CVE-2026-105699: CWE-639: Authorization Bypass Through User-Controlled Key in langflow-ai langflow
Description
A vulnerability in langflow versions 1.6.8 through 1.9.0 allows authorization bypass via a user-controlled key. An attacker with access to any project-scoped MCP endpoint could access another user's private flow-backed files without proper authorization. The issue exposes uploaded documents, structured data, prompts, and other private flow artifacts but does not allow modification of victim files or flows. This vulnerability is fixed in version 1.9.1.
CVSS v4.0
Score 7.1high
Affected software
langflow-ai
langflow
pkg:github/langflow-ai/langflowRun on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
Weaknesses
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
Langflow versions from 1.6.8 up to and including 1.9.0 contain an authorization bypass vulnerability (CWE-639) where the project-scoped MCP connection authenticated the project identifier but failed to authorize the resource URI supplied to resources/read. The read_resource function forwarded an attacker-controlled URI to handle_read_resource, which parsed flow_id and filename and called storage_service.get_file without verifying ownership or project association. This allowed users with access to any project-scoped MCP endpoint to retrieve private flow-backed files belonging to other users. Additionally, global handlers could disclose flow and file identifiers, facilitating targeted attacks. The vulnerability exposes private data but does not permit modification of victim files or flows. The issue is resolved in version 1.9.1.
Potential Impact
An attacker with access to any project-scoped MCP endpoint can read private flow-backed files of other users, including uploaded documents, structured data, and prompts. This leads to unauthorized disclosure of sensitive information across tenants. There is no impact on data integrity as victim files and stored flows cannot be modified through this vulnerability.
Mitigation Recommendations
Upgrade langflow to version 1.9.1 or later, where this authorization bypass vulnerability is fixed. No other mitigation actions are indicated by the vendor advisory.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- GitHub_M
- Date Reserved
- 2026-10-05T17:48:58.627Z
- Cvss Version
- 4.0
- State
- PUBLISHED
Threat ID: 6ac415412cdf04f6563b26ad
Added to database: 10/05/2026, 21:23:13 UTC
Last enriched: 10/05/2026, 21:33:27 UTC
Last updated: 10/05/2026, 21:33:27 UTC
Views: 8
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.