CVE-2026-26019
Awaiting Analysis Awaiting Analysis - Queue
Improper URL Validation in LangChain RecursiveUrlLoader Enables SSRF

Publication date: 2026-02-11

Last updated on: 2026-02-19

Assigner: GitHub, Inc.

Description
LangChain is a framework for building LLM-powered applications. Prior to 1.1.14, the RecursiveUrlLoader class in @langchain/community is a web crawler that recursively follows links from a starting URL. Its preventOutside option (enabled by default) is intended to restrict crawling to the same site as the base URL. The implementation used String.startsWith() to compare URLs, which does not perform semantic URL validation. An attacker who controls content on a crawled page could include links to domains that share a string prefix with the target, causing the crawler to follow links to attacker-controlled or internal infrastructure. Additionally, the crawler performed no validation against private or reserved IP addresses. A crawled page could include links targeting cloud metadata services, localhost, or RFC 1918 addresses, and the crawler would fetch them without restriction. This vulnerability is fixed in 1.1.14.
CVSS Scores
EPSS Scores
Probability:
Percentile:
Meta Information
Published
2026-02-11
Last Modified
2026-02-19
Generated
2026-06-16
AI Q&A
2026-02-12
EPSS Evaluated
2026-06-14
NVD
CVE-2026-26019
Affected Vendors & Products
Showing 1 associated CPE
Vendor Product Version / Range
langchain langchain_community to 1.1.14 (exc)
Helpful Resources
Exploitability
CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-918 The web server receives a URL or similar request from an upstream component and retrieves the contents of this URL, but it does not sufficiently ensure that the request is being sent to the expected destination.
Attack-Flow Graph
AI Quick Actions
Instant insights powered by AI
Executive Summary

This vulnerability exists in the RecursiveUrlLoader class of the LangChain framework prior to version 1.1.14. The class is a web crawler that follows links recursively starting from a given URL. It has a preventOutside option meant to restrict crawling to the same site as the base URL. However, the implementation used a simple string prefix check (String.startsWith()) to compare URLs, which does not properly validate URLs semantically.

Because of this, an attacker controlling content on a crawled page could insert links to domains that share a string prefix with the target domain, tricking the crawler into following links to attacker-controlled or internal infrastructure. Additionally, the crawler did not validate against private or reserved IP addresses, allowing it to fetch links pointing to cloud metadata services, localhost, or RFC 1918 addresses without restriction.

This flaw could lead to unintended crawling of malicious or sensitive internal resources. The vulnerability was fixed in version 1.1.14.

Impact Analysis

This vulnerability can impact you by allowing an attacker to manipulate the web crawler to access unintended domains or internal network resources. This could lead to exposure of sensitive internal infrastructure or cloud metadata services that are normally protected.

Such unauthorized crawling could result in information disclosure, potentially leaking internal IP addresses or sensitive data accessible via internal services. It may also allow attackers to use the crawler as a proxy to access restricted resources.

Compliance Impact

I don't know

Detection Guidance

I don't know

Mitigation Strategies

To mitigate this vulnerability, upgrade the LangChain framework to version 1.1.14 or later, where the issue with RecursiveUrlLoader's URL validation is fixed.

Chat Assistant
Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-26019. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70
EPSS Chart