Agent Protocols news and analysis on Web Pulse: 3461 curated articles tracking the latest Agent Protocols developments, tools, and research, updated continuously from vetted sources.
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…
Artbucket launched as an open-source brand asset management system that serves the same answers to people, AI agents and code through a web app, MCP server, REST API, CLI and Git. The tool exposes MCP tools named search_…
A developer published a session-reset acceptance test for AI agent memory, arguing that a successful save proves only that a correction was written, not that it was retrieved or applied before the relevant decision. The …
A developer built Quran Sanity Agent, an open-source bilingual (Arabic and English) research workspace that anchors every verse, translation, and scholarly commentary to immutable records in Sanity's Content Lake so the …
A team of AWS developers open-sourced Pizza Bot, an Apache 2.0-licensed, self-hosted inbox application that lets AI agents run scheduled or webhook-triggered tasks in the background and return results through an email-st…
WebMCP lets a web page expose its features to AI agents as callable tools by adding three HTML attributes — toolname, tooldescription and toolparamdescription — plus the optional toolautosubmit, to an existing form, with…
Supabase launched a BYO-MCP feature at its Select 2026 conference on October 2 that lets any Supabase app expose a Model Context Protocol server to users with a single command, npx shadcn@latest add @supabase/mcp-server,…
A developer has released mcpscan, a lightweight static supply-chain security scanner built specifically for Model Context Protocol (MCP) servers and Claude Code projects. The tool audits MCP server implementations and lo…
CallChatSyn launched a free public API that lets small businesses expose their answering bot, order status and booking calendar to websites, apps and AI agents, with founder access free for the first 1,000 businesses at …
Squarespace launched an official Model Context Protocol server at https://mcp.squarespace.com/mcp that connects AI assistants including Claude, ChatGPT, and Gemini to Squarespace domain tools with no API key, OAuth flow,…
Six global banks — Bank of America, Capital One, ING, NatWest, ASB Bank, and Commonwealth Bank of Australia — published a September 22 paper, "Building Trust in Agentic Commerce," laying out principles for authorizing AI…
TypeSafe AI, a two-year-old stealth startup founded by ex-OpenAI researcher Diogo Almeida and backed by a $40M DCVC-led seed round, launched Jev, a decision-only model that returns typed Choice, Score and Noul answers wi…
ScriptMasterLabs has published a receipt-first implementation of per-call x402 payments for MCP servers, in which each unpaid tool call returns a 402 Payment Required challenge carrying price, asset, network, payTo addre…
Ikanos, an open-source capability runtime published as the naftiko/ikanos Docker Hub image, argues that exposing microservices one-to-one as MCP tools forces AI agents to re-make five decisions — granularity, formats, pa…
Pi 1.0 adds native Model Context Protocol support through a code mode that lets agents write JavaScript to orchestrate tool calls and return only condensed results, alongside a custom model router, deferred tool loading,…
OpenAI published an SDK and plugin extensions specification that adds sidebar apps, conversation panels, file viewers, settings, and context sharing to ChatGPT, with the company planning to replace custom GPTs with plugi…
A developer outlined a practical threat model for teams deploying Model Context Protocol (MCP) servers internally, identifying indirect prompt injection via tool output, over-broad tool credentials, confused-deputy token…
A developer's guide to the Model Context Protocol argues that teams wrapping internal platforms in MCP should map capabilities onto its three root primitives — tools, resources, and prompts — rather than exposing every e…
A developer outlines a versioning discipline for MCP tool catalogs, arguing that agents re-read tool names, descriptions, and JSON Schemas every session, so renames, tightened parameters, or changed error semantics break…