// guides · explainers
Explainers
This topic gathers the Drydock explainers on AI agents and the infrastructure behind them. Each one takes a single question, answers it in the first paragraph, and then gives the background: what the thing is, how it works, what it costs and where it falls short, with links to primary sources.
There are 25 guides here, published from Oct 05, 2026 to Nov 19, 2026. The tools they cover most often are Claude Code, OpenAI Codex and Model Context Protocol.
How to choose: if you need a quick answer, read the opening paragraph and the FAQ at the end. If you are about to make a decision, read the whole piece and follow its links to the related comparisons and how to guides. Explainers are rechecked when the underlying product or paper changes, and the date shows the last check.
9 min readListen
Superpowers for Claude Code: Setup, Skills, Cost
Superpowers for Claude Code is a free skills plugin that makes the agent brainstorm, plan, test first and review. How to install it, what it does and its token cost.
8 min readListen
Qwen Code 1.0: What Changed and Should You Switch?
Qwen Code 1.0 Preview explained: the new native installer, Qoder harness, provider config and permission modes, what it lacks so far, and who should stay on 0.27.
8 min readListen
Is Replit Free to Deploy? Costs and Limits
Is Replit free to deploy? The Starter plan publishes one app free, but the link goes down after 30 days. Here is what Autoscale, Static and Reserved VM cost.
8 min readListen
Is E2B Sandbox Free? Hobby Credits and Limits
Is E2B sandbox free? The Hobby plan has no monthly fee and a one time $100 usage credit, then billing is per second. Here is what the credit buys and the limits.
10 min readListen
Is Tavily MCP Free? Yes, Within Credit Limits
Is Tavily MCP free? Yes. The server costs nothing extra. Keyless mode covers search and extract, and a free key covers all five tools with 1,000 credits a month.
8 min readListen
Claude Dreaming: How Dreams Fix Agent Memory
Claude dreaming explained: how Anthropic's dreams reread past sessions to fix agent memory, what the Managed Agents API does, what it costs, and the alternatives.
10 min readListen
GitHub Copilot Fusion: What It Is and What It Costs
GitHub Copilot Fusion explained: how it routes and combines models, Standard vs High vs Max, which plans get each, the 30% discount, and how it differs from Auto.
· updated
10 min readListen
Codex Usage Limits: What Changes in December
Codex usage limits explained for Plus, Pro and Business, what the December shared ChatGPT allowance changes, how to check resets, and how credits work.
· updated
9 min readListen
MCP Server vs Vector Database: Key Differences
MCP server vs vector database explained: one stores and searches embeddings, the other gives an agent tools. How they stack, which servers exist, when to skip MCP.
9 min readListen
What Is Agentic Coding? A Plain English Guide
What is agentic coding? An AI agent that reads your code, edits files, runs tests and repeats until the task is done. How it works, the tools, and how to start.
· updated
10 min readListen
What Is Agent Memory in Claude Code? Explained
What is agent memory in Claude Code? Plain Markdown files: CLAUDE.md you write and auto memory Claude writes. Where it lives, its limits and how to edit it.
10 min readListen
Power BI MCP Server: Which One to Use (2026)
Which Power BI MCP server to use in 2026: Fabric IQ for questions, the Authoring server (hosted or local) for model edits, with setup for Copilot and Claude.
8 min readListen
Is LangChain Free for Commercial Use?
Is LangChain free for commercial use? Yes. LangChain and LangGraph are MIT licensed. Here is what that allows, where costs start, and the one server that isn't MIT.
8 min readListen
What Is MCP? The Model Context Protocol Explained
What is MCP? The Model Context Protocol explained: hosts, clients, servers, tools, resources, the stateless 2026 spec, security risks and a tested first run.
8 min readListen
Cloudflare Security Audit Skill: How to Use It
How the Cloudflare security audit skill works, how to install it for Claude Code or Codex, what its six phases produce, and the sandbox rules it needs.
10 min readListen
NVIDIA OpenShell: What It Is and How to Use It
NVIDIA OpenShell explained: how the open source agent sandbox enforces file, network and credential policy, how to install it, and how it compares to Docker.
· updated
8 min readListen
What Is the Most Popular AI Agent Framework?
What is the most popular AI agent framework in 2026? LangChain leads developer surveys, PyPI downloads and GitHub stars. Here are the numbers and what they hide.
10 min readListen
Claude Managed Agents: Pricing and How It Works
Claude Managed Agents explained: what Anthropic hosts, the $0.08 session hour plus token pricing, self hosted sandboxes, budgets, limits and open source options.
· updated
10 min readListen
What Is Weaviate? The Vector Database Explained
Weaviate is an open source vector database in Go with hybrid search and a built in MCP server. How it works, a tested memory example, versions and cloud pricing.
· updated
6 min readListen
Is Gemini CLI Free for Pro Users? No
Is Gemini CLI free for Pro users? No. Google AI Pro access ended June 18, 2026. Pro and Ultra now use Antigravity CLI. What Gemini CLI costs if you still want it.
9 min readListen
Claude Code Mods: What They Are, How to Build One
Claude Code mods are plugins whose JavaScript runs inside Claude Code to add panes, commands and tool call rules. How to install, build and vet one safely.
7 min readListen
What Is an Agent Harness? (2026 Guide)
An agent harness wraps an LLM with tools, memory, sandboxes, and guardrails. How it differs from a framework, plus Claude, LangChain, and Docker examples.
· updated
10 min readListen
What Is Docker Agent (cagent)? A Practical Guide
Docker Agent, formerly cagent, runs AI agent teams from a YAML file. What it is, how to install and configure it, its MCP and sandbox options, and who it suits.
· updated
10 min readListen
Code Mode MCP Explained: Tool Calls as Code
Code Mode lets an AI agent write a short program that calls its MCP tools instead of calling them one by one. How it works, who supports it, and when to use it.
· updated
8 min readListen
MCP vs A2A: Which Agent Protocol Do You Need?
MCP connects an agent to tools and data; A2A connects agents to other agents. Learn when to use each, how they fit together, and what changed in 2026.
· updated

















