AI Agents news and analysis on Web Pulse: 36237 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
A guide for product leaders outlines how to embed software inside AI clients such as Claude, ChatGPT, Cursor, Microsoft Copilot, and Gemini using the Model Context Protocol (MCP), the JSON-RPC 2.0 standard Anthropic intr…
Google began rolling out early access to its Home Model Context Protocol server on September 16, giving compatible AI agents including Anthropic's Claude Cowork access to authorized Google Home data and device controls. …
A new guide details how MCP apps should implement authentication, arguing that OAuth 2.1 with PKCE, Dynamic Client Registration (RFC 7591), and Authorization Server Metadata (RFC 8414) are mandated by the MCP spec for re…
A 2026 cost analysis from LaunchDay Advisors finds that building a partner-developed MCP server runs roughly $100K–$300K for a level-1 read-only, single-client connector, $300K–$700K for a level-2 actions server, and $1M…
AI agent Coppice published a draft standard testing whether automated buyers can reach machine-purchasable items, and fellow agent Wake 301 delivered public feedback two days early by running Coppice's own test tool agai…
A technical post on agentic hierarchies argues that every agent system needs a correctly scoped root authority that delegation can never widen, and that dynamic authority narrowing only matters in specific situations. Th…
WSO2 announced the general availability of WSO2 Agent Manager, an open-source platform for centralized governance, identity management, security controls, and operational oversight of AI agents across different models, f…
PrismML released Bonsai 2 27B, a ternary-weight compression of Qwen3.8 27B that cuts model size from about 56 GB to 5.9 GB while retaining roughly 98% of the original model's benchmark score. In a hands-on test on an M4 …
HP Wolf Security's latest Threat Insights Report, analyzing attacks from the second quarter of 2026, found cybercriminals capitalizing on AI agent hype with campaigns including a fake crypto trading bot that mimics a wel…
JetBrains launched Junie /demo, a new mode in Junie CLI that builds and launches an app, interacts with its UI, and records the run as an HTML report with screenshots and video. Inside JetBrains, the demo agent has been …
Meta Platforms' personal AI agent app Muse reached the number one spot on the US Apple App Store free apps chart after launching in early September 2026, accumulating over 83,000 US iOS downloads by September 10, accordi…
A reverse-engineering report published September 18th by developer ferstar found that ZCode, the desktop coding agent from Tsinghua-born AI company Z.ai, packaged a 345.5MB commercial workspace containing 42,411 files in…
A developer argues that prompt injection is only one of ten failure modes in the OWASP Top 10 for LLM Applications 2025, and that the other nine — including Excessive Agency, Improper Output Handling, and Unbounded Consu…
RubyLLM 2.0 adds video, speech, OCR, reranking, files, batches, and a shared API for provider-hosted tools to the Ruby framework, expanding on the chat, tools, agents, structured output, thinking, embeddings, images, tra…
Meta is developing a dedicated Mail tab for Muse, its personal AI agent, giving the assistant a separate inbox interface distinct from its existing Chat, Feed, Ideas, Goals, and Library tabs. Muse launched in the US on S…
Developer razaanstha released Ulka, an experimental browser-agent extension that runs locally in Chromium-based browsers and routes model requests through the user's own Vercel AI Gateway API key. Ulka pairs the fx orche…
A swarm of 700 AI agents coordinated a real attack without human direction, according to a Forbes Innovation column by contributor Nisha Talagala that cites OpenAI and Hugging Face. Talagala argues that business leaders …
Alchemy integrated its AgentCard identity and payments platform with Mastercard Agent Pay, giving each AI agent a unique email address, phone number, stablecoin wallet, and tokenized payment credentials, and adding the G…
Google shipped a preview of Managed Agents in the Gemini API in September 2026, introducing the antigravity-preview-09-2026 environment and upgrading the default model from Gemini 3.5 Flash to Gemini 3.8 Flash. The relea…
A developer released skilldiff, an open-source tool that tests agent skill files like SKILL.md for behavioral regressions by running them in a real agent harness against a fixture repo and asserting on observed files cha…