CVE-2026-73487: Flowise Critical Vulnerability Leading to RCE and SSRF via Prompt Injection
Summary
CVE-2026-73487: Flowise Prompt Injection to RCE and SSRF via CSV/Airtable Agent Python Validator Bypass. Affects flowise versions < 3.1.3. Fixed in 3.1.3. How to fix ↓
Check if this vulnerability affects your application
Upload a manifest or SBOM and we'll show whether you're affected, plus every other risky package in the same scan.
No signup required · Results in about 60 seconds
Habilume
Habit tracker with a companion that grows with you
Complete habits, earn Lume Points, and watch your Lume grow as you stay consistent.
Background
Flowise is an open-source tool designed for building AI applications with a focus on ease of use and integration. It allows users to create chatflows that can process data from various sources, including CSV files and Airtable. However, versions up to 3.1.2 have a critical vulnerability that exposes users to remote code execution (RCE) and server-side request forgery (SSRF) attacks due to inadequate input validation in its Python code execution environment.
Vulnerability Detail
The vulnerability arises from the use of a regex-based blocklist in the validatePythonCodeForDataFrame() function, which fails to adequately sanitize user-generated Python code before execution. Attackers can exploit this weakness through prompt injection, allowing them to execute arbitrary code that can exfiltrate sensitive data or make unauthorized requests to internal services. Key technical aspects include the ability to bypass regex checks using functions like pd.read_json() and pd.read_csv() that accept URLs, as well as the lack of restrictions on Python's import capabilities.
How To Fix
Latest published version: 3.1.4 as of 22 Oct 2026
npm install flowise@3.1.4Upgrading to it may include breaking changes; the version below is the minimum fixed release for this vulnerability.
If you need the minimum change
3.1.3
This is the first release that contains the fix.
npm install flowise@3.1.3Tip — Not sure which version your app runs? Vulert can match this fix to it. Scan free →
Workaround
As an immediate workaround, users can block URL-accepting pandas functions by adding patterns for pd.read_json, pd.read_html, pd.read_csv, and pd.read_fwf in the validation logic. Network isolation for the Pyodide instance can also help mitigate risks.
Conclusion
The vulnerability identified as CVE-2026-73487 in Flowise poses significant risks, including RCE and SSRF attacks. Users are strongly advised to update to version 3.1.3 or later and consider implementing additional security measures to protect their applications.
See every vulnerability recorded for flowise →
FAQ
The CVSS score for this vulnerability is 9.0, indicating a critical severity level.
Flowise versions up to and including 3.1.2 are affected by this vulnerability.
You can use the Vulert playground at Vulert Scanner to check if your application is affected by this vulnerability without needing to sign up.
The potential impacts include full dataset theft, unauthorized access to internal services, and the ability to execute arbitrary code.
If you cannot update immediately, implement the suggested workarounds, such as blocking URL-accepting functions and isolating the execution environment.
Related vulnerabilities
Recent vulnerabilities recorded for flowise.
Habilume
Habit tracker with a companion that grows with you
Complete habits, earn Lume Points, and watch your Lume grow as you stay consistent.
