[KongchangAI]

#MCP

523 related articles

Building an AI Smart Home That Understands Natural Language with n8n + MCP + ESP8266
·3 min

Building an AI Smart Home That Understands Natural Language with n8n + MCP + ESP8266

A smart home project using n8n workflows, the MCP protocol, and ESP8266: the Gemini-powered voice Agent "Bob" reads sensor data and voice-controls 6 devices, showing how LLM tool calling works in IoT.

Read more →
MCP Server Hosts AI Answer Service: Track Search Visibility Across Platforms with One Integration
·2 min

MCP Server Hosts AI Answer Service: Track Search Visibility Across Platforms with One Integration

AnswerLine launches a hosted MCP server returning answers and citations from ChatGPT, Gemini, Google AI Mode, and Google Search as structured JSON, with 500 free credits for new users.

Read more →
n8n AI Agent in Practice: A Complete Breakdown of Models, Memory, and MCP Tools
·3 min

n8n AI Agent in Practice: A Complete Breakdown of Models, Memory, and MCP Tools

A hands-on walkthrough of the n8n AI Agent's three core components: Chat Model switching, Simple Memory, and Atlassian MCP Server integration for reading Jira stories.

Read more →
Driving Pi with ChatGPT Sessions: MCP and Computer Use in Practice
·2 min

Driving Pi with ChatGPT Sessions: MCP and Computer Use in Practice

How to use ChatGPT tasks to organize persistent Pi sessions, connect MCP tools, and fill capability gaps with Computer Use for continuous workflows without losing context.

Read more →
Getting Started with GitHub Copilot: Teaching AI to Understand Your Codebase
·2 min

Getting Started with GitHub Copilot: Teaching AI to Understand Your Codebase

GitHub Copilot beginner guide: learn how custom instructions, skills, custom agents, and MCP servers like Playwright let Copilot truly understand your codebase.

Read more →
Connecting Blender Cloud Rendering to AI Agents: The 5-Second MCP Timeout Trap and How to Fix It
·2 min

Connecting Blender Cloud Rendering to AI Agents: The 5-Second MCP Timeout Trap and How to Fix It

JANCTION Render wraps Blender cloud rendering as an MCP Server for OpenAI Agents SDK and LangChain. This article dissects the McpError caused by the 5-second default timeout and offers a 120-second timeout config, LangChain integration code, a pure Python client approach, and cost control strategies.

Read more →
SAP MCP Security Checklist: 7 Checks You Must Verify Before Connecting AI Agents
·3 min

SAP MCP Security Checklist: 7 Checks You Must Verify Before Connecting AI Agents

A security checklist for MCP in SAP environments: from authentication, least privilege, network segmentation, and data protection to SIEM monitoring—covering the essential checks and common failure modes before connecting AI Agents to SAP.

Read more →
Building an MCP Server in Python: One Server for Both Claude and OpenAI
·2 min

Building an MCP Server in Python: One Server for Both Claude and OpenAI

Learn how to write an MCP Server in Python that connects to both Claude Code and OpenAI Agent SDK. Covers local stdio and remote streamable HTTP, allowed tools security, and ToolError best practices.

Read more →
Breaking the L402 Micropayment Bottleneck for AI Agents: Batch Macaroons and Bitcoin Anchoring
·2 min

Breaking the L402 Micropayment Bottleneck for AI Agents: Batch Macaroons and Bitcoin Anchoring

Bitcoin Stratigraphy Data Engine team solves the L402 micropayment bottleneck for AI agents via 100-call macro Macaroon batching and edge caching, integrating OpenTimestamps for state anchoring.

Read more →
Beyond MCP: Wiring Your AI Coding Agent to Jira and Bitbucket with the TWG CLI
·5 min

Beyond MCP: Wiring Your AI Coding Agent to Jira and Bitbucket with the TWG CLI

Atlassian's TWG CLI lets AI coding agents read and write Jira, Confluence, and Bitbucket. We compare it to Rovo MCP Server on token cost, composability, and use cases, with a corporate Windows install guide and security guardrails.

Read more →
AI Agent Writes SCL Programs for Siemens S7-1500 via MCP
·4 min

AI Agent Writes SCL Programs for Siemens S7-1500 via MCP

An AI Agent bridges Siemens TIA Portal V19 via MCP to independently build a conveyor sorting control program for the S7-1500, from creating a CPU and writing the SCL state machine to delivering 14 test cases — while proactively flagging 12 race conditions and boundary issues and explicitly refusing to handle safety functions.

Read more →
MCP Sampling Deprecated: What's Actually Wrong with Delegating Generation to the Client?
·4 min

MCP Sampling Deprecated: What's Actually Wrong with Delegating Generation to the Client?

MCP deprecated sampling in its July 2026 spec. This article explains how sampling works, why it was introduced, the three reasons it was deprecated (complexity, security, alternatives), and when delegating generation to clients still makes sense.

Read more →
Building a Secure MCP Server from a REST API: OAuth, Permissions, and Risk-Level Classification in Practice
·3 min

Building a Secure MCP Server from a REST API: OAuth, Permissions, and Risk-Level Classification in Practice

Step-by-step guide to turning a REST API into a secure MCP server: OAuth auth, scope & claim authorization, Read/Write/Destructive risk levels, and dual credential isolation for Claude, Cursor, and AI agents.

Read more →
mcp.tc: A Free Directory of Over a Thousand MCP Servers
·2 min

mcp.tc: A Free Directory of Over a Thousand MCP Servers

mcp.tc is a free, independent MCP server directory listing ~1,000 servers. Supports one-click setup for Claude, ChatGPT, Cursor & more, with daily reviews and a public API.

Read more →
How MCP Connectors Enable LLMs to Handle the Hardcore Math of Environmental Emissions Compliance
·2 min

How MCP Connectors Enable LLMs to Handle the Hardcore Math of Environmental Emissions Compliance

How MCP connectors wrap deterministic EPA stack testing math as structured tools, eliminating LLM hallucination risk in emissions compliance, with Vinkius sandbox execution and governance.

Read more →
AI Agent Identity Governance: A Five-Layer Architecture for Securing Agents and MCP
·4 min

AI Agent Identity Governance: A Five-Layer Architecture for Securing Agents and MCP

PlainID & iC Consult CTOs break down the 5-layer Agentic IAM framework covering agent identity, discovery, policy, MCP enforcement, and audit for AI agent security.

Read more →
MCP Protocol Explained: The "USB-C" That Connects AI to the Real World
·1 min

MCP Protocol Explained: The "USB-C" That Connects AI to the Real World

MCP (Model Context Protocol) by Anthropic — adopted by Google, Microsoft & OpenAI — acts like USB-C for AI, enabling it to connect to Gmail, GitHub, and more via a unified standard.

Read more →
River AI in Practice: Building a Voice-Controlled ESP32 Agent with n8n + MCP
·3 min

River AI in Practice: Building a Voice-Controlled ESP32 Agent with n8n + MCP

River AI is a voice-controlled AI Agent built with n8n, MCP, and ESP32. It reads sensor data and controls 6 appliances. A practical IoT + AI Agent walkthrough.

Read more →