GHSA-HJWH-XVFW-QRWJ: Credential Disclosure via Diagnostic Boundaries in mcp-searxng
Vulnerability ID: GHSA-HJWH-XVFW-QRWJ
CVSS Score: 5.5
Published: 2026-08-19
A credential disclosure vulnerability in the mcp-searxng NPM package prior to version 1.12.0 allows attackers to recover plain-text SearXNG Basic Authentication credentials. The application exposes these credentials via console logs (stderr), MCP logging notifications, validation error messages, and JSON-RPC error responses. This occurs because the application lacks comprehensive sanitization across diagnostic boundaries when credentials are parsed from the SEARXNG_URL environment variable.
TL;DR
mcp-searxng before 1.12.0 leaks plain-text SearXNG Basic Authentication credentials via standard error logs, MCP notifications, and JSON-RPC error outputs. Upgrading to 1.12.0 remediates the issue via a robust diagnostic sanitization layer.
⚠️ Exploit Status: POC
Technical Details
- CWE ID: CWE-209, CWE-532
- Attack Vector: Local (AV:L) / Logical exposure via MCP clients
- CVSS Score: 5.5 (Medium)
- Exploit Status: Proof of Concept (PoC) documented
- Affected Versions: < 1.12.0
- Patched Version: 1.12.0
- KEV Status: Not Listed
Affected Systems
- mcp-searxng NPM package
-
mcp-searxng: < 1.12.0 (Fixed in:
1.12.0)
Mitigation Strategies
- Upgrade mcp-searxng to version 1.12.0 or newer to enable centralized diagnostic sanitization.
- Rotate any exposed credentials stored in the SEARXNG_URL environment variable.
- Configure parameter properties in server.json to mark authentication parameters as secret.
Remediation Steps:
- Update the package dependency via 'npm install mcp-searxng@latest'.
- Modify the SearXNG instance configuration to invalidate previously exposed Basic Auth keys.
- Purge system error logs and centralized log aggregator data containing unredacted credentials.
References
Read the full report for GHSA-HJWH-XVFW-QRWJ on our website for more details including interactive diagrams and full exploit analysis.
Top comments (0)