Severity by source
AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Network-reachable MCP endpoint with no authentication or interaction (AV:N/AC:L/PR:N/UI:N); missing authorization yields unauthorized resource access and operation execution, supporting high confidentiality and integrity, with availability high retained per vendor rating.
Primary rating from Vendor (ibm).
CVSS VectorNVD
Lifecycle Timeline
5DescriptionNVD
IBM Langflow OSS 1.0.0 through 1.9.6 could allow unauthenticated attackers to access protected MCP project resources and execute MCP operations due to improper authorization enforcement in the Streamable MCP transport endpoint.
AnalysisAI
Improper authorization enforcement in IBM Langflow OSS 1.0.0 through 1.9.6 lets unauthenticated remote attackers reach protected MCP (Model Context Protocol) project resources and invoke MCP operations through the Streamable MCP transport endpoint. Because the flaw bypasses authentication entirely on a network-facing endpoint and is rated CVSS 9.8 with full confidentiality, integrity, and availability impact, any exposed Langflow instance is at risk. There is no public exploit identified at time of analysis, and the EPSS probability is low (0.24%, 15th percentile), indicating exploitation has not yet become widespread.
Technical ContextAI
Langflow OSS is an open-source, IBM-maintained low-code visual framework for building LLM-driven agents and workflows, which exposes tools and data to AI clients via the Model Context Protocol (MCP). The vulnerability lives in the Streamable HTTP MCP transport endpoint, the server-sent/streaming channel MCP clients use to call project-scoped tools and resources. The root cause is CWE-285 (Improper Authorization): the endpoint fails to enforce the authorization checks that should gate access to a given MCP project, so requests that should require an authenticated, authorized identity are honored regardless. In effect the authorization decision is missing or incorrectly applied at the transport layer, allowing operation-level access to resources scoped to other users' or projects' MCP servers.
RemediationAI
Patch available per vendor advisory: IBM has released a fix, so upgrade Langflow OSS to the vendor-specified fixed release above 1.9.6 as documented at https://www.ibm.com/support/pages/node/7277570 (consult that advisory for the exact patched version, which is not enumerated in the available data). As an interim compensating control until patching, restrict network reachability of the Streamable MCP transport endpoint - place Langflow behind an authenticating reverse proxy or gateway that enforces identity before requests reach the endpoint, and block direct inbound access to the MCP port from untrusted networks; the trade-off is that legitimate remote MCP clients must be routed through the proxy and re-authenticated. Where MCP project exposure is not required, disable or do not expose the MCP server feature for sensitive projects, accepting the loss of remote MCP tool access. Avoid exposing Langflow directly to the internet.
More in Langflow Oss
View allUnauthenticated remote code execution in IBM Langflow OSS versions 1.0.0 through 1.10.0 lets any network-reachable attac
Authenticated command injection in IBM Langflow OSS 1.0.0 through 1.11.1 lets any logged-in flow user escalate to arbitr
Remote code execution in IBM Langflow OSS 1.0.0 through 1.11.1 allows a network-accessible authenticated attacker with l
Unauthenticated remote code execution in IBM Langflow OSS versions 1.0.0 through 1.9.3 allows attackers to fully comprom
Arbitrary Python code execution in IBM Langflow OSS 1.0.0 through 1.10.0 (Langflow versions up to 1.9.2) allows authenti
Remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 allows attackers to run arbitrary code by planting a mali
Authenticated remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 lets any logged-in user run arbitrary Pytho
Arbitrary file write in IBM Langflow OSS 1.0.0 through 1.10.0 lets an attacker who controls an external HTTP server plan
Remote code injection in IBM Langflow OSS versions 1.0.0 through 1.10.0 allows an authenticated remote attacker to execu
Remote code execution in IBM Langflow OSS 1.0.0 through 1.10.1 stems from improper input validation in the PythonREPL sa
Privilege escalation to remote code execution in IBM Langflow OSS 1.0.0 through 1.10.0 allows an authenticated user to e
Code injection in IBM Langflow OSS versions 1.0.0 through 1.10.0 lets an authenticated user execute arbitrary operating-
Same weakness CWE-285 – Improper Authorization
View allSame technique Authentication Bypass
View allShare
External POC / Exploit Code
Leaving vuln.today
EUVD-2026-40384
GHSA-5wjx-cxqm-p8r4