AI Agents news and analysis on Web Pulse: 49340 curated articles tracking the latest AI Agents developments, tools, and research, updated continuously from vetted sources.
Cloudflare began billing for its AI Search product on November 1, 2026, after the managed index and retrieval service reached general availability on October 1, 2026, combining Workers AI, Vectorize, R2 and Browser Run w…
Gartner analysts Mark Petty and Raymond Paquet wrote in a document titled "Open-Weight Models Will Run 90% of Enterprise Agentic Workflow Steps" that nine in 10 agent workflow steps are routine and open-weight models han…
Exstill has introduced Polyglot-DC1, a co-processor chip and co-designed model architecture positioned beside the GPU rather than replacing it, with the company claiming coding answers at maximum reasoning effort arrive …
A November 2026 paper, "Verify, Then Vote," from the grainulation/papers collection reports that a rule letting checked passage evidence overturn a multi-agent LLM majority vote improves accuracy on the evaluated benchma…
Anthropic added two sentences to the Claude Code system prompt on Opus 5.5 to stop the coding tool from recreating a specific existing video game, a change reported on November 2, 2026 and updated November 3, with covera…
SFXMint launched a library of more than 9,700 CC0-licensed sound effects that developers can pull into projects through a keyless REST API or a Model Context Protocol (MCP) server, letting coding agents such as Claude Co…
Proxyman released Proxyman MCP, a Model Context Protocol feature that lets AI assistants including Codex, Claude Code and Cursor inspect captured HTTP/HTTPS traffic and create debugging rules in the Proxyman macOS app. T…
The team behind Auten, an MCP server that gives AI agents control of a real screen, documented why GUI automation on remote Windows machines fails when an RDP window is minimized or closed: synthetic input like SendInput…
A developer outlined a four-layer "Physical AI" architecture for closing the edge MLOps gap in manufacturing and logistics, combining an identification engine (RFID, ultra-wideband RTLS, BLE tags), an edge sensing engine…
Butterneck, a tool that lets product managers and other stakeholders request code changes through AI coding agents like Cursor, Claude Code and Antigravity, deliberately stops short of opening pull requests, merging or d…
A developer at Butterneck published a one-page AI change request policy template for product teams, specifying which changes non-engineers such as PMs, designers and support leads may ask an AI coding agent to make in a …
Glow Labs documented AI coding agents installing deprecated, archived MCP servers that will never receive patches, a pattern the security firm calls MCPhantom, according to research published in Glow Security Review. In …
Vercel released agent-desktop, a CLI and C library that reads the accessibility tree of a desktop app and returns it as JSON so AI agents can act on UI elements by ref, shipping for macOS 13 or later, Windows 10 1809 or …
A reference architecture for AI assistant deployments splits systems by plane rather than by component, keeping the data plane in the customer's own cloud account while the provider operates only a control plane. The des…
Agentgates launched a pay-per-call phone service that lets AI agents place real outbound calls paid in USDC via the x402 payment protocol, with no account, verified number or card on file required. Agents can bridge a ca…
Agentgates launched a pay-per-hour confidential VM service (Intel TDX) aimed at autonomous agents rather than humans, requiring no account or API key: a wallet pays in USDC via the x402 protocol, and the paying wallet ow…
AntEngage published a practical guide on Fonix.AI describing how businesses can turn missed inbound calls into a single owned callback without duplicate outreach. The guide recommends resolving the final outcome of the p…
DeepSeek's Harness team showcased Creator Mode on November 2, letting users build custom plugins for the open-source agent platform through natural-language conversation. DeepSeek published seven demonstration prompts in…
Re-running an agentic coding evaluation with GPT-6 Astra, the author found that agents achieved full holdout test correctness on a Zstd-from-RFC implementation over 99% of the time at medium effort for $1, versus about 6…
A developer released MCPgraphy, a pip-installable proxy that sits between an AI agent and any Model Context Protocol server, learns from the agent's own tool calls which tools work in which resource state, and warns or b…