CVE Intelligence
Analysis AI-assisted

CVE-2026-55450: Langflow: unauthenticated unlimited upload and path disclosure

Langflow versions before 1.9.1 let unauthenticated users upload unlimited data, which can exhaust server storage. The response also reveals the absolute path of uploaded files, an information leak that can help an attacker chain other weaknesses. The fix is available in 1.9.1, and no confirmed exploitation appears in the sources consulted.

Published Oct 13, 2026 · Updated Oct 13, 2026 · Ler em português

Overview

Langflow is a tool for building and deploying AI-powered agents and workflows. According to the CVE record, versions before 1.9.1 accept data uploads from unauthenticated users without any size limit. The weakness is classified as CWE-200 (Exposure of Sensitive Information), and the vendor advisory states that it is fixed in 1.9.1.

Impact

  • Storage exhaustion: an unauthenticated party with network access could fill server space, which may disrupt the service (inference from the record).
  • Information disclosure: responses reveal the absolute path of uploaded files, which the advisory notes can assist in chaining other primitives.
  • GitHub assigned a CVSS 3.1 base score of 9.3 (critical). The vector indicates network attack vector, no privileges or user interaction required, high availability impact, and low confidentiality impact.

Affected products and versions

  • Langflow (Langflow Langflow): versions less than 1.9.1, according to the record.

Exploitation status

The CISA KEV catalog does not list this CVE, and the record shows no KEV entry. No confirmed active exploitation appears in the sources consulted. The advisory reference is tagged as Exploit in the record, but this tag describes the reference type and does not by itself confirm exploitation in the wild.

Remediation and mitigation

  • Upgrade Langflow to version 1.9.1 or later, as indicated by the vendor advisory and the fix pull request.
  • Until patching is complete, limit network exposure of Langflow instances and keep them away from untrusted networks.
  • Place Langflow behind an access layer that enforces authentication for sensitive endpoints, where feasible.
  • Apply size and rate limits for upload traffic at a reverse proxy or gateway as a compensating control.

Detection guidance

  • Monitor disk utilization on hosts running Langflow and alert on rapid or unexpected growth in upload storage locations.
  • Review web server and application logs for upload requests that lack valid session or authentication context.
  • Track unusual volumes of inbound data to Langflow endpoints compared with a normal baseline.

Limitations of the available information

  • The record does not specify the exact vulnerable endpoint, default configuration, or whether any deployment-level limits already reduce exposure.
  • No EPSS score or SSVC data is present in the record.
  • The record title was empty; the title in this analysis was written from the description.
  • The sources consulted do not describe real-world attack activity, so the actual exposure of deployed systems is unknown.

Frequently asked questions

Is this Langflow vulnerability being actively exploited?

The CISA KEV catalog does not list it, and the sources consulted show no confirmed active exploitation. This does not guarantee that exploitation has not occurred, so monitoring remains advisable.

What should administrators do first?

Upgrade Langflow to version 1.9.1 or later. Until then, restrict network exposure and monitor disk usage on affected hosts.

This analysis was drafted with AI assistance from verified source data and automatically checked against it.