to add or modify properties of the global Object.prototype by parsing a maliciously crafted TOML input. This is fixed in version 1.0.2.
@nestjs/devtools-integration's CSRF to Sandbox Escape Allows for RCE against JS Developers
Nest is a framework for building scalable Node.js server-side applications. In versions 0.2.0 and below, a critical Remote Code Execution (RCE) vulnerability was discovered in the @nestjs/devtools-integration package. When enabled, the package exposes a local development HTTP server with an API endpoint that uses an unsafe JavaScript sandbox (safe-eval-like implementation). Due to improper sandboxing and missing cross-origin protections, any malicious website visited by a developer can execute arbitrary code on their local machine. The package adds HTTP endpoints to a locally running NestJS development server. One of these endpoints, /inspect
webfinger.js is vulnerable to Blind SSRF attacks through localhost
webfinger.js is a TypeScript-based WebFinger client that runs in both browsers and Node.js environments. In versions 2.8.0 and below, the lookup function accepts user addresses for account checking. However, the ActivityPub specification requires preventing access to localhost services in production. This library does not prevent localhost access, only checking for hosts that start with "localhost" and end with a port. Users can exploit this by creating servers that send GET requests with controlled host, path, and port parameters to query services on the instance's host or local network, enabling blind SSRF attacks. This is fixed in version
nyariv sandboxjs 0.8.23 Prototype Pollution Sandbox Escape DoS
A prototype pollution vulnerability exists in @nyariv/sandboxjs versions <= 0.8.23, allowing attackers to inject arbitrary properties into Object.prototype via crafted JavaScript code. This can result in a denial-of-service (DoS) condition or, under certain conditions, escape the sandboxed environment intended to restrict code execution. The vulnerability stems from insufficient prototype access checks in the sandbox’s executor logic, particularly in the handling of JavaScript function objects returned.
GitProxy is susceptible to a hidden commits injection attack
GitProxy is an application that stands between developers and a Git remote endpoint. In versions 1.19.1 and below, attackers can inject extra commits into the pack sent to GitHub, commits that aren’t pointed to by any branch. Although these “hidden” commits never show up in the repository’s visible history, GitHub still serves them at their direct commit URLs. This lets an attacker exfiltrate sensitive data without ever leaving a trace in the branch view. We rate this a High‑impact vulnerability because it completely compromises repository confidentiality. This is fixed in version 1.19.2.
GitProxy is vulnerable to a new branch approval exploit
GitProxy is an application that stands between developers and a Git remote endpoint. In versions 1.19.1 and below, attackers can exploit the way GitProxy handles new branch creation to bypass the approval of prior commits on the parent branch. The vulnerability impacts all users or organizations relying on GitProxy to enforce policy and prevent unapproved changes. It requires no elevated privileges beyond regular push access, and no extra user interaction. It does however, require a GitProxy administrator or designated user (canUserApproveRejectPush) to approve pushes to the child branch. This is fixed in version 1.19.2.
GitProxy is vulnerable to a packfile parsing exploit
GitProxy is an application that stands between developers and a Git remote endpoint (e.g., github.com). In versions 1.19.1 and below, an attacker can craft a malicious Git packfile to exploit the PACK signature detection in the parsePush.ts file. By embedding a misleading PACK signature within commit content and carefully constructing the packet structure, the attacker can trick the parser into treating invalid or unintended data as the packfile. Potentially, this would allow bypassing approval or hiding commits. This issue is fixed in version 1.19.2.
GitProxy bypasses approvals when pushing multiple branches
GitProxy is an application that stands between developers and a Git remote endpoint (e.g., github.com). Versions 1.19.1 and below allow users to push to remote repositories while bypassing policies and explicit approvals. Since checks and plugins are skipped, code containing secrets or unwanted changes could be pushed into a repository. This is fixed in version 1.19.2.
A SAML library not dependent on any frameworks that runs in Node. In version 5.0.1, Node-SAML loads the assertion from the (unsigned) original response document. This is different than the parts that are verified when checking signature. This allows an attacker to modify authentication details within a valid SAML assertion. For example, in one attack it is possible to remove any character from the SAML assertion username. To conduct the attack an attacker would need a validly signed document from the identity provider (IdP). This is fixed in version 5.1.0.