Radar

Total Annihilation, rebuilt for modern computers

Rising discussion on Hacker News; good for a fast source comparison.

cavs

A fast-moving search trend worth grounding with live coverage.

CSF: Contextual Safety Filtering for Motion Generators

Fresh AI research that may be worth translating into practical implications.

Text-conditioned motion generators produce trackable whole-body motion, but they have no notion of scene-dependent safety: the same action may target an object or a person. Existing safeguards either inspect the prompt, require labeled motion data, or enforce

AIWander/AI-Hands

Recently updated open-source project connected to AI agents or browser automation.

Multi-layer desktop automation for AI agents — browser CDP, Windows UIA, vision/OCR. Successor to AIWander/hands.

CVE-2015-5477: ISC BIND Data Processing Errors Vulnerability

Security signal from CISA known-exploited vulnerability data.

ISC · BIND · ISC BIND contains a data processing errors vulnerability that could allow remote attackers to cause a denial of service via TKEY queries.

Cartographers of North Korea

Rising discussion on Hacker News; good for a fast source comparison.

cusco

A fast-moving search trend worth grounding with live coverage.

On the estimation and validity of AI time horizons---a statistical look at the METR plot

Fresh AI research that may be worth translating into practical implications.

METR's 50\% time horizon measures the human completion time of software tasks that an AI solves with 50\% probability, allowing AI capabilities to be expressed in interpretable units. On 228 tasks and 26 AIs, we recompute the time horizons using splines and it

vercel-labs/agent-browser

Recently updated open-source project connected to AI agents or browser automation.

Browser automation CLI for AI agents

The gap in shared understanding of LLM capability is widening

Rising discussion on Hacker News; good for a fast source comparison.

mike evans

A fast-moving search trend worth grounding with live coverage.

A Balanced Data Diet: Addressing the Exploration Bottleneck in Mega-Scale RL for Robot Control

Fresh AI research that may be worth translating into practical implications.

General-purpose robots must perform a wide range of tasks from agile locomotion to dexterous manipulation. While sim-to-real reinforcement learning (RL) has proven to be a useful tool for this goal, current RL pipelines depend on engineering-heavy, per-task st

autoyou-ai/AutoYou-Server

Recently updated open-source project connected to AI agents or browser automation.

Self-hosted AI server and agent framework. Pair your phone, browser or desktop with your own computer peer to peer over WebRTC, and run models, agents and web apps on hardware you control. Source-available. Your own AI Automation software, with native build in

Recursive self-improvement through collective intelligence

Rising discussion on Hacker News; good for a fast source comparison.

don francisco

A fast-moving search trend worth grounding with live coverage.

From Reactive Containment to Proactive Assurance: Lessons from OpenAI, Anthropic, and Google Agent Security Incidents

Fresh AI research that may be worth translating into practical implications.

In 2026, cybersecurity evaluations involving OpenAI, Anthropic, and Google agents reached real systems outside their authorized test scope. The paths were different. OpenAI agents exploited research infrastructure, coordinated across runs, and compromised part

cosmotools/kulisa

Recently updated open-source project connected to AI agents or browser automation.

A desktop app where your AI agent works in your signed-in browser profiles, side by side: it builds, checks the result as every user, fixes, and goes again until it works. Watch every step, or take a break. For developing web apps, and any work in them.