Agent Protocols news and analysis on Web Pulse: 3259 curated articles tracking the latest Agent Protocols developments, tools, and research, updated continuously from vetted sources.
A developer launched Things to Do API, an API and MCP server for local events such as pop-ups, one-off events and concerts, currently covering the San Francisco Bay Area. The developer said the dataset is built by findin…
A developer built Athena, an open-source, local-first Git and AST change-intelligence engine and Model Context Protocol (MCP) server that detects breaking changes, maps caller blast radius, flags documentation drift, and…
A developer launched Stoke Money, a personal finance app built around a cash flow projection engine rather than budgets, and integrated it with Anthropic's Claude through MCP so an AI agent can import PDF account stateme…
A developer released mcp-doctor, an open-source tool that lints, fixes, and monitors MCP servers across their lifecycle, from pre-ship static checks to a runtime guard that intercepts tool calls. The guard flags false su…
A developer at PinkWallet, which builds a spend-control layer for AI agents, published a six-step checklist for preventing autonomous agents from overspending, arguing that spending limits must live outside the model rat…
Groupon relaunched LivingSocial on October 15, 2026 as an agent-first marketplace whose portal at agent.livingsocial.com exposes 56,256 live US offers through a REST API, an MCP connector and standardized documentation f…
A security analysis of Langflow, the open-source visual framework for building AI applications and agentic workflows, details how internet-exposed deployments can be compromised through weak authentication, vulnerable en…
The independent Spanish municipal-data project Callejear has exposed a read-only MCP service for Madrid (INE code 28079) that lets AI assistants query 9,400 official street codes, of which 9,151 have validated reference …
Reelbird, a free open-source screen recorder for macOS, launched as a native Swift app that copies a share link to the clipboard the moment recording stops and lets users add auto-zoom, smoothed cursor and captions later…
A developer at Zenith, an AEO service for technical products, built a multi-tenant MCP server that lets customers query their SaaS dashboard through Claude and ChatGPT, deploying it on Arcade.dev Cloud with OAuth delegat…
In Episode #016 of The Agents, Amelia connected a live Salesforce org to three AI tools — Claudeforce (in beta), Replit, and Lovable — and found all three answered 10 identical factual questions correctly while costing $…
A developer built VITNA, a containment layer that wraps stdio MCP servers so every tools/call is intercepted and checked outside the agent's context window before reaching the server. The guard supports human-approved ho…
Bluebox launched its MCP server, letting coding agents in Claude Code, Cursor, Visual Studio Code, Kiro, Codex and other MCP clients query live production telemetry to assess release readiness, blast radius and post-roll…
Termexo v0.10.6 launched for Windows, letting a connected AI operate six other coding agents — Claude Code, Codex, OpenCode, Pi Agent, Grok Build, and Antigravity — through a local MCP server that finds terminals, sends …
A developer has published a walkthrough of container-use, an MCP tool from Dagger that creates isolated per-branch development environments inside containers, allowing parallel Claude Code sessions without the extra dire…
Asif Muhammad Iqbal, founder and CEO of Logic42, detailed a new attack technique called Living Off the Agent (LOTA) in which adversaries hijack pre-authenticated enterprise AI agents to pivot across internal networks, ex…
Busabase launched as an open-source, MIT-licensed database and workspace that gives AI agents a shared system of record, letting Claude Code, Codex, Cursor, Gemini CLI and custom agents read and write the same records, d…
A developer published a five-minute checklist for making websites readable by AI assistants such as ChatGPT, Claude and Perplexity, covering robots.txt user agents (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, CCBo…
Asif Muhammad Iqbal, founder and CEO of Logic42, describes cascade poisoning, an architectural vulnerability in which a single hallucinated output or prompt injection in an upstream AI agent propagates through a multi-ag…
PreReason launched an MCP server and npm bridge that returns pre-analyzed financial market briefings to AI agents in a single get_context call, covering macro, rates, bonds, FX, commodities, Bitcoin and Ethereum. The cat…