AI Tools news and analysis on Web Pulse: 38327 curated articles tracking the latest AI Tools developments, tools, and research, updated continuously from vetted sources.
A hands-on experiment published by Towards Data Science benchmarked four AI retrieval architectures — plain RAG, graph RAG, and placing all documents directly in a frontier model's context window — against the same set o…
DocEngine now runs a shared agent workflow on a self-hosted OpenHands environment, where agents implement GitHub Issues tasks, open pull requests, and run a review-and-fix loop before a human decides whether to merge. Me…
A developer reports that the specialized vision-driven browser agent Mano-CUA 1.1 scored 41.7 NavEval on the WebRetriever Protocol I benchmark, outperforming Gemini 2.5 Pro Computer Use at 40.9 and Claude 4.5 Computer Us…
Google launched Gemini 3.8 Live Extended Thinking on Sept. 15, a voice model that continues reasoning through multistep tasks and runs tools in the background while audio streams, alongside the lower-latency Gemini 3.8 L…
NodePilot launched as an agentless Windows workflow orchestration tool positioned as a modern, open replacement for Microsoft System Center Orchestrator, importing native .ois_export XML runbooks with activities, links, …
A developer who tested AI-powered frontend debugging tools found that they combine source-code parsing with runtime data analysis—error logs, variable values, and event traces—fed into large language models to suggest fi…
ThinkingAI announced the Agentic Engine, an autonomous growth engine whose AI agents instrument data, diagnose causes, decide responses and execute them — from tracking plans to segments to live campaigns — running on th…
A senior backend engineer writing as Devrim spent a year reviewing every production bug his team shipped and found AI-written code fails differently from human code rather than more often, with AI bugs passing review and…
Five of 16 AI models tested with a web search tool attached still named King Harald V as Norway's current monarch after his death on 28 August, according to a test published by stale.jock.pl creator jock.pl. One model, G…
Mandiant reported that an attacker hijacked an active AI coding-assistant session at an unnamed software-as-a-service provider and spread the Shai-Hulud worm across roughly 100 internal code repositories, stealing reposi…
Google launched Agent Anomaly Detection for the Gemini Enterprise Agent Platform in private preview on September 11, the first platform-native monitoring layer grounded in the OWASP Top 10 for Agentic Applications. The f…
Ontos-AI released Knowhere 2.0, a document parsing and retrieval system that reconstructs document hierarchy for AI agents through dual Vision and Text tracks, with the full stack open-sourced on May 7, 2026. The Septemb…
A developer at NeuraGrowth found that nine of ten pinned MCP package versions in a guide failed to resolve in the npm registry, causing 404 errors at install time even though the packages themselves exist. The fix, shipp…
A senior DevOps engineer described how a conversation with engineering manager Naveed Sanghera at a company conference reshaped his view on AI's role in his work. Sanghera advised treating AI as a tool rather than a full…
JPMorgan has imposed a $2,000 monthly spending limit on some engineers using Anthropic's Claude Code, according to internal Teams messages seen by Business Insider, with an "ExceededBudget" error code and "Budget=2000.0"…
Hamilton Ulmer built a DuckDB extension that uses TypeSafeAI's Jev to classify rows in any CSV, Parquet file, or DuckDB table, processing roughly 1,000 rows in about 10 seconds. Ulmer said the approach is faster than usi…
A blog series titled "Coding Agent Diaries" published three posts dated September 15-17, 2026, covering the author's experience working with coding agents: "Day one," which lays out the rules the author is working under …
Anthropic engineer Thariq Shihipar said on X that MCPs are better than CLIs for most integrations, citing improved model tool calling, deferred tools, and MCP now being stateless. Shihipar added that the one remaining is…
An engineer has released five open-source tools forming a layered strategy for testing AI systems, arguing that traditional software testing assumptions break down because LLM outputs are nondeterministic and often have …
Citrolabs released ego (lite) v2.0.0, an MIT-licensed Chromium browser that lets AI agents such as Claude Code, Codex, and Cursor drive the user's own logged-in browser session through a connection layer called ego-brows…