Intezer published a July 2026 vendor analysis arguing that AI-assisted attackers are forcing detection-engineering teams to move beyond static indicators and periodic rule tuning, requiring continuous…
On July 7, Noma Security disclosed a critical prompt injection flaw in GitHub's Agentic Workflows, dubbed GitLost, that allows any free GitHub account to extract private repository contents by posting…
Noma Labs disclosed GitLost, a July 2026 prompt-injection technique that could make GitHub Agentic Workflows leak private repository contents through a public issue comment. The attack exploits an arc…
GitHub's Agentic Workflows, which use AI agents to automate tasks from plain-English instructions, are vulnerable to the GitLost indirect prompt injection attack. Noma Security demonstrated that an at…
Researchers at Hong Kong University of Science and Technology introduced SkillCloak, a framework that demonstrates how malicious third-party coding-agent skills can evade static install-time scanners …
Google has been silently downloading its 4GB Gemini Nano AI model onto Chrome users' devices without consent since at least April, while a malicious Chrome extension disguised as Perplexity AI logged …
A flaw in Amazon Q Developer allowed malicious repositories to inject rogue Model Context Protocol (MCP) configurations, leading to arbitrary code execution. The vulnerability exploited the agent's tr…
Varonis Threat Labs disclosed SearchLeak, a three-stage vulnerability chain in Microsoft 365 Copilot Enterprise that can exfiltrate emails, MFA codes, calendar items, SharePoint and OneDrive files via…
Microsoft's Defender Security Research Team disclosed AutoJack on June 18, 2026, a chained exploit that achieves host-level remote code execution via AI browsing agents in AutoGen Studio pre-release b…
Enterprises are accumulating orphaned AI agents—autonomous tools left running after their creators depart—and standing privileges that increase attack surface, according to The Hacker News. Compromise…
Cybersecurity researchers at Tenet Security disclosed a new attack class called 'Agentjacking' that tricks AI coding agents into executing attacker-controlled code via maliciously crafted Sentry error…
A vulnerability chain in LangGraph, a widely deployed agentic AI framework, exposed self-hosted deployments to remote code execution. Attackers manipulated agent tool-calling behavior through prompt i…
A new class of attack called 'Agentjacking' targets AI coding agents by manipulating their tool-execution pipeline, exploiting the observation-action loop that makes them productive. The attack inject…
Tenet Security researchers disclosed a new supply-chain attack called "Agentjacking" that tricks AI coding agents into executing attacker-controlled code by injecting malicious error events into Sentr…
Check Point Research's June 2026 analysis of LangGraph and LangChain, building on Cyera Research's March 2026 findings, identified three critical vulnerabilities including a CVSS 9.3 deserialization f…
OpenAI has introduced a Lockdown Mode for ChatGPT that restricts connected tools and integrations to prevent data exfiltration, responding to evidence that LLM-connected tooling serves as a viable vec…
Independent researcher Buchodi found that free apps on smart-TV platforms from Samsung, LG, and others are embedding a Bright Data SDK that turns home devices into residential proxy nodes for web scra…
Microsoft Threat Intelligence identified a prompt-injection vulnerability in Anthropic's Claude Code GitHub Action that could expose CI/CD workflow secrets, according to the Microsoft Security Blog. S…
Security researchers at Flatt.tech disclosed a critical supply-chain vulnerability in Anthropic's Claude Code GitHub Actions integration that a single malicious GitHub issue, pull request, or comment …
An unknown threat actor exploited a public Marimo notebook via CVE-2026-39987 on May 10, 2026, using a large language model agent to drive post-compromise activity. The intruder extracted cloud creden…