Nova Browser takes security, memory safety, and user privacy seriously. This document outlines our security policies, vulnerability disclosure guidelines, response timelines, and threat model.
Nova Browser operates on a rolling release cadence. Security patches and hardening updates are delivered directly to the main branch and published in the latest tagged release.
| Version | Supported | Notes |
|---|---|---|
| 1.4.x | Yes | Current active release branch with active security patches |
| 1.3.x | No | End of life; upgrade to 1.4.x or later |
| < 1.3.0 | No | Unsupported legacy versions |
Nova Browser employs defense-in-depth principles across all application boundaries:
- Multi-Process Sandboxing: Chromium renderers operate in isolated sandbox environments with Node.js integration disabled and context isolation strictly enforced (
contextIsolation: true,nodeIntegration: false,sandbox: true). - Webview Preload Scoping: Specialized preloads (
webstore-preload.cjs) are restricted exclusively to authorized Chrome Web Store origins (chromewebstore.google.comandchrome.google.com). All general webviews receive standard, restricted sandboxed preloads. - IPC Origin Validation: All IPC channels evaluate the sender's frame origin and webContents identifier using
isTrustedSender. Subframe iframes and foreign webContents cannot invoke privileged host APIs. - Credential Isolation: Master keys and authentication tokens are encrypted using platform keystores (macOS Keychain, Windows DPAPI, Linux Secret Service via Electron
safeStorage). Plaintext tokens and password hashes are never persisted in unencryptedlocalStorage. - Local MCP Server Security: The internal Model Context Protocol (MCP) server binds strictly to the IPv4 loopback interface (
127.0.0.1), enforces dynamic per-session bearer token authorization, and validates all incomingHostheaders against DNS rebinding attacks.
| In Scope | Out of Scope |
|---|---|
Sandbox escape or privilege escalation from <webview> to host process |
Attacks requiring physical access or root/administrator privileges on the user device |
| Remote Code Execution (RCE) or Command Injection via IPC or URL schemes | Self-XSS requiring users to paste malicious code into Developer Tools |
| Insecure credential storage or plaintext token leaks to renderer DOM | Vulnerabilities in third-party websites visited within the browser webview |
| SSRF vulnerabilities in internal proxy handlers or MCP server endpoints | Denial of Service through resource exhaustion on severely resource-constrained machines |
| Bypass of Safe Navigation URL filter policies or phishing blocklists | Issues in unmaintained or deprecated upstream third-party dependencies without proof of exploitability in Nova |
Please do not report security vulnerabilities via public GitHub issues or public forums.
- GitHub Security Advisories (Preferred): Submit a private report via GitHub Security Advisories.
- Encrypted Email: Send your report to
security@nova-browser.orgencrypted with our PGP public key.
To expedite triage and validation, please include:
- A clear description of the vulnerability and its potential security impact.
- Step-by-step reproduction instructions or a minimal reproducible proof-of-concept (PoC).
- Target platform (macOS Apple Silicon/Intel, Windows x64, Linux) and Nova Browser version.
- Any relevant logs, stack traces, or screenshots.
Key ID: 0x4F8A9B1C2D3E4F5A
Fingerprint: 8A2C 4F91 D3B6 E872 90C4 1B5E 4F8A 9B1C 2D3E 4F5A
UID: Nova Browser Security Team <security@nova-browser.org>
We commit to the following response timeline for valid security reports:
- Initial Acknowledgment: Within 24 hours of report receipt.
- Triage & Severity Assessment: Within 48 hours with assigned severity classification (CVSS v3.1).
- Patch Development & Release:
- Critical (CVSS 9.0 - 10.0): Targeted patch release within 72 hours.
- High (CVSS 7.0 - 8.9): Targeted patch release within 7 days.
- Medium / Low (CVSS < 7.0): Addressed in the subsequent scheduled release cycle (within 14 to 30 days).
- Public Disclosure: Coordinated disclosure within 90 days of reporting, or upon public release of the patch.
We consider security research conducted under this policy to be authorized. We will not pursue legal action against researchers who:
- Act in good faith to avoid privacy violations, data destruction, and service interruption.
- Give us reasonable time to remediate vulnerabilities before public disclosure.
- Do not exploit identified vulnerabilities beyond what is necessary to establish proof-of-concept.