CRITICAL: CVE-2025-71336 (CVSS 9.8) — multiple products
CVE: CVE-2025-71336
CVSS: 9.8 (3.1) — CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Severity: CRITICAL
Status: Critical advisory
Affected
_See vendor advisory_
Summary
Flowise before 3.0.6 (affected versions 2.2.7-patch.1 and earlier) contains an unsandboxed remote code execution vulnerability in the Custom MCP feature, which is designed to execute OS commands such as launching local MCP servers. Because Flowise's authentication and authorization model is minimal and lacks role-based access control, and the default installation runs without authentication unless FLOWISE_USERNAME and FLOWISE_PASSWORD are set, an attacker can send a crafted JSON payload with the header 'x-request-from: internal' to the /api/v1/node-load-method/customMCP endpoint to execute arbitrary OS commands, resulting in complete compromise of the platform container or server.
Verified Sources
References
- https://github.com/FlowiseAI/Flowise/security/advisories/GHSA-6933-jpx5-q87q
- https://www.vulncheck.com/advisories/flowise-unsandboxed-remote-code-execution-via-custom-mcp
_Validated by the Lyrie Threat Intelligence Pipeline — 3 independent sources confirmed before publication. No speculation._
Lyrie Verdict
A vulnerability of this severity is exactly what Lyrie's anti-rogue-AI defense is built for: continuous, autonomous monitoring that doesn't wait for human reaction time.
Validated sources
- [1]NVD
- [2]GitHub Advisory
- [3]MITRE