DeerFlow 安全漏洞

DeerFlow is an open-source orchestration framework developed by Bytedance, used to coordinate sub-agents and skill executions. DeerFlow has a security vulnerability, which stems from the bypass of agent name validation during the creation of custom agents in boot mode. This vulnerability may lead...

GHSA-3XX2-MQJM-HG9X Paperclip: Cross-tenant agent API key IDOR in `/agents/:id/keys` routes allows full victim-company compromise

Summary The GET, POST, and DELETE handlers under /agents/:id/keys in the Paperclip control-plane API only call assertBoardreq, which verifies that the caller has a board-type session but does not verify that the caller has access to the company owning the target agent. A board user whose membersh...

Paperclip: Cross-tenant agent API token minting via missing assertCompanyAccess on /api/agents/:id/keys

Isolated paperclip instance running in authenticated mode default config on a clean Docker image matching commit b649bd4 2026.411.0-canary.8, post the 2026.410.0 patch. This advisory was verified on an unmodified build. Summary POST /api/agents/:id/keys, GET /api/agents/:id/keys, and DELETE...

Securing AI Applications From Inception to Deployment

Extending the Wiz AI APP into the code layer to detect AI-specific risks at inception, validate exploitability at runtime, and orchestrate remediation with agents that understand your codebase...

RHEL 8 : resource-agents (RHSA-2026:1241)

The remote Redhat Enterprise Linux 8 host has packages installed that are affected by multiple vulnerabilities as referenced in the RHSA-2026:1241 advisory. The resource-agents packages provide the Pacemaker and RGManager service managers with a set of scripts. These scripts interface with severa...

AlmaLinux 8 : fence-agents (ALSA-2026:1240)

The remote AlmaLinux 8 host has packages installed that are affected by multiple vulnerabilities as referenced in the ALSA-2026:1240 advisory. urllib3: urllib3: Unbounded decompression chain leads to resource exhaustion CVE-2025-66418 urllib3: urllib3 Streaming API improperly handles highly...

RHEL 8 : fence-agents (RHSA-2026:1240)

The remote Redhat Enterprise Linux 8 host has packages installed that are affected by multiple vulnerabilities as referenced in the RHSA-2026:1240 advisory. The fence-agents packages provide a collection of scripts for handling remote power management for cluster devices. They allow failed or...

Towards Personalizing Secure Programming Education with LLM-Injected Vulnerabilities

According to constructivist theory, students learn software security more effectively when examples are grounded in their own code. Generic examples often fail to connect with students' prior work, limiting engagement and understanding. Advances in LLMs are now making it possible to automatically...

MCP Server Kubernetes 安全漏洞

MCP Server Kubernetes is a Kubernetes management server developed by Suyog Sonwalkar. Versions of MCP Server Kubernetes prior to 3.4.0 have security vulnerabilities. These vulnerabilities stem from parameter injection issues in the port-forward tool, which may lead to exposure of internal...

Challenges and Future Directions in Agentic Reverse Engineering Systems

Agentic systems built on large language models LLMs are increasingly being used for complex security tasks, including binary reverse engineering RE. Despite recent growth in popularity and capability, these systems continue to face limitations in realistic settings. Cutting-edge systems still fai...

User Impersonation

Overview Affected versions of this package are vulnerable to User Impersonation via the isHealthCheckRequest function in pkg/middleware/healthcheck.go. An attacker can reach protected endpoints by sending a request with a configured health-check User-Agent, causing the middleware to treat the...

The vulnerability of the Agents component in Google Chrome and Microsoft Edge browsers allows a hacker to trigger a service failure.

The vulnerability of the Agents component in Google Chrome and Microsoft Edge relates to the use of memory after it is freed. Exploiting this vulnerability can allow a malicious actor to cause service interruptions through a specially crafted HTML page...

CVE-2026-40160

PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, webcrawl's httpx fallback path passes user-supplied URLs directly to httpx.AsyncClient.get with followredirects=True and no host validation. An LLM agent tricked into crawling an internal URL can reach cloud metadata endpoints...

CVE-2026-40153

PraisonAIAgents is a multi-agent teams system. Prior to 1.5.128, the executecommand function in shelltools.py calls os.path.expandvars on every command argument at line 64, manually re-implementing shell-level environment variable expansion despite using shell=False line 88 for security. This...

What Is CAASM? Cyber Asset Attack Surface Management Explained

Your security team runs scans from five different tools. Each one gives you a different number of assets, a different count of vulnerabilities, and a different view of your risk. Meanwhile, your CMDB is outdated, shadow IT keeps expanding, and nobody can confidently answer a basic question: "What...

SIR-Bench: Evaluating Investigation Depth in Security Incident Response Agents

We present SIR-Bench, a benchmark of 794 test cases for evaluating autonomous security incident response agents that distinguishes genuine forensic investigation from alert parroting. Derived from 129 anonymized incident patterns with expert-validated ground truth, SIR-Bench measures not only...

EUVD-2026-21696

A vulnerability was identified in FoundationAgents MetaGPT up to 0.8.1. This affects the function generatethoughts of the file metagpt/strategy/tot.py of the component Tree-of-Thought Solver. The manipulation leads to code injection. It is possible to initiate the attack remotely. The exploit is...

EUVD-2026-21694

A vulnerability was determined in FoundationAgents MetaGPT up to 0.8.1. The impacted element is the function evaluateCode of the file metagpt/environment/minecraft/mineflayer/index.js of the component Mineflayer HTTP API. Executing a manipulation can lead to cross-site request forgery. The attack...

Beyond Static Sandboxing: Learned Capability Governance for Autonomous AI Agents

Autonomous AI agents built on open-source runtimes such as OpenClaw expose every available tool to every session by default, regardless of the task. A summarization task receives the same shell execution, subagent spawning, and credential access capabilities as a code deployment task, a 15x...

PT-2026-32144

Name of the Vulnerable Software and Affected Versions FoundationAgents MetaGPT versions up to 0.8.1 Description A security flaw exists in FoundationAgents MetaGPT versions up to 0.8.1. The decode image function within the metagpt/utils/common.py file is susceptible to server-side request forgery...