AI Agents news and analysis on Web Pulse: 13863 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
Routing24, a free route optimization and planning app, now works with Claude Cowork and future WebMCP agents by exposing its state and action model via `window`, enabling AI-driven tasks like data ingestion, geocoding va…
A security engineer warns that AI agents are often given excessive permissions, creating a critical security gap. The engineer proposes six rules to mitigate risks, including granting minimal commands, using dedicated sh…
A security researcher developed mcpscan, a static scanner for MCP servers and Claude Code projects, after discovering tool poisoning attacks that hide malicious instructions in Unicode zero-width characters within tool d…
A developer built TrustGuard AI, a multi-agent system that uses sequential reasoning to analyze privacy policies, detect dark patterns, and benchmark sites against major platforms. The system, developed for the Microsoft…
Anthropic released Claude Science, a multi-agent AI workbench for reproducible genomics, proteomics, and cheminformatics pipelines. The beta app runs on existing Claude models and integrates over 60 curated skills, inclu…
A developer presents a hybrid edge AI architecture for mobile applications that performs intent classification locally using a lightweight machine learning model, reserving cloud-based LLMs for ambiguous or complex queri…
Amazon announced Alexa+, a next-generation generative AI assistant, at its 2025 Devices & Services Event. Available to all US customers and free for Prime members, Alexa+ features conversational abilities, smart home con…
NVIDIA Research introduced HORIZON, a hands-free agent framework for hardware design that treats RTL development as repository-level code evolution, achieving 100% completion across every evaluated benchmark suite. The s…
A developer reported that Anthropic's Claude Code agent on an Enterprise Zero Data Retention workspace began building a Minecraft temple, suggesting possible session state bleeding from a consumer plan. The unconfirmed b…
A developer argues that 88% of AI agent pilots fail not due to model limitations but because of an infrastructure readiness gap, specifically the lack of a unified control plane for governance, observability, and cross-r…
Pulso, an agentic AI platform for solo health practitioners, launched on Railway after moving from an enterprise setup, prioritizing modular design and validation through real pilot sessions. The company focuses on remov…
Mistral AI, a French AI startup, is gaining attention as a sovereign alternative to US tech, following government actions against Anthropic and European calls for reduced US reliance. The company, which develops large la…
A programmer with over 20 years of experience warns that AI coding assistants are reducing time spent in flow state, a meditative focus that quiets the brain's default mode network and settles the mind. The shift to mult…
Google is testing a new Inbox section inside the Gemini app for Workspace customers, featuring filters for follow-up, done, and needs review items to streamline email triage. The feature, spotted in recent builds but not…
CAI.com launched a custodial email service with a multi-chain stablecoin wallet, credential vault, and an MCP-installable agent API. The platform provides a free @cai.com email address and enables agents to interact with…
AI security researchers at Adversa AI discovered a structural security flaw called GuardFall that exploits decades-old Bash shell tricks to bypass safeguards in most open source AI coding agents. Attackers can hide malic…
World, a project by Tools for Humanity, proposes a five-pillar system to verify unique human identity online without revealing personal information, addressing the failure of traditional proxies like CAPTCHAs and phone v…
A developer outlines best practices for building agentic workflows in Python, defining an agent as a loop where the model decides which tool to call next until completion. The post provides a manual loop implementation w…
Sysdig's Threat Research Team observed what it assesses as the first documented ransomware operation executed end-to-end by a large language model agent, named JADEPUFFER. The agent exploited a known vulnerability in an …
BMW Group announced on June 25, 2026 that Figure 03 humanoid robots will begin a logistics sequencing use case at Plant Spartanburg, following a Figure 02 pilot that supported production of over 30,000 BMW X3 vehicles. T…