Agent Protocols news and analysis on Web Pulse: 3429 curated articles tracking the latest Agent Protocols developments, tools, and research, updated continuously from vetted sources.
Amazon unveiled Amazon Ads Agent at its annual UnBoxed advertising event, an overhaul of its core ad platform that integrates chat-based features and AI into all advertiser workflows, including two campaign types, DVA+ a…
Cloudflare launched Cloudflare Artifacts in open beta on October 1, 2026, a distributed, versioned filesystem that speaks Git and is built specifically for agent workloads on Cloudflare Durable Objects, with a custom Git…
A developer expanded a URL metadata service with two new paid x402 endpoints for AI agents: one that actively opens an HTTP/2 connection and exchanges SETTINGS frames to grade a server's protocol support, and another tha…
Sierra and Meta announced the Personal Agent Protocol (PAP) on October 6, 2026, a third protocol layer for AI agent infrastructure that adds OAuth-based identity and authentication governance on top of the Model Context …
Atlassian introduced AMP, the Agentic Multiplayer Protocol, at its Team '26 Europe conference in Amsterdam on October 7th, proposing that AI agents work as visible, identifiable and governed participants in shared team s…
Daisy, a local-first meeting recorder for Mac released on GitHub and at mydaisy.io, transcribes meetings on-device and runs a local MCP server on port 54321 so Claude Desktop, Claude Code, Cursor or Codex can search and …
Drupal creator Dries Buytaert used his DrupalCon Rotterdam keynote to launch an advocacy program aimed at closing the gap between Drupal's current capabilities and its outside reputation, following the release of Drupal …
OX Security analyzed 15,465 publicly indexed MCP servers across 5 MCP registries, deduplicated to 5,095 unique hostnames, and found no marketplace vetting or code signing, according to a report titled "15,465 MCP Servers…
Meta unveiled a set of AI advertising tools at Advertising Week 2026 in New York on October 6, including in-chat checkout through Messenger and Meta AI, an expanded Ads MCP server supporting third-party AI tool integrati…
Atlassian launched Agentic Multiplayer Protocol (AMP), a foundation for human-agent collaboration across its platform, along with Teamwork Graph, which indexes source code down to functions, symbols and classes and provi…
Atlassian Corp. announced the Agentic Multiplayer Protocol (AMP) at its Team '26 Europe event, a platform update that assigns each AI agent an administrator-set identity with defined authority and scope so humans and age…
REA (Reverse Engineer Anything) shipped on October 7, 2026 as an open-source CLI and MCP server that gives AI agents 41 native binary inspection tools and 14 guided investigation workflows for decompiling and analyzing n…
Eighty-eight percent of IT leaders reported at least one security breach involving AI agents in the last 12 months, according to the AvePoint State of AI 2026 report, while the ngCERT GHOSTJACKING advisory released Octob…
A developer is building a reusable Robinhood Trading MCP Risk Gateway in TypeScript that inserts deterministic permission, risk, and approval layers between an AI agent and trade execution. The gateway validates trade in…
Tuskira released AI Agent Gateway, an open-source tool that sits between AI agents and the MCP tool servers and model providers they call, keeping credentials out of agent configurations. The gateway runs in a user's own…
OWASP's Q1 2026 exploit round-up documents eight real-world AI agent security incidents, including a three-hour PyPI backdoor of the LiteLLM gateway that pulled an autonomous attack bot into roughly 47,000 downloads, a p…
A developer argues that the Model Context Protocol (MCP) beats plain function calling for agentic AI teams whose agents must outlive a single model, framework or application, since one MCP server integration works across…
Sierra and Meta announced the Personal Agent Protocol (PAP) on October 6, 2026, an open standard built on OAuth that governs how personal AI agents authenticate with businesses and declare intent, backed by a founding co…
A developer demonstrated a pattern for designing APIs that LLM-based agents can call reliably, arguing that human-oriented APIs fail machine clients through vague errors, non-idempotent writes, and bloated responses. The…
A developer argues that multi-party computation (MPC) wallets, not multi-signature schemes, are the correct primitive for autonomous AI agent payments, since multi-sig requires human co-signers that add 2-5 seconds per t…