CVE-2026-19306

published 4 Sep 2026 · updated 4 Sep 2026

Langflow is vulnerable to arbitrary local file read due to path traversal in ChatInput, bundle FileInput, and GitExtractor components

7.7HighCVSS 7.7EPSS 1%

No exploitation in the wild has been reported. EPSS puts the chance of exploitation in the next 30 days at 1%. No fixed version is recorded.

IBM Langflow OSS 1.0.0 through 1.11.2 allows an authenticated attacker to read arbitrary files from the server filesystem — including server secret material (secret_key, JWT signing keys, the application database, /proc/self/environ, and other tenants' upload directories) — by supplying absolute paths or traversal sequences in the files parameter of an authenticated build request. The file contents were embedded as text attachments in the language model prompt and transmitted to the configured model endpoint, resulting in confidential data exfiltration. This bypassed the LANGFLOW_RESTRICT_LOCAL_FILE_ACCESS=true containment boundary, which was enforced for other file-reading components but not for the Chat Input to Message attachment pipeline.

Affected

Vendors, products and packages with version ranges

VendorProductVersionsFixed
IBMLangflow>=1.0.0 <=1.11.2

Timeline

Every event on this vulnerability

Related

Other vulnerabilities in Langflow, strongest first

Embed

A live severity badge for an advisory, README or ticket

CVE-2026-19306 severity on Zeptric[![CVE-2026-19306](https://zeptric.io/badge/cve/CVE-2026-19306.svg)](https://zeptric.io/cve/CVE-2026-19306)