DEV Community

CVE Reports
CVE Reports

Posted on • Originally published at cvereports.com

GHSA-X2G5-FVC2-GQVP: GHSA-X2G5-FVC2-GQVP: Insufficient Bcrypt Salt Rounds in Flowise

GHSA-X2G5-FVC2-GQVP: Insufficient Bcrypt Salt Rounds in Flowise

Vulnerability ID: GHSA-X2G5-FVC2-GQVP
CVSS Score: Medium
Published: 2026-03-05

Flowise, an open-source low-code tool for LLM applications, contains a cryptographic weakness where user passwords were hashed using bcrypt with an insufficient work factor (salt rounds). Versions prior to 2.2.6 defaulted to 5 salt rounds, significantly below industry standards. This low computational cost allows attackers who obtain the database to crack password hashes via offline brute-force attacks at high speeds.

TL;DR

Flowise configured bcrypt with only 5 salt rounds, making password hashes trivial to crack if the database is compromised. Fixed in version 2.2.6 by increasing rounds to 10 and auto-upgrading hashes on login.


Technical Details

  • CWE ID: CWE-916
  • Weakness Name: Use of Password Hash with Insufficient Computational Effort
  • Attack Vector: Local / Physical (Database Access Required)
  • Impact: Credential Compromise
  • Severity: Medium
  • Default Salt Rounds: 5 (Vulnerable) vs 10 (Fixed)

Affected Systems

  • Flowise (npm package)
  • flowise: < 2.2.6 (Fixed in: 2.2.6)

Mitigation Strategies

  • Update Flowise to version 2.2.6 or later.
  • Rotate passwords for all administrative accounts.
  • Configure higher salt rounds via environment variables.

Remediation Steps:

  1. Stop the running Flowise instance.
  2. Pull the latest Docker image (flowiseai/flowise:latest) or update the npm package (npm install -g flowise).
  3. Restart the service.
  4. Log out and log back in with all administrative accounts to trigger the automatic hash upgrade logic.
  5. Optionally, set PASSWORD_SALT_HASH_ROUNDS=12 in your .env file for higher security.

References


Read the full report for GHSA-X2G5-FVC2-GQVP on our website for more details including interactive diagrams and full exploit analysis.

Top comments (0)