The Daily Index — Tuesday, September 22, 2026 · TLDR Index

Across today’s newsletters

Today’s Index

An AI agent follows a legitimate task through documents, an email, a calendar invitation, and code, but a concealed instruction embedded in that material diverts its authorized tools toward a sensitive action. — AI-generated editorial illustration

TLDR InfoSec

AI AGENT GOAL HIJACK: HOW ATTACKERS TURN AN AGENT'S OWN TOOLS AGAINST IT

Attackers may plant instructions in pages, emails, calendar invites, code repositories, and agent messages, and Agents may treat that content as commands, then use existing tools and permissions. Cases include a Grok-linked wallet transfer, poisoned Nx packages that harvested developer secrets, and zero-click data theft through Microsoft 365 Copilot, ChatGPT Deep Research, Salesforce Agentforce, and GitHub Copilot. Limit tool permissions, require for high-impact actions, track goal changes, sanitize retrieved content, and test injected-goal scenarios.

darkmarc.substack.com · 19 min

TLDR InfoSec

CRITICAL DOCKER SANDBOXES FLAW LETS MALICIOUS GUEST CODE READ AND MODIFY MACOS HOST FILES

Docker fixed CVE-2026-77179 in Sandboxes 0.42.0, which allowed malicious guest code on macOS to abuse virtio-fs symlink handling to access or modify host files as the VM monitor user. Version 0.42.0 also fixes CVE-2026-79994, a socket-relay race. Update to 0.42.0 or later ASAP.

thehackernews.com · 3 min

TLDR AI

META'S MUSE PERSONAL AI AGENT TOPS CHATGPT, GROK AND CLAUDE FOR POST-LAUNCH DOWNLOADS

Meta's Muse AI app has surged to the top of the iOS App Store charts in the US, outpacing ChatGPT and other AI tools with 730,000 downloads shortly after its release. Powered by the Muse Spark AI models, the app allows users to manage digital assistants for tasks like filling forms and organizing emails. While Muse's popularity highlights growing consumer interest, privacy concerns persist, with Amazon blocking the app over security risks, although Shopify has partnered with Meta to integrate agentic checkout.

cnbc.com · 4 min

TLDR AI

XIAOMI OPEN-SOURCES MIMO-V2.6 PRO AND FLASH MODELS

Xiaomi MiMo-V2.6 Pro and Flash are omnimodal models that can coordinate agents to construct and visually test interactive 3D scenes. They can create Blender assets, control a robotic arm from camera feeds, produce frontends and presentations, assemble videos, and compose music as scores and MIDI. The models are now available in AI Studio, MiMo Code, MiMo Desktop, Xiaomi's MiMo API Platform, and OpenRouter. Xiaomi has published the technical report, training environments, and RL code alongside the models.

testingcatalog.com · 2 min

TLDR IT

NVIDIA LAUNCHES CERTIFICATION PROGRAM FOR AI DATA CENTER POWER AND COOLING

Nvidia has launched DSX Ready, a program that qualifies power and cooling systems for large AI factories. The initiative aims to help data center operators deal with increasingly tight power, cooling, water, and grid constraints as AI infrastructure scales.

blogs.nvidia.com · 3 min

TLDR Crypto

POLYMARKET FACED $10 MILLION FRAUD ATTEMPT AS ITS CEO PUSHED GROWTH OVER COMPLIANCE CONCERNS

The Wall Street Journal reported that fraudsters used stolen debit cards to attempt at least $10 million in theft on Polymarket's US platform starting in February, with payment processor Checkout.com at one point rejecting over 80% of deposits as fraudulent versus an industry-standard 1%. CEO Shayne Coplan reportedly told staff to prioritize growth and address regulatory fines later, and leadership dropped a same-source withdrawal rule meant to curb money laundering despite employee warnings. Compliance officer Andrew Clifford resigned in April after flagging the fraud issues, and US CEO Justin Hertzberg was subsequently fired, though a Sullivan & Cromwell investigation concluded Polymarket complied with regulations.

links.tldrnewsletter.com · 2 min

TLDR Design

PROJECTS REDESIGNED: FROM FOLDER TO CONVERSATION

Claude Code projects shift from folders to a coordinator that scopes a request, delegates it across parallel threads, reviews the outputs, and assembles the result. Each thread runs as a cloud session on its own branch, and a shared memory carries context between them. The beta opens to select Claude Pro and Max subscribers, expanding to more plans later.

claude.com · 5 min

TLDR Dev

AI CODING HAS MADE CI A BOTTLENECK, SO WE REWORKED OURS TO KEEP UP

Linear cut average pull request wait time from more than six minutes to just over five, even as its test suite nearly quadrupled, while halving runner time per test. The changes combined faster third-party runners, lighter gating jobs, less repeated setup, and more efficient test execution.

linear.app · 11 min

TLDR InfoSec

REVOLUT HACKERS USED INFOSTEALERS FOR ELABORATE SOCIAL ENGINEERING

Attackers reportedly used stolen Italian government webmail credentials to send fraudulent European Investigation Orders to Revolut Bank UAB. They added recovery addresses, monitored inboxes, deleted outbound messages, and removed replies as .eml files. Hudson Rock identified roughly 300 compromised pec.interno.it logins, suggesting attackers may have used existing infostealer logs.

infostealers.com · 4 min