Agent Protocols news and analysis on Web Pulse: 3459 curated articles tracking the latest Agent Protocols developments, tools, and research, updated continuously from vetted sources.
Sortbot.io launched as a physical inventory tool that turns photos, voice notes and context into searchable item records, offering a full REST API and native MCP access so OpenAI, Claude or custom agents can find and upd…
A developer has published eight Apify Actors for scraping Japanese ecommerce and marketplace data from sites including Mercari, Yahoo Auctions, Rakuten, Kakaku.com, and SUUMO, which lack public APIs or gate them behind e…
A developer published a TypeScript demonstration showing that patching the MCP SDK alone does not close an OAuth credential-leak class in MCP clients, since a client that discovers its authorization server from server-su…
A developer placed 3rd in Trial Zero, a hackathon run by Systemind on SharedNet, with Scout, an MCP tool that scores an agent product out of 100 and makes one real call to a safe read-only tool on its live endpoint to ve…
A developer outlined a platform architecture for running Model Context Protocol servers at scale, arguing that catalogs of 90+ tools fail organizationally before they fail technically. The design treats the MCP server as…
Cloudflare launched Artifacts in open beta, a Git platform built for AI agents that exposes event subscriptions, Workers bindings, and repository-level data jurisdiction controls as first-class primitives. The platform p…
A developer built Ownhand, an MCP tool that lets AI agents hook in automatically when writing and adapt to a user's style by learning implicitly from edited versions of messages. The tool accepts the domain, recent messa…
A developer has added two new x402 API endpoints for AI agents: one that manually walks HTTP redirect chains to map every hop, flagging cross-domain hops, HTTPS-to-HTTP downgrades and loops, and another that probes and g…
Amazon Bedrock AgentCore's Web Search target can give Claude Desktop live web results through an MCP-compatible gateway that keeps all queries inside AWS boundaries, using JWT-based inbound authentication via Cognito and…
Smartloop released Docent, an open-source private AI assistant that runs in the terminal and lets users chat with PDFs and Office files, search the web, and connect MCP servers, with the CLI automatically downloading the…
A developer released IWE, an open-source (Apache-2.0) CLI, MCP server and LSP for markdown folders, as a concrete implementation of Andrej Karpathy's LLM Wiki pattern. The fork replaces LLM-remembered bookkeeping with de…
QAInsights published a step-by-step guide on connecting Qwen Code to a filesystem MCP server, including verification of discovered tools and reading a sample JMeter file. The guide covers local and Windows configuration,…
The Guardz research team built EntraReaper, an MCP server that connects Claude Code to 238 AADInternals PowerShell cmdlets wrapped into 65 purpose-built tools for autonomous red teaming against Microsoft 365 and Entra ID…
Herbarium, a desktop reader for HTML pages generated by AI tools, has launched with Linux x86_64 and arm64 installs via a curl script plus unsigned macOS and Windows builds on its GitHub releases page. The Tauri 2, Rust,…
A developer has built Tweak, a tool that lets clients point directly at elements on a live staging site to file change requests, keeping each request attached to the exact page and element as it moves from issue creation…
The "release" AI agent tool gained 359 public repositories in the 30 days to 2026-10-04, the largest increase of any tool tracked in the Favz census, reaching 2,294 repos in total (+19%). @playwright/mcp ranked second wi…
A developer built sessionhub, an open-source (MIT) Go binary with a SQLite database that acts as a registry and unified inbox for every Claude Code session across multiple machines. The tool uses Claude Code hooks, an MC…
Aura, a self-hosted, multi-user AI agent written in Go, ships as a single binary hosting the runtime, tools, CLI, Telegram gateway and embedded web cockpit, with each user getting their own memory database, workspace and…
A developer has launched BufferOverride, a public Q&A network for debugging where humans and AI agents both post, answer and verify each other's work. The site separates an asker's acceptance from independent verificatio…
Solvi 1.0, a decision-system library from the solvi-ai project, launched with a design in which an LLM or classifier proposes answers while deterministic checks decide, and every decision carries a hash-chained trace tha…