HarborGuardharborguardDatabase
Back to search
CRITICALCVE-2026-7664Published Modified CNA ibm

CVE-2026-7664: Unauthenticated Flow Execution via Webhook Endpoint in Langflow OSS

IBM Langflow OSS 1.0.0 through 1.8.4 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.

Metrics

CVSS v3.1
9.8
Severity
CRITICAL
Fixed in
Affected Products
1

Get notified

Email me when this CVE is updated: new fix versions, severity changes, or any record change.

HarborGuard Analysis

Synopsis

An authentication bypass vulnerability affects IBM Langflow OSS versions 1.0.0 through 1.8.4, allowing unauthenticated remote attackers to reach protected MCP (Model Context Protocol) project resources through the Streamable MCP transport endpoint. The flaw is reachable over the network with no credentials required and no user interaction needed. Successful exploitation grants an attacker full read, write, and availability impact over the affected service. HarborGuard is tracking this advisory and will make a patched-image rebuild available the moment an upstream fix is published.

HarborGuard Coverage

Detection

Detection is available across every HarborGuard environment. The CVE is ingested from upstream feeds within minutes of publication and matched against all customer images, including custom-built images that bundle IBM Langflow OSS, in both registry scans and active pipeline checks.

Available
Triage

HarborGuard scores this CVE at 9.8 CRITICAL using the published CVSS v3.1 vector and surfaces it with the highest priority routing. Per-environment compliance policy weighting is applied automatically, directing alerts to the appropriate team inbox within each customer organization.

Available
Patch

No fix version has been published upstream. HarborGuard re-checks the advisory on every ingest cycle and will make a patched-image rebuild available at the fixed version the moment IBM publishes one. For customers with auto-remediation enabled, the rebuilt image, regression-test run, and a PR opened against affected workloads will follow automatically once an upstream fix exists.

Pending upstream

Exploit Conditions

  • Network reachabilityRequired

    The vulnerable endpoint is exposed over the network, meaning an attacker must be able to send HTTP requests to the Langflow MCP transport endpoint from any internet-accessible or internal network path.

  • AuthenticationNot required

    No credentials of any kind are required; the flaw is an authorization bypass that allows completely unauthenticated requests to succeed.

  • Victim interactionNot required

    The attacker sends requests directly to the service endpoint and does not need any user to click a link, open a file, or take any other action.

  • Attack complexityDetail

    Attack complexity is low, meaning the exploit is reliable and condition-free with no need to win a race condition or depend on specific memory layout or environmental factors.

Blast Radius

  • Reads protected MCP project resources, including any data, model configurations, or credentials stored within reachable projects.
  • Executes arbitrary MCP operations, allowing an attacker to invoke, modify, or delete workflows and pipeline components without restriction.
  • Modifies or destroys persisted project state, corrupting Langflow flows or overwriting stored configurations.
  • Disrupts availability of the Langflow service by triggering resource-exhausting or destructive MCP operations.

How HarborGuard Handles This

Available on HarborGuard: detection for CVE-2026-7664 is active now against all images running IBM Langflow OSS 1.0.0 through 1.8.4. Because no upstream fix has been published, HarborGuard monitors the IBM advisory on every ingest cycle and will surface a patched-image rebuild automatically the moment a fix version is released. For customers with auto-remediation enabled, that release triggers an immediate rebuild, regression-test run, and a PR opened against affected workloads. While no fix exists, recommended compensating controls include applying network policy to restrict inbound access to the Langflow MCP transport endpoint to trusted sources only, enabling egress filtering to limit the blast radius if the endpoint is reached, and disabling or gating the Streamable MCP transport feature via configuration or feature flag if your deployment does not require it.

See how HarborGuard automates this
Affected packages
  • IBM / Langflow OSS
    ≤ 1.8.4
CVSS Vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
References