Hermes Agent 授权问题漏洞

Hermes Agent is an AI agent tool developed by Nous Research, featuring a self-learning mechanism. Version 0.8.0 of Hermes Agent contains an authorization vulnerability. This vulnerability arises from an unknown function in the Webhooks Endpoint component’s gateway/platforms/webhook.py file, which...

PT-2026-35519

The LatePoint – Calendar Booking Plugin for Appointments and Events plugin for WordPress is vulnerable to Privilege Escalation in versions up to and including 5.4.1. This is due to a missing authorization check in the execute method of the connect-customer-to-wp-user ability, which only requires...

PT-2026-35433

An issue in the /store/items/search endpoint of Agent Protocol server commit e9a89f allows attackers to cause a Denial of Service DoS via a crafted POST request...

Structured Security Auditing and Robustness Enhancement for Untrusted Agent Skills

Agent Skills package SKILL.md files, scripts, reference documents, and repository context into reusable capability units, turning pre-load auditing from single-prompt filtering into cross-file security review. Existing guardrails often flag risk but recover malicious intent inconsistently under...

PT-2026-35390

A vulnerability has been found in NousResearch hermes-agent 0.8.0. Affected by this vulnerability is the function check auth of the file gateway/platforms/api server.py of the component API SERVER KEY Handler. The manipulation leads to improper authentication. The attack can be initiated remotely...

Microsoft Entra Agent ID Flaw Enabled Tenant Takeover via Privilege Escalation

Microsoft Entra Agent ID flaw allowed privilege escalation and tenant takeover via Service Principal abuse, now fully patched by Microsoft...

smooth-pentest-agent

No d...

Constraint-Guided Multi-Agent Decompilation for Executable Binary Recovery

Decompilation -- recovering source code from compiled binaries -- is essential for security analysis, malware reverse engineering, and legacy software maintenance. However, existing decompilers produce code that often fails to compile or execute correctly, limiting their practical utility. We...

BrowserTools MCP 命令注入漏洞

BrowserTools MCP is an open-source browser monitoring and AI interaction tool developed by AgentDeskAI. Versions of BrowserTools MCP 1.2.0 and earlier contained a command injection vulnerability, which stemmed from the os command injection present in the browser-tools-server/browser-connector.ts...

CVE-2026-29181 vulnerabilities

Vulnerabilities for packages: ko-fips, crossplane-provider-aws-athena, crossplane-provider-aws-efs, gatus, kyverno, ipfs-cluster, beats-fips, cilium-cli, crossplane-provider-aws-kinesis, knative-net-istio-fips, crossplane-provider-aws-elbv2-fips, kubevirt-cdi-uploadserver, cloud-provider-vsphere,...

GHSA-MH2Q-Q3FH-2475 vulnerabilities

Vulnerabilities for packages: act, go-discover, kargo, osv-scanner, haproxy-ingress, hubble, net-kourier, emissary, consul-k8s, aws-privateca-issuer, db-operator, crossplane-provider-aws-elasticache, gitlab-pages, kubernetes-csi-node-driver-registrar, kubernetes-csi-driver-nfs, argo-rollouts,...

SUSE CVE-2009-1636

Multiple buffer overflows in the Internet Agent aka GWIA component in Novell GroupWise 7.x before 7.03 HP3 and 8.x before 8.0 HP2 allow remote attackers to execute arbitrary code via 1 a crafted e-mail address in an SMTP session or 2 an SMTP command...

[SECURITY] Fedora 44 Update: goose-1.23.2-8.fc44

Goose is your on-machine AI agent, capable of automating complex development tasks from start to finish. More than just code suggestions, goose can build entire projects from scratch, write and execute code, debug failures, orchestrate workflows, and interact with external APIs - autonomously...

Architecture Matters for Multi-Agent Security

Multi-agent systems MAS, composed of networks of two or more autonomous AI agents, have become increasingly popular in production deployments, yet introduce security risks that do not arise in single-agent settings. Even if individual agents exhibit robust security, architectural decisions...

Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents

Autonomous Large Language Model LLM agents are increasingly deployed to conduct complex tasks by interacting with external tools, APIs, and memory stores. However, processing untrusted external data exposes these agents to severe security threats, such as indirect prompt injection and unauthorize...

Security Bulletin: IBM Turbonomic Prometurbo agent used by IBM Turbonomic Application Resource Management is affected by a single vulnerability (CVE-2026-6389)

Summary IBM Turbonomic Prometurbo is an agent used by IBM Turbonomic Application Resource Management to integrate with Prometheus to collect application metrics and send them to Turbonomic for analysis and generation of optimization plans. A security vulnerability has been addressed in the IBM...

CVE-2026-41433 OpenTelemetry eBPF Instrumentation: Privileged Java agent injection allows arbitrary host file overwrite via untrusted TMPDIR

OpenTelemetry eBPF Instrumentation provides eBPF instrumentation based on the OpenTelemetry standard. From 0.4.0 to before 0.8.0, a flaw in the Java agent injection path allows a local attacker controlling a Java workload to overwrite arbitrary host files when Java injection is enabled and OBI is...

CVE-2026-41433

OpenTelemetry eBPF Instrumentation provides eBPF instrumentation based on the OpenTelemetry standard. From 0.4.0 to before 0.8.0, a flaw in the Java agent injection path allows a local attacker controlling a Java workload to overwrite arbitrary host files when Java injection is enabled and OBI is...

EUVD-2026-25620

OpenTelemetry eBPF Instrumentation provides eBPF instrumentation based on the OpenTelemetry standard. From 0.4.0 to before 0.8.0, a flaw in the Java agent injection path allows a local attacker controlling a Java workload to overwrite arbitrary host files when Java injection is enabled and OBI is...

CVE-2026-41433 OpenTelemetry eBPF Instrumentation: Privileged Java agent injection allows arbitrary host file overwrite via untrusted TMPDIR

OpenTelemetry eBPF Instrumentation provides eBPF instrumentation based on the OpenTelemetry standard. From 0.4.0 to before 0.8.0, a flaw in the Java agent injection path allows a local attacker controlling a Java workload to overwrite arbitrary host files when Java injection is enabled and OBI is...