What changed in AI security, Apr 6 to Apr 12, 2026
Apr 6 to Apr 12, 2026 (ISO week 2026-W15). Weeks run Monday to Sunday in UTC.
199 records published, +24 on the previous week: 61 vulnerabilities (+2), 0 incidents (no change), 11 research items (-1), 126 news items (+23), 1 policy item (no change).
Critical and high advisories
Vulnerability records rated critical or high, newest first. Showing 25 of 30.- High
CVE-2026-6129: A vulnerability was detected in zhayujie chatgpt-on-wechat CowAgent up to 2.0.4. This affects an unknown function of…
CVE-2026-6129NVD/CVE Database - High
CVE-2026-6126: A weakness has been identified in zhayujie chatgpt-on-wechat CowAgent 2.0.4. The affected element is an unknown…
CVE-2026-6126NVD/CVE Database - High
GHSA-75hx-xj24-mqrw: n8n-mcp has unauthenticated session termination and information disclosure in HTTP transport
GitHub Advisory Database - Critical
GHSA-8x8f-54wf-vv92: PraisonAI Browser Server allows unauthenticated WebSocket clients to hijack connected extension sessions
GitHub Advisory Database - High
CVE-2026-40217: LiteLLM through 2026-04-08 allows remote attackers to execute arbitrary code via bytecode rewriting at the…
CVE-2026-40217NVD/CVE Database - High
CVE-2026-40150: PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, the web_crawl() function in…
CVE-2026-40150NVD/CVE Database - High
CVE-2026-40116: PraisonAI is a multi-agent teams system. Prior to 4.5.128, the /media-stream WebSocket endpoint in PraisonAI's call…
CVE-2026-40116NVD/CVE Database - High
CVE-2026-40113: PraisonAI is a multi-agent teams system. Prior to 4.5.128, deploy.py constructs a single comma-delimited string for the…
CVE-2026-40113NVD/CVE Database - Critical
CVE-2026-40111: PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, he memory hooks executor in praisonaiagents passes a…
CVE-2026-40111NVD/CVE Database - High
CVE-2026-39981: AGiXT is a dynamic AI Agent Automation Platform. Prior to 1.9.2, the safe_join() function in the essential_abilities…
CVE-2026-39981NVD/CVE Database - High
GHSA-7437-7hg8-frrw: OpenClaw: HGRCPATH, CARGO_BUILD_RUSTC_WRAPPER, RUSTC_WRAPPER, and MAKEFLAGS missing from exec env denylist — RCE via build tool env injection (GHSA-cm8v-2vh9-cxf3 class)
GitHub Advisory Database - Critical
GHSA-2763-cj5r-c79m: PraisonAI Vulnerable to OS Command Injection
GitHub Advisory Database - High
GHSA-4ggg-h7ph-26qr: n8n-mcp has authenticated SSRF via instance-URL header in multi-tenant HTTP mode
GitHub Advisory Database - High
CVE-2026-34724: Zammad is a web based open source helpdesk/customer support system. Prior to 7.0.1, a server-side template injection…
CVE-2026-34724NVD/CVE Database - Critical
GHSA-qf73-2hrx-xprp: PraisonAI has sandbox escape via exception frame traversal in `execute_code` (subprocess mode)
CVE-2026-39888GitHub Advisory Database - High
CVE-2026-3357: IBM Langflow Desktop 1.6.0 through 1.8.2 Langflow could allow an authenticated user to execute arbitrary code on the…
CVE-2026-3357NVD/CVE Database - High
GHSA-69x8-hrgq-fjj8: LiteLLM: Password hash exposure and pass-the-hash authentication bypass
GitHub Advisory Database - High
GHSA-8jxr-pr72-r468: Java-SDK has a DNS Rebinding Vulnerability
CVE-2026-35568GitHub Advisory Database - High
CVE-2026-24175: NVIDIA Triton Inference Server contains a vulnerability where an attacker could cause a server crash by sending a…
CVE-2026-24175NVD/CVE Database - High
CVE-2026-24174: NVIDIA Triton Inference Server contains a vulnerability where an attacker could cause a server crash by sending a…
CVE-2026-24174NVD/CVE Database - High
CVE-2026-24173: NVIDIA Triton Inference Server contains a vulnerability where an attacker could cause a server crash by sending a…
CVE-2026-24173NVD/CVE Database - High
CVE-2026-24146: NVIDIA Triton Inference Server contains a vulnerability where insufficient input validation and a large number of…
CVE-2026-24146NVD/CVE Database - High
CVE-2026-35486: text-generation-webui is an open-source web interface for running Large Language Models. Prior to 4.3, he superbooga…
CVE-2026-35486NVD/CVE Database - High
CVE-2026-35485: text-generation-webui is an open-source web interface for running Large Language Models. Prior to 4.3, an…
CVE-2026-35485NVD/CVE Database - High
CVE-2026-1839: A vulnerability in the HuggingFace Transformers library, specifically in the `Trainer` class, allows for arbitrary code…
CVE-2026-1839NVD/CVE Database
Exploitation signals
Vulnerabilities published in the week that are listed in the CISA Known Exploited Vulnerabilities catalog or have an EPSS score of 10% or more.No vulnerability published in this week is listed as exploited or has an EPSS score of 10% or more.
Packages that began delegating to a language model
Exposure Registry packages whose first release declaring an LLM SDK, agent framework or MCP dependency was published in the week.| Package | Ecosystem | LLM SDKs | Release | Released |
|---|---|---|---|---|
| b12x | PyPI | Hugging Face Hub / Transformers | 0.7.6 |
Topics that moved
Largest increases over the mean of the 4 previous weeks, for topics with at least 3 records in the week.| Topic | Records | Weekly mean, previous 4 | Difference |
|---|---|---|---|
| Inference infrastructure | 5 | 2.5 | +2.5 |
Research
Peer-reviewed first, then newest. Showing 8 of 11.A Formal Lens on Android Permissions System: Modeling, Verification, and Exploitation Using LLMs and Model Checking
Peer-reviewedACM Digital Library (TOPS, DTRAP, CSUR)Exploring Visual Explanations for Defending Federated Learning against Poisoning Attacks: Enhancing LayerCAM with Autoencoders
Peer-reviewedACM Digital Library (TOPS, DTRAP, CSUR)AICCE: AI-Driven Compliance Checker Engine
Peer-reviewedIEEE Xplore (Security & AI Journals)Quadruplet Augmentation With Attribute and Structure Invariance for Online Continual Learning
Peer-reviewedIEEE Xplore (Security & AI Journals)ReSLC: Defending backdoor attacks on intelligent vulnerability detection via redundant semantic LLM compression
Peer-reviewedElsevier Security JournalsXFaceMark: Explainable deep fake watermarking using YOLO, and random MRFO
Peer-reviewedElsevier Security JournalsSBOMs into Agentic AIBOMs: Schema Extensions, Agentic Orchestration and Reproducibility Evaluation
Peer-reviewedACM Digital Library (TOPS, DTRAP, CSUR)Nearest Neighbor Projection Removal Adversarial Training
Peer-reviewedIEEE Xplore (Security & AI Journals)
Policy and regulation
Newest first.Generated from the AI Sec Watch database at . Every item links to its record.