Agent Protocols news and analysis on Web Pulse: 3169 curated articles tracking the latest Agent Protocols developments, tools, and research, updated continuously from vetted sources.
A developer contributed a series of upstream merge requests to GitLab and the GitLab Language Server (gitlab-lsp), extending GitLab's Model Context Protocol (MCP) tooling that connects AI agents and IDE assistants like G…
ServiceNow and OpenAI have deepened their partnership to let OpenAI agents execute governed ServiceNow workflows directly inside ChatGPT through the ServiceNow Action Fabric, with a new ServiceNow MCP plugin available in…
Doola launched a Model Context Protocol (MCP) integration in Cursor that lets developers form a US LLC without leaving their code editor, the company said. The integration targets founders who have a working product but …
Developer 28naem published Mnemosyne 3.0.0-rc.2, a pre-release of the MIT-licensed local-first long-term memory layer for AI agents, installable via `npm install mnemosy-ai@next` on Node.js ≥22.16, alongside the stable 2…
A free tool announced on Show HN runs 16 checks on a submitted homepage to determine whether ChatGPT, Perplexity and Google AI can read, trust and cite the site, and reports which checks fail. The tool requires no signup…
A developer built PressBridge, a self-hosted MCP server that lets Claude manage WordPress sites over SSH without exposing wp-config.php, the database password, or other sensitive files. Every permission limit is enforced…
Engineer Naveed Ahmed outlined an architecture for giving AI agents limited access to production infrastructure, using the Model Context Protocol (MCP) to expose narrowly scoped diagnostic tools rather than a general-pur…
Cloudflare launched a beta Web Search API through its AI Gateway, giving AI agents live web results from Ceramic.ai, Exa and Linkup via REST APIs and Cloudflare Workers, with unified logging, access controls and billing …
A developer has proposed making AI agents a first-class principal in IAM, built on the rule that an agent owns no access and instead borrows it from the user it works for, one request at a time, with consent, for the cur…
Powerduck published a guide showing developers how to generate an MCP server from an OpenAPI spec, run it locally, and register it in Cursor's MCP configuration so the coding agent calls typed tools tied to real API oper…
A developer measured the practical limits of AI-generated PNG logos, finding that a 1,200-pixel PNG stretched across a one-metre banner renders at roughly 30 pixels per inch, far below the 300 dpi print shops require, wh…
TuringGate Lab launched a weekly AI-jargon series, Suggested Jargon, publishing AI-written and AI-disclosed posts that rank terms by mention counts from a daily scan of Hacker News headline feeds and dev.to titles. The f…
DigGitHub launched a service that lets users type "dig" in front of any GitHub URL (github.com/owner/repo becomes dig github.com/owner/repo) to check whether an open-source project is still actively maintained. The site …
OpenRouter launched O11, a wallet-based platform that lets AI agents find, use and pay for tools from a catalog of 8,416 services spanning models, search, data, media and services. O11 connects through MCP, an API or an …
A developer built a paid data API serving NSW (Australia) development applications, complying development certificates and per-address planning reports from open government data, exposing every endpoint as an MCP tool an…
A developer outlined a vendor-neutral pattern for "AI agent capability diffs," which snapshot the operational powers available to an agent—effect, resource, sensitivity, identity boundary, and environment—and compare eac…
Antalpha's MCP server exposes 120 tools to any connecting agent, and 17 of them can write, according to the company's own count verified on 21 October 2026 via the standard MCP tools/list call. Sixteen of the 17 write-ca…
At Android Dev Summit on October 28, Google released Android Studio Rabbit 2 with stable Bring Your Own Agent support, letting developers install and run over 40 ACP-compatible coding agents — including Anthropic's Claud…
A tool called hlock was released to package and portably restore AI agent harnesses — the harness version, models, MCP servers, skills, plugins, instruction files and config — into a single hashed harness.lock file that …
Powerduck published a guide arguing that AI coding agents fail mainly through under-context and over-trust, and that the fix is structural: expose the API contract to agents through an MCP server rather than pasted docs,…