OpenPanel MCP Token Plaintext Logging Exposure
First seen Sep 20, 2026 · Updated Sep 20, 2026 · CVSS 3.3
OpenPanel logs Model Context Protocol authentication tokens passed via URL query parameters into plaintext application logs, without redacting sensitive fields. An attacker who already has access to application stdout or a centralized logging system could recover these tokens and replay MCP requests to access project analytics. The CVSS score of 3.3 reflects that this requires pre-existing privileged access to logging infrastructure, limiting practical severity.
Technical Analysis
The vulnerability arises because OpenPanel passes MCP auth tokens as URL query parameters, a design choice that causes them to be captured by standard web/application request logging. These logs are written in plaintext (base64-encoded, not encrypted) to stdout or log aggregation systems, creating a secondary, less-guarded exposure path for credentials that were meant to authenticate MCP sessions. An attacker with log access—such as a malicious insider, a compromised logging pipeline, or misconfigured log storage—can extract these tokens and use them to impersonate a legitimate MCP client, replaying requests to pull analytics data. This crosses the boundary between the application layer and the protocol authentication layer, turning an operational logging practice into a credential leakage vector for the MCP session.
Affected Systems
OpenPanel; protocols: MCP
Detection Signatures
- Search logs for URL query strings containing parameters like token=, auth=, or mcp_token= followed by base64-encoded strings
- Audit stdout/log aggregation pipelines for presence of authentication material in request URLs
- Monitor for anomalous MCP requests replayed from IPs/hosts not associated with legitimate clients
Remediation Steps
- 1
Move tokens out of URL parameters
Update OpenPanel to pass MCP authentication tokens via headers (e.g., Authorization) or request body instead of URL query strings to prevent automatic logging.
- 2
Redact sensitive fields in logging
Implement log sanitization/redaction middleware that strips or masks token-like query parameters before writing to stdout or log aggregation systems.
- 3
Rotate exposed credentials
Rotate all MCP tokens that may have been logged prior to the patch, and audit logging systems for historical exposure.
- 4
Restrict log access
Apply least-privilege access controls and encryption-at-rest to centralized logging systems to reduce the risk of credential harvesting from logs.
CVE / Advisory IDs
Industries Most Exposed
Respond to this threat
Pro subscribers get a full AI-generated incident-response playbook for this threat — detection, containment, eradication, and recovery steps — plus an unlimited AI Threat Advisor for questions about your environment.