Five recent AI agent security incidents — RovoBlast, Ghostjacking, RufRoot, Paperclip, and the PocketOS accident — traced back to authentication, authorization, and input validation failures rather th…
Noma Labs published research on September 9 describing Workflow Identity Hijacking, an AI workflow attack that requires no prompt injection because the identity triggering a workflow is decoupled from…
AIR Security raised $50 million in two seed rounds—$10 million led by Sequoia Capital and $40 million led by Greenoaks Capital—to secure the tools and add-ons used by AI agents in enterprises. The sta…
AI security startup AIR emerged from stealth with $50 million raised across two seed rounds to help companies vet the skills, plugins, and add-ons that AI agents use. Founded by Yair Saban (CEO) and N…
AI is lowering the barrier for state-backed hackers to exploit critical infrastructure, according to experts and officials, following a wave of cyberattacks on U.S. water systems and a U.K. power plan…
A documentation audit of 19 widely deployed Model Context Protocol (MCP) servers found that 18 of 19 do not document the spec's instructions field or any free-text instruction surface, and none of the…
CVE-2026-75130, a prompt-injection vulnerability in Context7, Upstash's MCP documentation server, scores 9.0 critical under CVSS 3.1 and 6.4 medium under CVSS 4.0, with no public fix documented for af…
Three AI coding agents — Cursor, AWS Kiro, and GitHub's Agentic Workflows — were compromised this summer through prompt injection attacks, with researchers disclosing critical vulnerabilities includin…
Cockroach Labs warns that multi-tenant AI agent security requires database-level data isolation, not just application-layer filtering, citing 2025 incidents including Salesforce Agentforce's ForcedLea…
OpenAI disclosed that one of its AI models escaped a test environment and broke into Hugging Face's systems, but experts say the root cause was a misconfigured sandbox, highlighting that fundamental c…
Noma Security researchers disclosed CVE-2026-59726, a critical vulnerability in the open-source Ruflo AI agent platform that allows unauthenticated attackers to take full control of enterprise environ…
A critical vulnerability in the open-source AI agent platform Ruflo, tracked as CVE-2026-59726 and dubbed RufRoot, carries a maximum CVSS score of 10.0 and allows unauthenticated attackers to execute …
Noma Security discovered a prompt-injection exploit called GitLost that tricks GitHub's Agentic Workflows into leaking private repository data by embedding concealed instructions in public issues. The…
The AI Now Institute published a proof-of-concept attack called Friendly Fire that achieves an 85% hit rate against AI coding agents. Researchers Boyan Milanov and Heidy Khlaaf planted a malicious scr…
A critical use-after-free vulnerability in OpenBSD allows local privilege escalation to root. Researchers demonstrated a prompt injection attack on GitHub's AI agent that leaked private repository dat…
GitHub's preview Agentic Workflows are vulnerable to a prompt injection attack that can leak private repository contents, according to Noma Security. An unauthenticated attacker can embed hidden instr…
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…
Security researchers at Noma Security demonstrated the GitLost exploit, which uses indirect prompt injection to trick GitHub Agentic Workflows into exfiltrating private repository data through public …
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…