UTC
Medium

Browser Use Upload Tool Reads Arbitrary Files from the n8n Server

GHSA-9qx8-fhxv-w794 · published 2026-10-28

Severity
Medium
CVSS score
6.0
Chance of exploitation (EPSS)
No CVE
Fixed in
3.0.1
What it is

When the Browser Use feature is enabled, the browser_upload tool accepted absolute file paths and passed them to Playwright. Because the browser session is connected remotely, Playwright resolved and read those paths on the n8n server rather than on the user's machine. An authenticated member who connects their own browser session could supply arbitrary absolute paths and have the file contents delivered into the page the assistant drives, where they could be retrieved. The approval prompt shown to the user named only the active page host; the file path was never surfaced or gated. Files readable by the n8n process user, including instance configuration files, were within scope. …

What to do

Upgrade n8n to 3.0.1 or later. If you start it from an MCP config with npx or uvx, pin that version so an old cached copy is not used.

The issue has been fixed in n8n versions 2.46.1, 2.45.5 and 3.0.1. Users should upgrade to one of these versions or later to remediate the vulnerability.

If you cannot upgrade yet

If upgrading is not immediately possible, administrators should consider the following temporary mitigations: Disable the Browser Use feature by setting N8N_INSTANCE_AI_BROWSER_USE_ENABLED=false or leaving it unset. Restrict n8n instance access to fully trusted users only. Remove global:member access from any users who are not fully trusted until the instance is patched. These workarounds do not fully remediate the risk and should only be used as short-term mitigation measures.

Affected
PackageRegistryAffectedFixed in
n8nnpm< 2.45.52.45.5
n8nnpm< 2.46.12.46.1
n8nnpm< 3.0.13.0.1

The impact, patch and workaround text comes from the advisory, shortened (published by n8n). Proof-of-concept and reproduction sections and code are left out; the original advisory has them.

Sources: GitHub Advisory Database, reviewed and malware advisories (CC BY 4.0, https://github.com/advisories); advisories published by Cursor (https://github.com/cursor/cursor/security/advisories); advisories published by OpenClaw (https://github.com/openclaw/openclaw/security/advisories); advisories published by vLLM (https://github.com/vllm-project/vllm/security/advisories); advisories published by Open WebUI (https://github.com/open-webui/open-webui/security/advisories); advisories published by n8n (https://github.com/n8n-io/n8n/security/advisories); advisories published by Flowise (https://github.com/FlowiseAI/Flowise/security/advisories); advisories published by Langflow (https://github.com/langflow-ai/langflow/security/advisories); advisories published by the MCP Python SDK project (https://github.com/modelcontextprotocol/python-sdk/security/advisories); advisories published by MLflow (https://github.com/mlflow/mlflow/security/advisories); CISA Known Exploited Vulnerabilities catalog (CC0). Data as of 2026-11-22. ShieldMap describes how each problem works and how to fix it; it does not publish exploit code. Report a correction to [email protected].