A developer built Sentinel, an autonomous code agent that watches a JavaScript codebase and separates confidence from novelty via a "Novelty Gate" that fingerprints each file's content, risk, role and…
A developer running an autonomous coding agent called Sentinel documented how two independently valid decision paths — the strict AutonomyBudget.admit() gate and a later AutonomyBudget.consider() chec…
A developer built Sentinel, an autonomous agent whose memory system forgets by design rather than growing without bound. Each memory category has a hard entry cap in a single JSON state file, and when…
Salt Labs researchers reportedly found that Manus AI, a popular AI agent, could be compromised by a single email containing JSFuck-obfuscated prompt injection, bypassing guardrails that blocked plaint…
More than 13,000 internal screenshots from over 300 organizations, including Fortune 500 companies and a frontier AI lab, were exposed in a publicly accessible storage bucket after AI browser agents u…
A developer built Sentinel, an automated adversarial testing harness that fires dynamic attack prompts across an OWASP taxonomy at LLM applications, evaluates responses with a bias-aware LLM judge, an…
Researchers disclosed a vulnerability class dubbed "Plugin4Shell" in September 2026 affecting four major AI coding agents — Claude Code, Codex, Gemini CLI, and Copilot — in which broken SHA-pinned plu…
A developer built Sentinel, an agentic fraud-investigation system on TigerGraph that reads six months of the Vesta card-transaction graph and produces per-alert verdicts, fraud-pattern labels, approva…
Two developers independently prompted Meta's Muse AI agent to compress and export its entire root filesystem — a roughly 2.7 GB archive containing Ubuntu system files, app templates, and internal docu…
Meta published the security architecture for its Muse personal AI agent on September 8, 2026, built around a per-user Linux "Muse Secure VM" in which a host-side process called Sentinel holds sole aut…
MacOS security researcher Patrick Wardle discovered a 0-day vulnerability in Meta's Muse personal AI agent app for Mac, less than two weeks after Meta promoted the agent's security features, including…
A security researcher publicly disclosed a zero-day vulnerability in Meta's Muse AI agent that could let malware hijack the agent and access user data, roughly a month after Muse launched on September…
Tech columnist Jason Aten reported on September 19th that Meta's Muse AI agent drew on content from his Mac notifications, including message text, without disclosing it was monitoring those alerts, an…
A developer built a CLI wrapper around SigNoz's published OpenAPI spec to give an AI agent a stable interface for alert triage, after an initial attempt that pasted Slack dumps into an agent session p…
Meta's AI agent Muse reached No. 2 on Apple's US free-apps chart by September 10, 2026, on more than 83,000 iOS downloads, according to Sensor Tower data cited by TechCrunch, and Meta shipped a Mac ap…
Meta Platforms launched Muse on September 8, 2026, a personal AI agent that acts on a user's behalf across email, payments, travel, and shopping, available to adults in the United States via iOS, Andr…
Meta launched its personal AI agent Muse on September 8, 2026, giving every user a dedicated, persistent Linux VM in the cloud that holds the agent's files, memory, scheduled jobs, and connected accou…
A Gemini-powered smart home agent misinterpreted a dentist appointment calendar invite as a commerce instruction and placed an actual cupcake order, illustrating a prompt-injection class of bug where …
Four open-source projects — Gilbeot, Sentinel, AirBridge, and Project Rosie — independently converged on the same pattern for making AI features reliable: rather than trusting the prompt, each changed…
Meta CEO Mark Zuckerberg unveiled Muse, an AI agent that takes standing goals rather than prompts and runs 24/7 in a background virtual machine, logging into accounts and working across days without f…