PYSEC-2026-2706 Open WebUI Vulnerable to Unauthenticated RAG Configuration Disclosure

Vulnerability Type: Information Disclosure / Missing Authentication Severity: Medium Component: backend/openwebui/routers/retrieval.py — getstatus GET / Affected Endpoint: GET /api/v1/retrieval/ Affected Version: Open WebUI main branch — confirmed unpatched through v0.9.2 Authentication Required:...

CVE-2026-15535 AkariAsai self-rag retrieval_lm index.py Indexer.deserialize_from deserialization

A vulnerability was determined in AkariAsai self-rag up to 1fcdc420e48f50a7d7ab1ece5494221b93252e99. Affected by this issue is the function Indexer.deserializefrom of the file retrievallm/src/index.py of the component retrievallm. Executing a manipulation of the argument indexmeta.faiss can lead ...

CVE-2026-15535

The CVE concerns AkariAsai self-rag’s retrieval_lm component. Affected code: file retrieval_lm/src/index.py, function Indexer.deserialize_from, where manipulating the argument index_meta.faiss can trigger deserialization. This may be exploitable remotely; public exploit details exist. No specific...

PT-2026-57259

Deloitte AI Assist for Customer exposed unauthenticated API endpoints that allowed an attacker with knowledge of additional parameters to read from or inject content into the retrieval-augmented generation RAG corpus. On 2026-03-25, AI Assist for Customer restricted network access and enforced...

Langflow < 1.9.2 Multiple Vulnerabilities

The version of Langflow installed on the remote host is prior to 1.9.2. It is, therefore, affected by multiple vulnerabilities: - The Shareable Playground feature enables execution of workflows by unauthenticated users. When the route /api/v1/buildpublictmp executes a flow, it allows providing...

CVE-2026-54602

FastGPT is a knowledge-based AI application platform. Prior to 4.15.0, GET /api/core/ai/record/getRecord authenticates the caller but loads LLM request and response traces only by requestId without team scoping, allowing any authenticated user to read another team's prompts, retrieved RAG chunks,...

PYSEC-2026-1739 Open WebUI Vulnerable to Cross-Site Request Forgery (CSRF)

In version v0.3.8 of open-webui/open-webui, sensitive actions such as deleting and resetting are performed using the GET method. This vulnerability allows an attacker to perform Cross-Site Request Forgery CSRF attacks, where an unaware user can unintentionally perform sensitive actions by simply...

CVE-2026-59098 LobeChat 2.2.9 - Cross-User Document Disclosure via Unscoped RAG Semantic Search

LobeChat through 2.2.9 contains a broken access control vulnerability in the retrieval-augmented-generation semantic search functionality that allows authenticated attackers to access other users' data by exploiting missing user-identifier predicates in the chunk model semanticSearch method...

CVE-2026-55447

Langflow is a tool for building and deploying AI-powered agents and workflows. Prior to 1.9.2, by controlling a files that are digested into the RAG, an attacker can direct the node to read any file on the file-system by absolute path. All components based on BaseFileComponent are vulnerable to t...

CVE-2026-54019 Open WebUI: RAG ACL Bypass in Milvus Multitenancy Mode

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.6, Open WebUI added collection-level ACL checks, but the patch can still be bypassed when Milvus multitenancy mode is enabled. The ACL allows unknown non-KB collection names as...

CVE-2026-55447

Langflow is a tool for building and deploying AI-powered agents and workflows. Prior to 1.9.2, by controlling a files that are digested into the RAG, an attacker can direct the node to read any file on the file-system by absolute path. All components based on BaseFileComponent are vulnerable to t...

GHSA-P5CP-R7RG-QPXC Open WebUI: RAG ACL Bypass in Milvus Multitenancy Mode

RAG ACL Bypass in Milvus Multitenancy Mode Summary This is a bypass of the fix for: - GHSA-h36f-rqpx-j5wx - CVE-2026-44560 - "Unauthorized File and Knowledge Base Content Access via RAG Vector Search" Open WebUI added collection-level ACL checks, but the patch can still be bypassed when Milvus...

Open WebUI: RAG ACL Bypass in Milvus Multitenancy Mode

RAG ACL Bypass in Milvus Multitenancy Mode Summary This is a bypass of the fix for: - GHSA-h36f-rqpx-j5wx - CVE-2026-44560 - "Unauthorized File and Knowledge Base Content Access via RAG Vector Search" Open WebUI added collection-level ACL checks, but the patch can still be bypassed when Milvus...

Malicious code in @mastra/rag (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 031ec31035fa78e914e64884be3019150c8f1f77b9c464bce74e63cb0cd13227 The package was found to contain malicious code or consuming dependency that contains malicious code Source: ghsa-malware...

MAL-2026-5961 Malicious code in @mastra/rag (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 031ec31035fa78e914e64884be3019150c8f1f77b9c464bce74e63cb0cd13227 The package was found to contain malicious code or consuming dependency that contains malicious code Source: ghsa-malware...

PT-2026-50591

Name of the Vulnerable Software and Affected Versions Open WebUI versions prior to 0.9.6 Description An issue exists in the self-hosted artificial intelligence platform where collection-level Access Control List ACL checks can be bypassed when Milvus multitenancy mode is enabled. The ACL permits...

CVE-2026-45312 RAGFlow: Server-Side Template Injection in Prompt Generator leads to Remote Code Execution

RAGFlow is an open-source RAG Retrieval-Augmented Generation engine. In 0.24.0 and earlier, a Jinja2 template injection in the prompt generator rag/prompts/generator.py allows any authenticated user to execute arbitrary OS commands on the server. Any normal user can register, create a Canvas...

CVE-2026-21836

The HCL DominoIQ RAG feature is affected by a Broken Access Control vulnerability. Under certain circumstances, document level access restrictions will be ignored when determining what data to return from an AI query. This could enable an authenticated attacker to view sensitive data...

CVE-2026-21836

The HCL DominoIQ RAG feature is affected by a Broken Access Control vulnerability. Under certain circumstances, document level access restrictions will be ignored when determining what data to return from an AI query. This could enable an authenticated attacker to view sensitive data...

CVE-2026-21836

The HCL DominoIQ RAG feature is affected by a Broken Access Control vulnerability. Under certain circumstances, document level access restrictions will be ignored when determining what data to return from an AI query. This could enable an authenticated attacker to view sensitive data...