Symantec Honeypot Captures Skynet LLMjacking Agent With 74 Attack Tools
A Symantec honeypot caught an autonomous agent built to hijack exposed AI models: 74 attack tools, a 100,000-step budget and a hidden XMRig miner. Nothing ran.
· 5 minDesk · AI systems and model risk
Coverage of AI application security, model abuse, agentic systems, and data exposure.
A Symantec honeypot caught an autonomous agent built to hijack exposed AI models: 74 attack tools, a 100,000-step budget and a hidden XMRig miner. Nothing ran.
· 5 minRapid7 added six Langflow modules to Metasploit, including CVSS 10.0 CVE-2026-10134. Upgrade to 1.12.3 or later and take Langflow off the open internet.
· 4 minBitsight found MONST3R-AI, a pentest agent that runs on stolen OpenAI, Anthropic and Google API keys, as 800+ fake AI-skill and MCP repos push SmartLoader.
· 5 minSix CVEs in DB-GPT, Langchain-Chatchat, EvoScientist, LangAlpha and OpenSRE let unauthenticated callers or injected text reach a shell. Only one has a fix.
· 6 minNorForge found 20+ exposed panels of Cairn, an AI intrusion harness three criminal clusters use on retailers: about 10% success at roughly $5 a task.
· 4 minCheck Point chains two LiteLLM proxy flaws: CVE-2026-35029 takes any login to admin and code execution, CVE-2026-59823 redirects calls. Upgrade to 1.83.9.
· 4 minFive new CVEs in the official MCP TypeScript SDK leak messages across clients and expose tasks. Upgrade to 1.32.0 or 2.3.0, then change two settings.
· 5 minSGLang's ZMQ decoder still unpickles attacker data with pickle IPC disabled. Multi-node DP-attention clusters are open to unauthenticated RCE; no fix yet.
· 3 min