CVE-2026-43534

OpenClaw has a vulnerability in input validation prior to version 2026.4.10, where external hook metadata can be enqueued as trusted system events. This allows attackers to supply malicious hook names to escalate untrusted input into higher-trust agent context. Affected software: OpenClaw (pre-20...

CVE-2026-43534 OpenClaw < 2026.4.10 - Unsanitized External Input in Agent Hook Events

OpenClaw before 2026.4.10 contains an input validation vulnerability that allows external hook metadata to be enqueued as trusted system events. Attackers can supply malicious hook names to escalate untrusted input into higher-trust agent context...

CVE-2026-43534 OpenClaw < 2026.4.10 - Unsanitized External Input in Agent Hook Events

OpenClaw before 2026.4.10 contains an input validation vulnerability that allows external hook metadata to be enqueued as trusted system events. Attackers can supply malicious hook names to escalate untrusted input into higher-trust agent context...

Automation-Exploit-Legacy

Automation-Exploit Legacy Prototype This repository contain...

SUSE CVE-2026-6948

Velociraptor versions prior to 0.76.4 contain a resource exhaustion vulnerability in the server's agent control channel. This allows a compromised or rogue Velociraptor client to crash the server via out-of-memory OOM by sending crafted messages through the normal client communication channel...

Redefining AI Red Teaming in the Agentic Era: From Weeks to Hours

AI systems are entering critical domains like healthcare, finance, and defense, yet remain vulnerable to adversarial attacks. While AI red teaming is a primary defense, current approaches force operators into manual, library-specific workflows. Operators spend weeks hand-crafting workflows -...

Generating Proof-Of-Vulnerability Tests to Help Enhance the Security of Complex Software

Developers create modern software applications Apps on top of third-party libraries Libs. When library vulnerabilities are reachable through application code, the applications can be vulnerable to software supply chain attacks. Prior work shows that developers often require concrete and executabl...

Medium: ecs-service-connect-agent

Issue Overview: Envoy is a cloud-native high-performance edge/middle/service proxy. A security vulnerability in Envoy allows external clients to manipulate Envoy headers, potentially leading to unauthorized access or other malicious actions within the mesh. This issue arises due to Envoy's defaul...

ARGUS: Defending LLM Agents against Context-Aware Prompt Injection

The rise of Large Language Model LLM agents, augmented with tool use, skills, and external knowledge, has introduced new security risks. Among them, prompt injection attacks, where adversaries embed malicious instructions into the agent workflow, have emerged as the primary threat. However,...

PT-2026-37309

Name of the Vulnerable Software and Affected Versions YetAnotherForum.NET YAF.NET versions prior to 4.0.5 YetAnotherForum.NET YAF.NET versions prior to 3.2.12 Description Stored Cross-Site Scripting XSS occurs when attacker-controlled input is persisted and later rendered without proper...

Medium: docker

Issue Overview: Moby is an open source container framework. Prior to version 29.3.1, a security vulnerability has been detected that allows plugins privilege validation to be bypassed during docker plugin install. Due to an error in the daemon's privilege comparison logic, the daemon may...

CVE-2026-7784 RTGS2017 NagaAgent Skills Endpoint extensions.py path traversal

A vulnerability has been found in RTGS2017 NagaAgent up to 5.1.0. This issue affects some unknown processing of the file apiserver/routes/extensions.py of the component Skills Endpoint. Such manipulation of the argument Name leads to path traversal. It is possible to launch the attack remotely. T...

GHSA-Q3JJ-46PQ-826R OpenClaw's ACP child sessions inherit subagent security envelope constraints

Summary ACP child sessions inherit subagent security envelope constraints. Affected Packages / Versions - Package: openclaw npm - Affected versions: = 2026.4.21 - Fixed version: 2026.4.22 Impact A restricted subagent spawning an ACP child session could fail to carry forward subagent-only...

CVE-2026-41686

Claude SDK for TypeScript provides access to the Claude API from server-side TypeScript or JavaScript applications. From version 0.79.0 to before version 0.91.1, the BetaLocalFilesystemMemoryTool in the Anthropic TypeScript SDK created memory files and directories using the Node.js default modes...

JLSEC-2026-431 When doing SSH-based transfers using either SCP or SFTP, and asked to do public key authentication,...

When doing SSH-based transfers using either SCP or SFTP, and asked to do public key authentication, curl would wrongly still ask and authenticate using a locally running SSH agent...

OPENSUSE-SU-2026:20669-1 Security update for google-cloud-sap-agent

This update for google-cloud-sap-agent fixes the following issue: - CVE-2026-34986: github.com/go-jose/go-jose/v4: processing of JWE object with empty encryptedkey field but key wrapping algorithm set can lead to a denial of service bsc1262936...

SUSE-SU-2026:21540-1 Security update for google-cloud-sap-agent

This update for google-cloud-sap-agent fixes the following issue: - CVE-2026-34986: github.com/go-jose/go-jose/v4: processing of JWE object with empty encryptedkey field but key wrapping algorithm set can lead to a denial of service bsc1262936...

Security Bulletin: IBM Instana Observability is affected by multiple vulnerabilities within Instana Agent container image

Summary Multiple vulnerabilities were addressed in IBM Observability with Instana within Instana Agent container image build 1.0.317 Vulnerability Details CVEID:CVE-2026-22184 DESCRIPTION: zlib versions up to and including 1.3.1.2 include a global buffer overflow in the untgz utility located unde...

Autonomous LLM Agent Worms: Cross-Platform Propagation, Automated Discovery and Temporal Re-Entry Defense

Autonomous LLM agents operate as long-running processes with persistent workspaces, memory files, scheduled task state, and messaging integrations. These features create a new propagation risk: attacker-influenced content can be written into persistent agent state, re-enter the LLM decision conte...

CVE-2026-6948

CVE-2026-6948 affects Velociraptor servers prior to version 0.76.4, where the VQLResponse Result-Set Writer can allocate memory unboundedly in the agent control channel. A compromised Velociraptor client can trigger an Out-Of-Memory (OOM) condition, crashing the server by sending crafted messages...