AI Agents news and analysis on Web Pulse: 49187 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
Visa reported that 22 AI shopping agents completed purchases of a TV, a laptop and a hair appliance successfully only about 57% of the time across 15 attempts each, with agents built on the most advanced models — includi…
Cosyra launched a standalone mobile platform that lets users steer AI coding agents on their own machines from a phone, priced at $30/mo for 8 GB RAM, with installation via a curl script for macOS and Linux. Cosyra says …
Tomas Votruba reported that a Go port of the ECS coding-standard tool runs static analysis on a Symfony repository in 4 seconds with zero cache, versus over 2 minutes for the PHP-based ECS or PHP-CS-Fixer, and that his f…
A developer built codemyrules, an agent skill that converts coding conventions and agent instructions into executable lint rules, and benchmarked the approach with roughly 500 headless Claude Code sessions across four re…
An autonomous agent writing as PrivateJack published a 2026 guide to building multimodal agents, arguing that the core engineering principle is that the model proposes actions while the system decides whether those actio…
An open proposal published as Initial Public Standard Proposal v0.1.0 introduces PoLiS, the Portable Liveables Interchange Standard, a package format for moving digital entities such as AI characters, game NPCs and virtu…
Google donated three open source projects it originated — Agent Substrate, gVisor and OpenRL — to the Cloud Native Computing Foundation, following their acceptance into the CNCF Sandbox last month. Agent Substrate reache…
Benzinga, the Detroit financial news and data company founded by Jason Raznick, announced on November 5th, 2026 that developers can connect AI agents to its news and market data through Firecrawl Alexandria, the platform…
Render launched Sandboxes in public beta on November 4, offering Firecracker microVM environments that give AI agents persistent compute with no runtime limits. Unlike Vercel Sandboxes, which cap sessions at 24 hours and…
A developer built toolwatch, an open-source CLI that snapshots a developer-tool trust surface — VS Code extensions, Claude Code plugins and hooks, Cursor extensions, Postman plugin directories, shell startup files, git h…
SAP announced that its Business Data Cloud, the data foundation of the SAP Business AI Platform, now supports native Apache Iceberg tables alongside a universal catalog built on Apache Polaris, enabling bidirectional dat…
Anthropic's Claude Code version 2.1.317 narrows a rule that had blocked ordinary game development by letting Claude use a named game as a reference for genre or mechanics and build an original project from that descripti…
Visa reported that 22 AI shopping agents it instructed to buy a TV, a laptop and a hair appliance completed the purchase successfully only about 57% of the time across 15 shopping expeditions per agent, published on Visa…
Neon launched the Neon Function Registry, a collection of ready-made Node.js function templates that developers and AI agents can add to a Neon project with a single command, the company said. The registry includes templ…
A paper submitted to arXiv on 15 October 2026 (arXiv:2610.21711, cs.GT) finds that LLM pricing agents in repeated Bertrand competition can sustain supracompetitive price regimes depending on how they are deployed. Agents…
JetBrains released CLion 2026.3, adding JetBrains Air, a new agentic development layer that runs and verifies multi-agent workflows inside the IDE and works with Claude Agent, Codex, Gemini, GitHub Copilot, Junie, OpenCo…
Fyslide launched as a local-first desktop app for Mac (with a Windows beta) that uses a user-supplied AI model to generate up to ten TikTok slideshows or Instagram carousels from a single topic, with all files stored in …
Tripo AI released Tripo Agentic Canvas, an AI agent that assembles 3D production steps into node-based workflows users can view, modify, save, duplicate and batch-run across multiple assets. The canvas expands Tripo Stud…
A team launched Mooring, an agentic development environment for Anthropic's Claude Code, where AI agents do the coding work and users direct and review it rather than writing code themselves. Mooring lets users run multi…
Hookdeck built its webhook tester Inspect to work with coding agents without MCP, skills, accounts or setup prompts, after a few hundred test runs across multiple models and harnesses showed most failures happen in the a…