# CVE-2026-51888: langflow directory traversal in knowledge base creation

Source: https://junglewise.ai/threats/cve-2026-51888-langflow-ai-langflow-v1-8-4-is-affected-by-directory-traversal  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-10-01. Updated: 2026-10-05  
License: CC BY 4.0 with a link back

CVE-2026-51888 · Severity: info · Published 2026-10-01

Langflow is an AI workflow automation platform that includes a knowledge base feature for organizing AI training data. A directory traversal vulnerability in the knowledge base creation endpoint allows attackers to write or overwrite files anywhere on the server by using relative path sequences like "../" in the knowledge base name, potentially compromising the server and other users' data.

Technologies: [Langflow-Ai Langflow](https://junglewise.ai/threats/technologies/langflow-ai-langflow). Vendors: [Langflow-Ai](https://junglewise.ai/threats/vendors/langflow-ai).

## Quick answers

**Is CVE-2026-51888 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-51888 being exploited in the wild.

**How severe is CVE-2026-51888?** Informational severity, published 1 October 2026.

**Which products and versions are affected?** Langflow-Ai Langflow up to 1.8.4.

**Is there a fix for CVE-2026-51888?** No fixed version is recorded yet. Check the vendor's advisory in the references below. If you run it, Junglewise drafts a mitigation that fits your setup and applies it once you approve, with senior security engineers behind it.

**How is CVE-2026-51888 attacked?** It is attacked over the network, in a software library.

**Am I affected by CVE-2026-51888?** Check whether you run Langflow-Ai Langflow at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## Technical details

The vulnerability is a relative path traversal (CWE-22) in the knowledge_bases-create_knowledge_base HTTP POST endpoint (src/backend/base/langflow/api/v1/knowledge_bases.py, line 51). An attacker can inject path traversal sequences (e.g., ../../) via the knowledge base name parameter to write files outside the intended storage directory. The attack requires network access to the Langflow service and may require authentication depending on endpoint configuration.

## Affected products

- langflow-ai langflow up to 1.8.4

## References

- [https://gist.github.com/Ro1ME/c85f513d1fd702c686cffe061cbf8ef5](https://gist.github.com/Ro1ME/c85f513d1fd702c686cffe061cbf8ef5)

## Related threats

- [Langflow IP spoofing in MCP configuration installation](https://junglewise.ai/threats/cve-2026-105741-langflow-is-a-tool-for-building-and-deploying-ai-powered-agents)
- [Langflow command injection in MCP server configuration](https://junglewise.ai/threats/cve-2026-105740-langflow-is-a-tool-for-building-and-deploying-ai-powered-agents)
- [Langflow path traversal and cross-user disclosure in MCP resources](https://junglewise.ai/threats/cve-2026-105699-langflow-is-a-tool-for-building-and-deploying-ai-powered-agents)
- [Langflow code injection in validate_code endpoint](https://junglewise.ai/threats/duplicate-advisory-title-authenticated-remote-code-execution-in-65a053a9)
- [langflow code injection in validate endpoint](https://junglewise.ai/threats/cve-2026-51886-langflow-ai-langflow-v1-9-3-is-affected-by-code-injection-the)