Shared-Workflow Editor Can Exfiltrate Credentials via Workflow Tool Node Inline Sub-Workflow
GHSA-4r56-g65c-fm83 · published 2026-08-19
- Severity
- High
- CVSS score
- 7.2
- Chance of exploitation (EPSS)
- No CVE
- Fixed in
- 2.36.2
n8n's credential check did not inspect the inline workflow JSON of every node type that executes an inline sub-workflow, so a reference to a credential the acting user does not hold was accepted and persisted, on both the workflow-update and workflow-creation paths. An editor on a shared workflow, or any user creating a workflow through the REST API, Public API, or MCP, could plant such a node; when the workflow was later executed under an identity that does hold the credential, the inline sub-workflow resolved the secret and could send it to an attacker-controlled endpoint. The credential was never shared with the attacker.
Upgrade n8n to 2.36.2 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 2.35.4 and 2.36.2. Users should upgrade to one of these versions or later to remediate the vulnerability.
If upgrading is not immediately possible, administrators should consider the following temporary mitigations: Restrict workflow sharing to fully trusted users only, and avoid granting Editor access to untrusted users on shared workflows. Restrict n8n instance access to fully trusted users only. Audit shared workflows for unexpected Workflow Tool nodes with inline sub-workflows referencing credentials not owned by the editor, and rotate any credentials that may have been exposed. These workarounds do not fully remediate the risk and should only be used as short-term mitigation measures.
- CriticalLiteLLM has SQL Injection in Proxy API key verificationlitellmExploited
- HighLiteLLM has a sandbox escape in custom-code guardraillitellmExploited
- HighLangflow affected by Remote Code Execution via validate_code() exec()langflowExploited
- CriticalJupyter MCP Server: Browser DNS rebinding can hijack the Jupyter backend through management routesjupyter-mcp-server
- Criticalmem0ai Databricks vector store vulnerable to SQL injectionmem0ai
- CriticalFlowise NodeVM sandbox escape via puppeteer allowlist - authenticated RCE and arbitrary file read via Chromiumflowise, flowise-components
- Criticaln8n Vulnerable to Remote Code Execution via Expression Injectionn8n, n8nExploited
- Criticaln8n Has an Arbitrary File Read via Git Noden8n, n8n, n8n
- Criticaln8n: HTTP Request Node Pagination Prototype Pollution to RCEn8n, n8n, n8n
- Criticaln8n Has an XML Node Prototype Pollution Patch Bypassn8n, n8n, n8n
- Criticaln8n has XML Node Prototype Pollution that to RCEn8n, n8n, n8n
- Criticaln8n has Prototype Pollution in XML Webhook Body Parser that Leads to RCEn8n, n8n, n8n
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.