AI Agent | Breaking Cybersecurity News | The Hacker News

Security firm AIR built a fake AI agent skill, pushed it through a popular skill marketplace and an Instagram ad, and says it reached roughly 26,000 agents, including some on corporate accounts. Every skill security scanner the firm tested it against marked it safe. The payload was harmless by design: it collected the user's email address and did nothing else. The point was to show that none of the signals people lean on to trust a skill caught it: not the scanners, not the GitHub stars, not the open-source reputation. A skill is a bundle of instructions an agent loads into its own context and follows with roughly the authority of a user prompt. That trust is the whole problem, and it is the reason skill-scanning tools exist in the first place. The skill, named  brand-landingpage , claimed to build a landing page using Google's Stitch design tool, aimed squarely at non-technical users. To make it look credible, AIR went after two trust signals: GitHub stars an...

Earlier this month, I spoke at the Gartner Security & Risk Management Summit about a blind spot most security programs are still not accounting for - how attackers are circumventing AI security programs by using legacy infrastructure to hijack AI agents. AI adoption is moving faster than security programs can account for. Roughly 71% of organizations are piloting AI agents across their enterprise applications, and 31% have already moved them into production workflows. For this reason, organizations are legitimately pouring resources into securing AI workloads against model poisoning, prompt injection, data leakage, and other emerging threats. Yet this focus misses everything underneath the AI layer. Because an unpatched server, a misconfigured Active Directory permission, or a cached credential on a developer's machine are exposures that give attackers a direct route to everything your AI agents depend on - knowledge bases, cloud storage, Lambda functions, SaaS integrati...

Microsoft researchers have detailed an exploit chain, named  AutoJack , that turns an AI browsing agent into a delivery vehicle for remote code execution. Steer the agent to load an attacker's web page, and that page's JavaScript can reach a privileged local service on the same machine and spawn a process on the host. No credentials, no sign-in screen, and no further user interaction once the agent loads the page. The attacker only has to get the agent to open it, and a planted link, a URL field, or a prompt injection will do. The flaw sits in  AutoGen Studio , the open-source prototyping interface for Microsoft Research's AutoGen multi-agent framework. This is not a bug that hits everyone who installs the package, and the packaging detail is worth getting right. A plain pip install autogenstudio pulls the current stable release, 0.4.2.2, the build Microsoft inspected, and it has no Model Context Protocol (MCP) route at all. That is the basis for Microsoft...

Cybersecurity researchers have disclosed details of three now-patched security flaws impacting LangGraph , including a critical vulnerability chain that could result in remote code execution. LangGraph is an open-source framework created by LangChain to build complex, stateful, and multi-agent artificial intelligence (AI) agentic applications. "An SQL injection in LangGraph's function could allow attackers to gain full control via remote code execution of a server by exploiting weaknesses in how the system processes and handles data," Check Point said . The list of identified vulnerabilities is as follows - CVE-2025-67644 (CVSS score: 7.3) - A SQL injection vulnerability exists in LangGraph's SQLite checkpoint implementation that allows attackers to manipulate SQL queries through metadata filter keys. (Affects langgraph-checkpoint-sqlite versions before 3.0.1) CVE-2026-28277 (CVSS score: 6.8) - An unsafe msgpack deserialization vulnerability in LangG...

Two security teams have shown, in separate research published this week, that OpenClaw , the popular self-hosted AI agent, can be driven to run attacker-controlled code or hand over sensitive data through ordinary-looking inputs. Imperva buried instructions inside shared contacts, vCards, and location pins that the agent executed without the victim ever seeing them. Varonis built a test agent on the platform, gave it a mailbox full of synthetic business data, and watched a single plain email talk it into forwarding mock AWS keys and a fake customer export to an outside address. The flaw Imperva found is patched in OpenClaw 2026.4.23, so update if you run it. The phishing weakness Varonis found is not something a patch fixes; it comes down to limiting what the agent can do on its own. Different doors into the same room: the agent trusts what reaches it, and its access becomes the attacker's. Hidden commands in a shared contact Imperva researcher Yohann Sillam looked at ...

The Fragmented State of Modern Enterprise Identity Enterprise IAM is approaching a breaking point. As organizations scale, identity becomes increasingly fragmented across thousands of applications, decentralized teams, machine identities, and autonomous systems. The result is Identity Dark Matter: identity activity that sits outside the visibility of centralized IAM and beyond the reach of security teams. According to Orchid Security 's analysis , 46% of enterprise identity activity occurs outside centralized IAM visibility. In other words, nearly half of the enterprise identity surface may be operating unseen. This hidden layer includes unmanaged applications, local accounts, opaque authentication flows, and over-permissioned non-human identities. It is further amplified by disconnected tools, siloed ownership, and the rapid rise of Agentic AI. The consequence is a widening gap between what the security organizations think they have and the access that actually exists. Th...

Microsoft has unveiled two new open-source tools called RAMPART and Clarity to assist developers in better testing the security of artificial intelligence (AI) agents. RAMPART , short for Risk Assessment and Measurement Platform for Agentic Red Teaming, functions as a Pytest-native safety and security testing framework for writing and running safety and security tests for AI agents, covering both adversarial and benign issues, as well as various harm categories. Users can write test cases to attack or probe an AI agent to explore possible safety violations like cross-prompt injections, where untrusted data reaches an AI system indirectly via a data source (e.g., email, file, or a web page) processed by it, or unintended behavioral regressions and data exfiltration. RAMPART then evaluates the outcome of those tests and reports the results. All it needs is an adapter that connects an agent to the test suite. The tool builds on PyRIT (short for Python Risk Identification Tool), ...

Threat actors have been observed attempting to exploit a recently disclosed security vulnerability in PraisonAI , an open-source multi-agent orchestration framework, within four hours of its public disclosure. The vulnerability in question is CVE-2026-44338 (CVSS score: 7.3), a case of missing authentication that exposes sensitive endpoints to anyone, potentially allowing an attacker to invoke the API server's protected functionality without a token.  " PraisonAI ships a legacy Flask API server with authentication disabled by default," according to an advisory released by the maintainers earlier this month. "When that server is used, any caller that can reach it can access /agents and trigger the configured agents.yaml workflow through /chat without providing a token." Specifically, the legacy Flask-based API server, src/praisonai/api_server.py, hard-codes AUTH_ENABLED = False and AUTH_TOKEN = None. According to PraisonAI, successful exploitation of the...

An administrative role meant for artificial intelligence (AI) agents within Microsoft Entra ID could enable privilege escalation and identity takeover attacks, according to new findings from Silverfort . Agent ID Administrator is a privileged built-in role introduced by Microsoft as part of its agent identity platform to handle all aspects of an AI agent's identity lifecycle operations in a tenant. The platform enables AI agents to authenticate securely and access necessary resources, as well as discover other agents. However, the shortcoming discovered by the identity security platform meant that users assigned the Agent ID Administrator role could take over arbitrary service principals , including those beyond agent-related identities, by becoming an owner and then add their own credentials to authenticate as that principal. "That's full service principal takeover," security researcher Noa Ariel said . "In tenants where high-privileged service principals...

AI agents are no longer just writing code. They are executing it. Tools like Copilot, Claude Code, and Codex can now build, test, and deploy software end-to-end in minutes. That speed is reshaping engineering—but it’s also creating a security gap most teams don’t see until something breaks. Behind every agentic workflow sits a layer few organizations are actively securing: Machine Control Protocols (MCPs) . These systems quietly decide what an AI agent can run, which tools it can call, which APIs it can access, and what infrastructure it can touch. Once that control plane is compromised or misconfigured, the agent doesn’t just make mistakes—it acts with authority. Ask the teams impacted by CVE-2025-6514 . One flaw turned a trusted OAuth proxy used by more than 500,000 developers into a remote code execution path. No exotic exploit chain. No noisy breach. Just automation doing exactly what it was allowed to do—at scale. That incident made one thing clear: if an AI agent can execute...

The AI browser wars are coming to a desktop near you, and you need to start worrying about their security challenges. For the last two decades, whether you used Chrome, Edge, or Firefox, the fundamental paradigm remained the same: a passive window through which a human user viewed and interacted with the internet. That era is over. We are currently witnessing a shift that renders the old OS-centric browser debates irrelevant. The new battleground is agentic AI browsers, and for security professionals, it represents a terrifying inversion of the traditional threat landscape. A new webinar dives into the issue of AI browsers , their risks, and how security teams can deal with them. Even today, the browser is the main interface for AI consumption; it is where most users access AI assistants such as ChatGPT or Gemini, use AI-enabled SaaS applications, and engage AI agents. AI providers were the first to recognize this, which is why we’ve seen a spate of new ‘agentic’ AI browsers bein...

Cybersecurity researchers have uncovered a jailbreak technique to bypass ethical guardrails erected by OpenAI in its latest large language model (LLM) GPT-5 and produce illicit instructions. Generative artificial intelligence (AI) security platform NeuralTrust said it combined a known technique called Echo Chamber with narrative-driven steering to trick the model into producing undesirable responses. "We use Echo Chamber to seed and reinforce a subtly poisonous conversational context, then guide the model with low-salience storytelling that avoids explicit intent signaling," security researcher Martí Jordà said . "This combination nudges the model toward the objective while minimizing triggerable refusal cues." Echo Chamber is a jailbreak approach that was detailed by the company back in June 2025 as a way to deceive an LLM into generating responses to prohibited topics using indirect references, semantic steering, and multi-step inference. In recent weeks, the...

Google on Tuesday revealed that its large language model (LLM)-assisted vulnerability discovery framework identified a security flaw in the SQLite open-source database engine before it could have been exploited in the wild. The vulnerability, tracked as CVE-2025-6965 (CVSS score: 7.2), is a memory corruption flaw affecting all versions prior to 3.50.2. It was discovered by Big Sleep , an artificial intelligence (AI) agent that was launched by Google last year as part of a collaboration between DeepMind and Google Project Zero. "An attacker who can inject arbitrary SQL statements into an application might be able to cause an integer overflow resulting in read off the end of an array," SQLite project maintainers said in an advisory. The tech giant described CVE-2025-6965 as a critical security issue that was "known only to threat actors and was at risk of being exploited." Google did not reveal who the threat actors were. "Through the combination of threa...