Vertical AI Agents Explained: What They Are, Examples
Vertical AI agents work inside one industry instead of many. See how they're built, real production examples, and when to choose one over a general agent.
7 min read
SubmitLog in
Blog
A collection of useful articles for developers and software enthusiasts. Learn about the latest trends and technologies in the community.
MCP vs REST API: What Actually Differs for AI Agents
MCP vs REST API comes down to who decides what to call: a developer at build time for REST, or the model itself at runtime for MCP.
8 min read
MCP Transports: stdio and Streamable HTTP Explained
MCP transports are stdio and Streamable HTTP. Here is how each carries messages, what the 2026-07-28 revision removed, and how Claude Code connects to them.
4 min read
From demo to production: what a decade of custom AI builds taught me about agents
Adam Riccoboni, author of The AI Age, on why agents beat chatbots for productivity and five lessons from a decade of custom AI builds at Critical Future.
8 min read
MCP Prompts: What They Are and How Servers Expose Them
MCP prompts are reusable message templates an MCP server exposes through prompts/list and prompts/get. Here is how they work, with the 2026-07-28 changes.
4 min read
How an agency logs and bills MCP agent runs (without pretending the timer is the invoice)
A Copenhagen agency founder on logging MCP agent runs, splitting agent time from human review, and turning a run into an invoice line without watching anyone's screen.
5 min read
MCP Elicitation: How Servers Ask Users for Input
MCP elicitation lets a server pause a tool call to ask the user for missing input, in form mode or URL mode, before it continues. Here's how it works.
1 min read
MCP Sampling: What It Is and Why the 2026-07-28 Spec Deprecates It
MCP sampling lets a server ask the client's model for a completion. The 2026-07-28 MCP spec deprecates it, so here is how it worked and how to migrate.
4 min read
MCP Inspector: How to Test and Debug an MCP Server
MCP Inspector is the official tool for testing and debugging MCP servers in a browser, a terminal or CI. Setup steps, CLI exit codes and what changed in v2.
5 min read
7 AI Sales Agents Compared for 2026
A comparison of seven AI sales agents for 2026, covering what each one automates, who it fits, and what it costs.
9 min read
AI Agent Guardrails Explained: Types, Tools and How They Work
AI agent guardrails are the enforceable checks that keep autonomous agents inside safe limits. How OpenAI, LangChain, NVIDIA and AWS build them, and when to use each type.
3 min read
MCP Resources vs Tools: What's the Real Difference?
MCP resources vs tools comes down to who decides when each is used: the model calls tools, the application loads resources. Here is how the two differ and when to use each.
6 min read
MCP Server Examples: 7 Reference Servers and Working Code
MCP server examples worth reading: the seven official reference servers, the weather tutorial in Python and TypeScript, and how to run and debug each one.
8 min read
Customer Service AI Agents: 8 Platforms Compared for 2026
Customer service AI agents resolve support tickets without a human at every step. Eight platforms compared on approach, pricing and best fit for 2026.
7 min read
Claude Subagents Explained: How They Work in Claude Code
Claude subagents are isolated AI assistants inside Claude Code with their own context window. Here is how Claude subagents work and when to use them.
6 min read
MCP Authentication: How OAuth 2.1 Secures MCP Servers
MCP authentication secures MCP servers with OAuth 2.1: how the authorization flow works step by step, how to implement it, and the mistakes that leak tokens.
7 min read
How to Test an MCP Server: Tools and Steps
Learn how to test an MCP server with the MCP Inspector, automated pytest checks, CI scripts and security red teaming, step by step.
9 min read
AI Agent Architecture: The Components and Patterns Explained
AI agent architecture is the set of layers and patterns that turn a language model into a system that plans, uses tools and acts. Here's how it breaks down.
10 min read
How to Add an MCP Server to Claude Code: Step-by-Step
How to add an MCP server to Claude Code with claude mcp add: hosted, local and OAuth servers, scopes, .mcp.json, status checks and fixes for common errors.
10 min read
Agentic Workflows: 5 Patterns and When to Use Each
Agentic workflows let a model choose its next step at runtime. Five patterns from Anthropic and LangGraph, real GitHub and IBM examples, and when to skip them.
8 min read
7 Open Source AI Agents Worth Building On in 2026
A comparison of open source AI agents: LangChain, CrewAI, OpenHands, goose, AutoGPT, Dify and n8n, with real license terms and 2026 release dates.
7 min read
Voice AI Agents Explained: How They Work and Who's Building One
Voice AI agents handle spoken calls with an LLM instead of a script. How the STT-LLM-TTS pipeline works, who builds voice AI agents, and what's new.
6 min read
What Is an MCP Client? How Hosts, Clients and Servers Fit
What is an MCP client? It is the part inside apps like Claude, Cursor and VS Code that opens one connection to one MCP server to fetch tools and data.
6 min read
Human in the Loop AI Agents: How HITL Works and When to Use It
Human in the loop AI agents pause before a risky action so a person can approve, reject or edit it. How the pattern works, real frameworks, and when to skip it.
5 min read