The AI security podcast. Everything worth knowing in AI security, LLM threats, and agentic AI risks — under 15 minutes. Curated by a human. Voiced by AI.
Reading the feed…
Top Story: OWASP put a name and a risk list on the layer nobody governs — A "skill" is packaged expertise you hand an AI agent: instructions and scripts that turn a general assistant into one that knows how to build a landing page, close a support ticket, or run a deployment. A backdoor that starts when code is loaded, not when it is installed. — Developers have been taught for years to watch what happens during installation, because that is where malicious packages historically fired. Rust's turn came the same week, one layer deeper. — Someone got into the account of a maintainer whose code s
Top Story: The firewall blocked it. The agent ran it. — Whoever controls a company's DNS controls where its web and email traffic goes. 🏛️ OpenAI moved the limit from the model to the door. — In the spring, OpenAI put its cyber-tuned GPT-5.5-Cyber model behind a vetted-access program. ⚔️ AI browsers still have no perfect fix. — At Black Hat, Brave Software security engineer Artem Chaikin found prompt-injection paths in all three browsers he demonstrated: Opera, Perplexity Comet, and ChatGPT Atlas. 🔬 Encrypted reasoning can leak what the final answer hides. — A new paper found that opaque rea
Top Story: The end of rare: what a vulnerability costs to find just collapsed — For thirty years, security has quietly been an economics argument wearing an engineering costume. The AI browser you installed can be taken over by a calendar invite, and the vendors do not agree on whether that is a flaw. — You install a browser that reads the web for you and acts with your permissions, so anything it reads can try to give it orders. An invisible sentence in a Word document escaped Microsoft 365 Copilot's cloud sandbox and reached the host underneath it. — Rubrik Zero Labs hid instructions behind
Top Story: Microsoft's Project Perception enters public preview today — Microsoft announced Project Perception on July 27, and it enters public preview inside Defender today. The Hugging Face intrusion ran through at least eight flaws in JFrog Artifactory, and there's a patch waiting. — Artifactory is the warehouse companies use to store and serve their software components. Hugging Face published how it investigated an AI-run attack, and the hardest part wasn't the attacker. — The intrusion started with a malicious dataset that abused two code-execution paths in the company's data-processing p
Top Story: The Reveal — On July 16, Hugging Face disclosed that something had broken into its production systems over the weekend of July 11-13, moved through internal servers for days, and left more than 17,000 recorded actions in its wake. A single link could plant a fake employee inside your company, with all their access. — Zenity Labs researcher Mike Takahashi found that ChatGPT's Agent Builder would silently follow instructions hidden in a link's web address the moment a logged-in victim clicked it, no further action needed. 💰 Funding — A stealth AI security startup surfaced already wor
Top Story: The Ransomware That Ran Itself — Sysdig's threat research team watched an AI agent break into an internet-facing instance of Langflow, a popular open-source AI workflow tool, through a flaw disclosed back in March (CVE-2025-3248) that many deployments still haven't patched. 🔬 Research: Check Point says defenders can no longer assume human pace. — Check Point's AI Security Report 2026 documents AI running full, multi-stage break-ins with barely any human direction over the past year, plus attackers planting hijack instructions in coding-agent config files (like CLAUDE.md) that reloa
Top Story: One Word Beat GitHub's Guardrail — An attacker doesn't need an account, a password, or a single line of malicious code. China flags a Claude Code "backdoor." — On July 8, China's national vulnerability database (run by the MIIT) warned that Anthropic's Claude Code versions 2.1.91 through 2.1.196 transmitted users' location and identity data back to Anthropic's servers, and urged users to uninstall or upgrade. CISA reportedly turns Mythos on its own code. — Reuters reports (sourced, not officially confirmed) that CISA's Attack Surface Evaluation team is running Anthropic's Mythos mod
Top Story: Anthropic Put Its Strongest Model Back Online. First It Had to Add a New Lock. — Three weeks ago, the US government took a commercial AI offline by order. Opening a repo with your AI assistant can hand your machine to a stranger. — Researchers at Mozilla's 0DIN group showed a clean-looking code project that compromises a developer's computer with no malicious code anywhere in it. A platform behind a million AI apps could let one customer's data leak to another. — Security firm Zafran found four flaws in Dify, a popular open-source tool for building AI apps, that let a low-privileged
Top Story: OpenAI Built Its Strongest Model Yet. Then Handed the Guest List to Washington — On Friday, June 26, OpenAI announced three new frontier models — Sol, its self-described "strongest model yet," plus Terra for everyday work and Luna as a cheaper option — and in the same breath said most people can't have them yet. The free, downloadable models are catching the paid frontier — fastest they ever have. — On the independent Artificial Analysis Intelligence Index, the leading open-weight model (Moonshot's Kimi K2.6) now ranks fourth overall and first among open models, about six points beh
Top Story: LiteLLM, Hit Again. This Time the Whole Gateway Falls — An AI "gateway" is the traffic controller that sits between your applications and the models they call. ⚔️ Attack: SearchLeak. One booby-trapped link could have turned Microsoft 365 Copilot into a silent data thief. Microsoft has already fixed it. — Varonis demonstrated the chain (they did not find it used in the wild): a crafted, Microsoft-hosted search link carried hidden instructions that Copilot read and obeyed, then quietly exfiltrated whatever the victim could access: emails, files, calendar. 🔬 Research: a guard for AI a
Top Story: A Directive at 5:21 PM, Two Frontier Models Gone by Morning — For the first time anyone can easily point to, the US government used export controls — the tool that governs missiles and advanced chips — to pull a deployed, commercial AI model off the market. The "Are you sure?" box in your AI coding assistant can lie about what you're approving. — Researchers at Adversa AI disclosed two flaws affecting popular AI coding tools. An attacker can turn a tool your AI agent trusts into a remote-control channel — without ever touching your infrastructure. — Tenet Security disclosed Agentjac
Top Story: Catching the Attack Isn't Enough Anymore — For two years, the standard defense against prompt injection — hiding malicious instructions in something an AI reads, so it mistakes them for orders — has leaned on a single hope: make the model smart enough to notice. Curated by Asaf Nakash. Voices by AI. Opinions by human. Show notes: https://contextwindowsec.com/episodes/2026-06-08.html
Top Story: Prompt Injection Goes Operational — For two years, prompt injection has mostly been a lab demo. ModelScope MS-Agent: a max-severity hole with no fix (CVE-2026-2256). — A command-injection flaw in Alibaba's widely used MS-Agent toolkit lets an attacker run arbitrary commands on the host running the agent (CVSS 9.8). LMDeploy: 13 hours from disclosure to exploitation (CVE-2026-33626). — Earlier this spring, a server-side request forgery flaw in the LMDeploy serving framework — think of it as tricking the server into making requests on the attacker's behalf — went from public advisory
Top Story: The Week Trust Broke Twice — Two stories landed in the same 72 hours that belong in the same frame. NVIDIA NemoClaw sandbox bypass (CVE-2026-24222). — Lasso Security demonstrated that AI agents running inside NVIDIA's NemoClaw/OpenShell sandbox can exfiltrate sensitive data through tools the sandbox explicitly allows. vm2 sandbox escape wave: 13 CVEs, CVSS 9.0–10.0. — Between May 4 and May 7, researchers disclosed 13 sandbox escape vulnerabilities in vm2, the popular Node.js library used to isolate untrusted JavaScript. Cisco: "Reading Between the Pixels" (multimodal prompt injectio
Top Story: The Race at Agentic Speed — Two things happened in the same week that belong in the same sentence. TeamPCP releases Shai-Hulud source code, launches BreachForums "supply chain challenge." — The group posted the complete worm framework to GitHub (since removed, but forked) with detailed deployment instructions, and announced a contest on BreachForums offering $1,000 in Monero to anyone who uses it to compromise open-source packages. TanStack CI cache poisoned, hitting OpenAI and Mistral AI. — A pull request from a throwaway fork (attributed to TeamPCP's ongoing supply-chain campaign)
Top Story: The Prompt Was the Payload — Two Agent-Framework RCEs in Seven Days — Two independent disclosures landed inside seven days, and they collapse to the same sentence: a model read an instruction it shouldn't have trusted, and a tool downstream did exactly what the parsed text said. Cisco announces intent to acquire Astrix Security. — Cisco's May 4 blog post by SVP Peter Bailey says Astrix will fold into Cisco Identity Intelligence, Cisco Secure Access, Duo IAM, and Splunk. An X user drained ~$150,000 from a Grok-linked Bankr wallet via Morse-encoded prompt injection (May 4, 2026). — Th
Top Story: TeamPCP Returns — "Mini Shai-Hulud" Hits Two Ecosystems Simultaneously — After a 26-day pause, TeamPCP is back. Curator's Corner: When Trust Is the Exploit Curated by Asaf Nakash. Voices by AI. Opinions by human. Show notes: https://contextwindowsec.com/episodes/2026-05-04.html
Top Story: MCP STDIO RCE — The Connector Layer Has an Authority Problem — On April 23, the Cloud Security Alliance — an independent industry research body — and OX Security, an established Israeli software-supply-chain security vendor (founded 2021, $34M seed from Insight Partners and Team8), jointly disclosed an architectural vulnerability in the Model Context Protocol's STDIO transport — the most common transport used by local MCP servers across the open-source agent ecosystem. Curator's Corner: Three Layers, Three Attack Surfaces, One Agent Curated by Asaf Nakash. Voices by AI. Opinions by
Top Story: Comment and Control — Three Coding Agents, One Bug Class, Zero CVEs — On April 15, researcher Aonan Guan — working with Johns Hopkins University's Zhengyu Liu and Gavin Zhong — published the first cross-vendor demonstration of a prompt-injection pattern that turns GitHub itself into the command-and-control channel for stealing runner credentials out of AI coding agents. Curator's Corner: Every Consultancy Is a Honey Pot Now Curated by Asaf Nakash. Voices by AI. Opinions by human. Show notes: https://contextwindowsec.com/episodes/2026-04-20.html
Top Story: Claude Mythos Preview + Project Glasswing — AI Reaches the Zero-Day Threshold — On April 7, Anthropic announced Claude Mythos Preview alongside Project Glasswing — the most consequential AI security development of 2026, and arguably of the decade.. Curator's Corner: Anthropic's Oppenheimer Moment Curated by Asaf Nakash. Voices by AI. Opinions by human. Show notes: https://contextwindowsec.com/episodes/2026-04-13.html
Anthropic accidentally published Claude Code's entire source code to npm — 512,000 lines of TypeScript, including an autonomous daemon called KAIROS that nobody was supposed to know about. North Korea compromised the Axios npm package through AI-assisted social engineering. Mercor, a $10B AI startup, got breached via the LiteLLM supply chain — 4TB exfiltrated. Plus: Microsoft open-sources the Agent Governance Toolkit, and Curator's Pick on why instructions are not guardrails. Curated by Asaf Nakash. Voices by AI. Opinions by human. Show notes: https://contextwindowsec.com/episodes/2026-04-06.h
Your vulnerability scanner just published malware. One threat actor — TeamPCP — hit five ecosystems in ten days: Trivy, Checkmarx KICS, LiteLLM, Telnyx, and npm via CanisterWorm. A supply chain worm that completes a full compromise cycle in under sixty seconds. Plus: RSA Conference drops its agent security agenda, Claude gets jailbroken, and Curator's Corner on why security tools became the attack surface. Curated by Asaf Nakash. Voices by AI. Opinions by human. Show notes: https://contextwindowsec.com/episodes/2026-03-30.html
Two major AI platform sandbox escapes dropped this week on the eve of RSA Conference. AWS Bedrock's "isolated" sandbox leaks DNS queries — researchers built a full reverse shell. Snowflake's Cortex Code CLI got jailbroken through a GitHub README. Plus: MCP rug pulls, VoidLink (88K lines of AI-generated malware), the biggest pre-RSA funding window ever, and Curator's Corner on why AI didn't create new vulnerabilities — it made old ones affordable. Curated by Asaf Nakash. Voices by AI. Opinions by human. Show notes: https://contextwindowsec.com/episodes/2026-03-23.html
A red-team agent compromised a Big Four consultancy's customer-facing chatbot in under two hours. The Chrome Gemini hijack lets browser extensions take over Google's built-in AI. Plus the biggest M&A week in AI security history — Google closed the $32B Wiz deal, OpenAI bought Promptfoo, and three stealth startups emerged with a quarter-billion in combined funding. Curated by Asaf Nakash. Voices by AI. Opinions by human. Show notes: https://contextwindowsec.com/episodes/2026-03-16.html
The week AI security shifted from theoretical to operational. Major breaches, new funding rounds, and the emerging challenge of securing autonomous AI agents. Curated by Asaf Nakash. Voices by AI. Opinions by human. Show notes: https://contextwindowsec.com/episodes/2026-03-08.html
