CVE-2026-61461
HighCVSS 8.8Exploitation Probability (EPSS)
Low risk28th percentile - higher than 28% of all known CVEs
Summary
A SQL injection vulnerability in Dify before version 1.16.0-rc1 in the MyScale vector store backend allows attackers to execute arbitrary SQL by supplying unsanitized search parameters to the search_by_full_text method without escaping or parameterization. Attackers can inject malicious SQL through the search parameters to read, modify, or delete data in the underlying ClickHouse database.
Risk Assessment
The organization is at risk of unauthorized access to data stored in the ClickHouse database, including data theft, modification, or permanent deletion, which could compromise the confidentiality, integrity, and availability of the system.
Recommendation
Immediately upgrade Dify to version 1.16.0-rc1 or later, which includes a fix for this vulnerability. Additionally, implement input validation and parameterization for all search parameters in query methods.
Other vulnerabilities in Dify
See all- CVE-2026-41948Critical
Dify version 1.14.1 and prior contain a path traversal vulnerability that allows authenticated users to manipulate requests forwarded to the Plugin Daemon's internal REST API. Attackers can exploit insufficient URL path sanitization to access internal endpoints.
- CVE-2026-41947Critical
Dify before version 1.14.2 contains an authorization bypass vulnerability that allows authenticated editor users to set and enable trace configurations for any application regardless of tenant ownership. This vulnerability allows attackers to redirect all messages and responses from victim applications to attacker-controlled LLM trace providers.
- CVE-2025-56157Critical
Dify up to version 1.5.1 includes default PostgreSQL credentials (username and password) in its docker-compose.yaml file. The vendor reports that PostgreSQL is not exposed on TCP port 5432 by default since version 1.0.1.
- CVE-2025-63388Critical
CORS misconfiguration vulnerability in Dify v1.9.1 in the /console/api/system-features endpoint. The endpoint reflects arbitrary Origin headers and sets Access-Control-Allow-Credentials: true, allowing any external domain to make authenticated cross-origin requests. NOTE: The supplier disputes this, arguing that credentialed requests do not provide additional access compared to unauthenticated requests.
- CVE-2026-18632Medium
A security flaw in langgenius dify up to 1.14.2, affecting the jinja2.Template function in api/core/helper/code_executor/jinja2/jinja2_transformer.py of the Jinja2 Handler component, results in improper neutralization of special elements used in a template engine. The attack can be launched remotely, and the exploit is publicly available.
- CVE-2026-18266Medium
An Open Redirect vulnerability in Dify AI Workflow's AppInitializer component. An attacker can redirect a user to a malicious site during OAuth flow, requiring user interaction (visiting a malicious page or file).
- CVE-2026-41949Medium
Dify before version 1.14.2 contains an authorization bypass vulnerability in the file preview endpoint that allows any authenticated user to read up to 3,000 characters of any uploaded document using only the file's UUID. Attackers can access the /console/api/files/{file_id}/preview endpoint to extract sensitive content from documents without ownership or workspace permission verification.
- CVE-2026-41950Medium
Dify before version 1.14.0 contains an authorization bypass vulnerability that allows authenticated users to read the full contents of files uploaded by other users within the same tenant by supplying an arbitrary file UUID in the chat-messages request.
Original NVD description (English source)
Dify before 1.16.0-rc1 contains a SQL injection vulnerability in the MyScale vector store backend that allows attackers to execute arbitrary SQL by supplying unsanitized search parameters to the search_by_full_text method without escaping or parameterization. Attackers can inject malicious SQL through the search parameters to read, modify, or delete data in the underlying ClickHouse database.