2026-05 Security and Quality Rollup for .NET Framework 3.5, 4.6.2, 4.7, 4.7.1, 4.7.2, 4.8 for Windows Server 2012 for x64 (KB5088860)

2026-05 Security and Quality Rollup for .NET Framework 3.5, 4.6.2, 4.7, 4.7.1, 4.7.2, 4.8 for Windows Server 2012 for x64 KB5088860...

2026-05 Cumulative Update for Windows 10 Version 1809 for x86-based Systems (KB5087538)

2026-05 Cumulative Update for Windows 10 Version 1809 for x86-based Systems KB5087538...

2026-05 .NET 8.0.27 Security Update for x64 Client (KB5093447)

2026-05 .NET 8.0.27 Security Update for x64 Client KB5093447...

2026-05 Cumulative Update for Windows 10 Version 21H2 for x86-based Systems (KB5087544)

2026-05 Cumulative Update for Windows 10 Version 21H2 for x86-based Systems KB5087544...

Langflow 路径遍历漏洞

Langflow is an open-source visualization framework developed by Langflow for building multi-agent and RAG applications. Versions of Langflow prior to 1.9.0 contained a path traversal vulnerability. This vulnerability stemmed from the path traversal in the knowledge base API, which could allow...

VulTriage: Triple-Path Context Augmentation for LLM-Based Vulnerability Detection

Automated vulnerability detection is a fundamental task in software security, yet existing learning-based methods still struggle to capture the structural dependencies, domain-specific vulnerability knowledge, and complex program semantics required for accurate detection. Recent Large Language...

When LLMs Team Up: A Coordinated Attack Framework for Automated Cyber Intrusions

Automated intrusion-style workflows require LLM agents to reason over partial observations, tool outputs, and executable artifacts under bounded budgets. A single LLM instance often compresses evidence extraction, planning, execution, and validation into one context, which increases the risk of...

PraisonAI knowledge-store backends interpolate unvalidated collection names into SQL and CQL queries

Summary PraisonAI exposes optional SQL/CQL-backed knowledge-store implementations that build table and index identifiers from unvalidated name and collection arguments. Applications that pass untrusted collection names into these backends can trigger SQL or CQL injection. Details This issue affec...

Oracle Poisoning: Corrupting Knowledge Graphs to Weaponise AI Agent Reasoning

We define Oracle Poisoning, an attack class in which an adversary corrupts a structured knowledge graph that AI agents query at runtime via tool-use protocols, causing incorrect conclusions through correct reasoning. Unlike prompt injection, Oracle Poisoning manipulates the data agents reason ove...

GHSA-H36F-RQPX-J5WX Open WebUI has Unauthorized File and Knowledge Base Content Access via RAG Vector Search

Unauthorized File and Knowledge Base Content Access via RAG Vector Search Affected Component RAG source resolution in chat completion pipeline: - backend/openwebui/retrieval/utils.py lines 963-965, 1063-1068, 1126-1131 in getsourcesfromitems Affected Versions Current main branch commit 6fdd19bf1...

Open WebUI has Unauthorized File and Knowledge Base Content Access via RAG Vector Search

Unauthorized File and Knowledge Base Content Access via RAG Vector Search Affected Component RAG source resolution in chat completion pipeline: - backend/openwebui/retrieval/utils.py lines 963-965, 1063-1068, 1126-1131 in getsourcesfromitems Affected Versions Current main branch commit 6fdd19bf1...

Missing Authorization

Overview open-webui is an Open WebUI Affected versions of this package are vulnerable to Missing Authorization via the validatecollectionaccess function. An attacker can obtain sensitive metadata, such as IDs, names, and descriptions of all knowledge bases across users, by sending crafted API...

Open WebUI vulnerable to Global Knowledge Base Enumeration via knowledge-bases Meta-Collection

Global Knowledge Base Enumeration via knowledge-bases Meta-Collection Affected Component Retrieval collection access validation: - backend/openwebui/routers/retrieval.py lines 2330-2355, validatecollectionaccess - backend/openwebui/routers/retrieval.py query endpoints, e.g. POST /query/doc Affect...

GHSA-6C2X-GCP3-GP73 Open WebUI vulnerable to Global Knowledge Base Enumeration via knowledge-bases Meta-Collection

Global Knowledge Base Enumeration via knowledge-bases Meta-Collection Affected Component Retrieval collection access validation: - backend/openwebui/routers/retrieval.py lines 2330-2355, validatecollectionaccess - backend/openwebui/routers/retrieval.py query endpoints, e.g. POST /query/doc Affect...

Missing Authorization

Overview open-webui is an Open WebUI Affected versions of this package are vulnerable to Missing Authorization in the processweb and processyoutube endpoints when the overwrite parameter is set to true and the collectionname is attacker-controlled. An attacker can overwrite or delete another user...

Open WebUI has Knowledge Base Destruction and RAG Poisoning via Unauthorized Collection Overwrite

Knowledge Base Destruction and RAG Poisoning via Unauthorized Collection Overwrite Affected Component Retrieval web/YouTube processing endpoints: - backend/openwebui/routers/retrieval.py lines 1810-1837, processweb - backend/openwebui/routers/retrieval.py the parallel processyoutube endpoint -...

GHSA-7R82-QHG4-6WVJ Open WebUI has Knowledge Base Destruction and RAG Poisoning via Unauthorized Collection Overwrite

Knowledge Base Destruction and RAG Poisoning via Unauthorized Collection Overwrite Affected Component Retrieval web/YouTube processing endpoints: - backend/openwebui/routers/retrieval.py lines 1810-1837, processweb - backend/openwebui/routers/retrieval.py the parallel processyoutube endpoint -...

CVE-2026-44337

PraisonAI is a multi-agent teams system. From version 2.4.1 to before version 4.6.34, PraisonAI exposes optional SQL/CQL-backed knowledge-store implementations that build table and index identifiers from unvalidated name and collection arguments. Applications that pass untrusted collection names...

CVE-2026-44337

PraisionAI across versions 2.4.1–4.6.34 exposes optional SQL/CQL-backed knowledge-store backends that derive table and index identifiers from unvalidated collection names. This can enable SQL/CQL injection when applications pass untrusted collection names into these backends. The issue is fixed i...

CVE-2026-44337 PraisonAI knowledge-store backends interpolate unvalidated collection names into SQL and CQL queries

PraisonAI is a multi-agent teams system. From version 2.4.1 to before version 4.6.34, PraisonAI exposes optional SQL/CQL-backed knowledge-store implementations that build table and index identifiers from unvalidated name and collection arguments. Applications that pass untrusted collection names...