Skip to main content

Dify EUVDEUVD-2026-50488

| CVE-2026-18266 MEDIUM
URL Redirection to Untrusted Site (Open Redirect) (CWE-601)
2026-07-29 zdi GHSA-hvq6-f5p6-23f6
5.4
CVSS 3.0 · Vendor: zdi
View

Attack ChainAIDerived

Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain

  1. Access
    Craft OAuth URL with malicious redirect_url
  2. Delivery
    Deliver link via phishing to Dify user
  3. Exploit
    User authenticates against legitimate Dify instance
  4. Execution
    OAuth flow redirects user to attacker-controlled site
  5. Impact
    Attacker captures OAuth token or authorization code from redirect URI

Vulnerability Assessment

Weakness
CWE
OWASP Top 10:2025
A01:2025 - Broken Access Control
ASVS 5.0
V3 - Web Frontend Security Redirect URL validation
MITRE ATT&CK

1 technique across 1 tactic: Initial Access

The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up.

Exploitation Conditions

The target user must click a crafted OAuth link and complete the Dify authentication flow - passive exploitation without user interaction is not possible given UI:R in the CVSS vector. The vulnerability is triggered specifically within the OAuth flow handled by the AppInitializer component, meaning OAuth login must be enabled and in use on the Dify deployment. No attacker authentication is required (PR:N). Exploitation is limited by: (1) mandatory user interaction - the victim must actively follow the malicious link and authenticate; (2) OAuth must be configured and accessible on the target instance; (3) impact is bounded to information disclosure (token/credential leakage) rather than direct code execution or system compromise.

Impact

The CVSS 3.0 score of 5.4 (Medium) reflects a realistic assessment: network-accessible (AV:N), low complexity (AC:L), no attacker authentication required (PR:N), but mandatory user interaction (UI:R), with limited confidentiality and integrity impact (C:L/I:L) and no availability impact (A:N). No EPSS data was provided, and no KEV listing exists, so automated widespread exploitation is not confirmed. The primary real-world risk is targeted phishing: an attacker embeds the crafted OAuth URL in a spear-phishing email, the user authenticates against the legitimate Dify instance, and is then redirected to a malicious site. This threat model is particularly relevant for Dify deployments with third-party OAuth providers (e.g., Google, GitHub) where token leakage could grant broader access. The CPE wildcard ('*' for all versions) suggests no safe version exists yet, but the absence of a KEV entry and the UI:R requirement meaningfully constrain opportunistic mass exploitation.

Exploit ScenarioDerived

An attacker crafts a Dify OAuth authorization URL substituting the redirect_url parameter with a domain they control (e.g., https://attacker.com/capture). The attacker delivers this URL via phishing email or malicious web page to a Dify user. The user clicks the link, authenticates normally against the Dify instance, and is transparently redirected to the attacker's server - which logs the OAuth authorization code or access token appended to the redirect URL, enabling the attacker to impersonate the victim within the application.

This Month Plan within 30 days. Why: MEDIUM severity (CVSS 4.0-6.9).
No patch listed yet 0.2% exploitation probability (EPSS)

Get an e-mail when this CVE is exploited, patched or re-scored. Free account.

Alert me

AnalysisAIv1 – Jul 29, 2026

Open redirect in Dify's OAuth flow enables remote attackers to hijack post-authentication redirections, sending users to attacker-controlled sites that can harvest OAuth tokens or credentials. The flaw resides in the AppInitializer component's oauth_redirect_url handling, where the redirect destination is not validated against an allowlist. Exploitation requires user interaction - a target must follow a crafted link - but no authentication is required of the attacker, making phishing-based delivery straightforward. No public exploit code or active exploitation (CISA KEV) has been identified at time of analysis.

RemediationAI

No specific patched version has been confirmed from available data - consult the ZDI advisory at https://www.zerodayinitiative.com/advisories/ZDI-26-452/ and the langgenius Dify GitHub repository for an official fix. As an immediate compensating control, administrators should configure a strict redirect_url allowlist in the OAuth provider settings, restricting accepted redirect URIs to only the exact registered Dify instance URL - this prevents arbitrary external destinations from being accepted. Additionally, deploying a WAF rule or reverse-proxy filter that rejects OAuth callback requests whose redirect_url parameter points to external domains provides a network-layer mitigation; note this may break legitimate OAuth flows if misconfigured. Until a vendor patch is confirmed, disabling third-party OAuth login entirely (if operationally feasible) eliminates the attack surface at the cost of forcing local authentication.

Affected VersionsENISA

dify 1.11.4

Affected Packages

euvd dify no fix

Source: OSV (Open Source Vulnerabilities)

Affects your stack

Affects 1 package in euvd

Checking this CVE against your own stacks and servers is part of the Basic plan and up.

Technical ContextAI · show details

CWE-601 (URL Redirection to Untrusted Site / Open Redirect) describes the root cause: the application accepts a caller-controlled redirect_url parameter in its OAuth flow and forwards the user there after authentication without validating the destination against a trusted allowlist or origin check. The vulnerable component is the AppInitializer in Dify - an open-source LLM application development platform by langgenius (cpe:2.3:a:langgenius:dify:*:*:*:*:*:*:*:*, covering all versions). Because the flaw is in the OAuth handoff path, an attacker can supply a redirect_url pointing to an external domain, and after the user authenticates, the platform will silently redirect them off-site - potentially exposing OAuth authorization codes or tokens embedded in the redirect URI to the attacker's server logs.

Know when Dify gets its next vulnerability

  • Watch products and get an email when something new is published
  • Check this CVE against your own stack and servers
  • Exploit and PoC references, threat intelligence from 22 sources

No credit card · 7-day full trial

Severity by source

Vendor (zdi) PRIMARY 5.4 MEDIUM
AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:L/A:N
vuln.today AI 5.4 MEDIUM

Network-delivered via crafted URL (AV:N), no special conditions (AC:L), no attacker auth needed (PR:N), user must click link (UI:R); impact limited to token disclosure and misdirection with no availability consequence.

3.1 AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:L/A:N
4.0 AV:N/AC:L/AT:N/PR:N/UI:A/VC:L/VI:L/VA:N/SC:N/SI:N/SA:N

Primary rating from Vendor (zdi).

CVSS VectorVendor: zdi

Attack Vector
Network
Attack Complexity
Low
Privileges Required
None
User Interaction
Required
Scope
Unchanged
Confidentiality
Low
Integrity
Low
Availability
None

DescriptionCVE.org

Dify AI Workflow oauth_redirect_url Open Redirect Vulnerability. This vulnerability allows remote attackers to disclose sensitive information on affected installations of Dify. User interaction is required to exploit this vulnerability in that the target must visit a malicious page or open a malicious file.

The specific flaw exists within the OAuth flow handling in the AppInitializer component. An attacker can force a redirection to a site that serves malicious content. An attacker can leverage this vulnerability to disclose information in the context of the application. Was ZDI-CAN-29196.

Threat Intelligence Overview

EPSS Score
0.2%
Top 84% exploitable
Prevalence
Medium
popular open-source AI app platform, growing but not ubiquitous
Data Quality
5 sources
Confidence: Medium

How this decision was reached

The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.

SignalValue
Severity MEDIUM
CISA KEV No
Public POC No
EPSS 0.2%
Patch available No
Active threat No
Deciding rule
MEDIUM severity (CVSS 4.0-6.9)
What every action level means
Emergency Immediate action required. Criteria: CRITICAL + in CISA KEV (actively exploited) without patch, OR CRITICAL + EPSS > 30% without patch, OR CRITICAL + active threat signals without patch.
Act Now Prioritize within 24-48h. Criteria: in KEV with patch available, OR CRITICAL + public POC, OR HIGH + EPSS > 10%, OR CRITICAL + active threat, OR any CRITICAL severity.
This Week Schedule remediation this week. Criteria: HIGH severity (CVSS 7.0-8.9). Patch within normal change window.
This Month Plan within 30 days. Criteria: MEDIUM severity (CVSS 4.0-6.9), lower exploitation probability. Include in regular patch cycle.
Monitor Track and reassess. Criteria: LOW severity (CVSS 0.1-3.9), minimal exploitation risk. No immediate action needed.

SSVC AssessmentENISA

SSVC Decision· derived
Scheduled
Exploitation
None
System Exposure· derived
Open
Automatable
No
Human Impact· derived
Medium
Technical Impact
Partial

Priority Score

27
Low Medium High Critical
KEV: 0
EPSS: 0
CVSS: +27
PoC: 0
Weaponized: 0
Threat intel: 0
Gov advisory: 0

More in Dify

View all
CVE-2025-63387 HIGH POC
7.5 Dec 18

Dify 1.9.1 permits unauthenticated remote GET requests to the /console/api/system-features console endpoint, letting ano

CVE-2025-56157 CRITICAL POC
9.8 Dec 18

Hard-coded default PostgreSQL credentials shipped in the docker-compose.yaml of langgenius Dify through version 1.5.1 al

CVE-2026-41948 CRITICAL POC
9.3 May 18

Path traversal in Dify versions 0 through 1.14.1 allows authenticated tenants to escape their authorized tenant path and

CVE-2026-41947 CRITICAL POC
9.3 May 18

Cross-tenant authorization bypass in LangGenius Dify versions through 1.14.1 lets any logged-in editor reroute another t

CVE-2025-1796 HIGH POC
8.8 Mar 20

A vulnerability in langgenius/dify v0.10.1 allows an attacker to take over any account, including administrator accounts

CVE-2026-61461 HIGH POC
8.7 Jul 10

SQL injection in Dify's MyScale vector store backend (versions before 1.16.0-rc1) lets authenticated attackers execute a

CVE-2025-67732 HIGH POC
8.4 Jan 05

Plaintext exposure of third-party service API keys in Dify prior to version 1.11.0 allows any authenticated non-administ

CVE-2026-41949 HIGH POC
8.2 May 18

Cross-tenant document disclosure in Dify 1.14.1 and prior allows any authenticated user to read up to 3,000 characters o

CVE-2024-12039 HIGH POC
8.1 Mar 20

langgenius/dify version v0.10.1 contains a vulnerability where there are no limits applied to the number of code guess a

CVE-2024-12776 HIGH POC
8.1 Mar 20

In langgenius/dify v0.10.1, the `/forgot-password/resets` endpoint does not verify the password reset code, allowing an

CVE-2025-43862 HIGH POC
7.6 Apr 25

Dify is an open-source LLM app development platform. Rated high severity (CVSS 7.6), this vulnerability is remotely expl

CVE-2024-11824 HIGH POC
7.6 Mar 20

A stored cross-site scripting (XSS) vulnerability exists in langgenius/dify version latest, specifically in the chat log

This site uses cookies essential for authentication and security. No tracking or analytics cookies are used. Privacy Policy