Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.
Sources: National Vulnerability Database, CISA KEV, Exploit-DB
CVE-2026-106106 Overview
CVE-2026-106106 affects Quasar Framework, a Vue.js framework for building high-performance user interfaces. The flaw resides in renderSSRError() within utils/render-ssr-error/src/index.js, which serialized process.env, request headers, and cookies into the HTTP response returned by serve.devError(). Because the development server binds to all interfaces by default, any network-adjacent client that triggers a server-side rendering (SSR) or static site generation (SSG) failure can retrieve shell environment secrets. The renderer also escaped only one exact lowercase script closing-tag spelling, enabling reflected cross-site scripting (XSS) through case variants and alternative closing-tag delimiters.
Critical Impact
Network-adjacent attackers can harvest shell environment variables, cookies, and HTTP headers from Quasar developer machines, and inject script markup into error pages rendered in a developer's browser.
Affected Products
- @quasar/render-ssr-error prior to 2.2.4
- @quasar/app-vite prior to 3.3.0
- Quasar Framework projects using SSR or SSG development mode
Discovery Timeline
- 2026-10-06 - CVE-2026-106106 published to NVD
- 2026-10-06 - Last updated in NVD database
Technical Details for CVE-2026-106106
Vulnerability Analysis
The vulnerability combines two defects in the development-mode SSR error renderer. First, diagnostic data gathered in utils/render-ssr-error/src/env.js is serialized directly into the HTML error page, exposing process.env entries, inbound request headers, and cookies. Second, the HTML serializer only neutralized the exact lowercase string </script>, missing case-variant and alternate delimiter forms that allow a reflected payload to break out of the script element.
The issue is categorized as CWE-79: Improper Neutralization of Input During Web Page Generation. Because the Quasar development server listens on 0.0.0.0 by default, any host on the same subnet (including café Wi-Fi, VPN peers, or lab segments) can reach the error path and read secrets without authentication.
Root Cause
The renderer treats developer-only diagnostic context as safe for inclusion in HTTP responses. env.js collected shell environment variables and all request headers into a plain object that was later stringified into the error page template. The script-tag escaping relied on a single hardcoded pattern, so inputs like </Script> or </script > survived sanitization intact.
Attack Vector
An adjacent attacker sends a crafted request that forces the Vite/Quasar dev server into an SSR or SSG render failure. The resulting error page echoes back environment variables, cookies, and headers controlled or observable by the attacker. If a developer visits the same failing URL with an attacker-shaped payload, the injected markup executes in the developer's browser context.
// Patch excerpt: utils/render-ssr-error/src/env.js
return Object.keys(req.headers).reduce((acc, name) => {
acc[name] = req.headers[name]
return acc
- }, {})
+ }, Object.create(null))
+}
+
+function decodeCookieValue(value) {
+ try {
+ return decodeURIComponent(value)
+ } catch {
+ return value
+ }
}
// Source: https://github.com/quasarframework/quasar/commit/61c2bd8a607785fdade72cce20e8faf1de7eee15
The companion patch removes the Shell environment variables block from the serialized diagnostic payload entirely, eliminating the information disclosure path.
Detection Methods for CVE-2026-106106
Indicators of Compromise
- Unexpected inbound connections to Quasar dev server ports (commonly 9000, 9100, or 3000) from non-localhost addresses.
- HTTP responses from the dev server containing Shell environment variables, process.env, or large Cookies dumps.
- Request logs showing crafted paths or headers designed to trigger SSR render exceptions.
- Access logs with case-variant </Script> or </SCRIPT > payloads in query strings or headers.
Detection Strategies
- Inspect running Node.js processes for @quasar/app-vite versions earlier than 3.3.0 and @quasar/render-ssr-error earlier than 2.2.4.
- Alert on dev servers bound to 0.0.0.0 that are reachable from corporate or guest networks.
- Review proxy or endpoint telemetry for outbound exfiltration of environment variable names such as SSH_AUTH_SOCK, AWS_SECRET_ACCESS_KEY, or GITHUB_TOKEN sourced from developer workstations.
Monitoring Recommendations
- Monitor developer endpoints for network listeners exposed beyond 127.0.0.1 using host-based telemetry.
- Capture HTTP response bodies from internal dev-server scans and flag content matching the diagnostic data signature.
- Track package.json and pnpm-lock.yaml changes in source repositories to confirm upgrades to patched Quasar versions.
How to Mitigate CVE-2026-106106
Immediate Actions Required
- Upgrade @quasar/render-ssr-error to 2.2.4 and @quasar/app-vite to 3.3.0 or later across all developer workstations and CI runners.
- Rotate any secrets, tokens, or credentials that were present in developer shell environments while vulnerable versions were in use.
- Restrict the Quasar development server to 127.0.0.1 to prevent adjacent-network access.
Patch Information
The fix is tracked in GitHub Security Advisory GHSA-r5mf-4r5x-q78f and delivered via the Quasar App Vite 3.3.0 release. The upstream change set is visible in the GitHub commit overview, which removes shell environment serialization and hardens cookie decoding.
Workarounds
- Bind the dev server to loopback by setting the host option to 127.0.0.1 in quasar.config.js until the upgrade is deployed.
- Run quasar dev inside an isolated network namespace or firewall the dev server ports from all non-local sources.
- Avoid starting the dev server from shells that export sensitive secrets such as cloud credentials or SSH agent sockets.
# quasar.config.js - restrict dev server to loopback
devServer: {
host: '127.0.0.1',
port: 9000,
strictPort: true
}
Disclaimer: This content was generated using AI. While we strive for accuracy, please verify critical information with official sources.
Sources: National Vulnerability Database, CISA KEV, Exploit-DB

