Researchers at GitGuardian uncovered 321 instances of n8n APIs accepting token exposure via public GitHub commits, illustrating four methods attackers can exploit these credentials for unauthorized access to sensitive information and downstream tokens This article explores n8n security reports. . Organizations leverage this automation platform to integrate databases, source code repositories, cloud environments, artificial intelligence services, customer support platforms, and other internal systems.
The hostname-and-token pairing was often found under various variable names, including: Since the hostname was usually present alongside the token in the same commit, no separate infrastructure discovery step was required for our pipeline.
The audit endpoint provides an attack map for n8n's security reports: It identifies potential SQL injection exposures in workflows, nodes with filesystem access, unprotected webhooks, the running version of n8n, unused credentials, high-risk or community-installed nodes, enabled security features, node allowlists and blocklists, and telemetry settings. This endpoint returned details such as name, type, and identifier for each stored credential object, including one labeled "OpenAI account." By creating a workflow with a Schedule trigger and an OpenAI node that referenced the credential by its ID, we activated it.
These techniques reveal how an attacker can escalate from a leaked N8N token to broader credential and data exposure through legitimate platform functionality: Deleting the malicious workflow also removed associated execution records, potentially leaving defenders with limited evidence for investigation.
