Zukko Blog
Guides to Claude Code, Cursor, Codex and the Claude and GPT APIs: setup, token savings and troubleshooting.
Refactoring code with AI: small steps and tests
5 min · · development
An agent can rewrite a module in a minute and break it just as fast. How to organize refactoring: tests first, one kind of change at a time, a commit per step, rollback.
AI for code analysis: how to understand someone else's project faster
5 min · · development
How to get a map of an unfamiliar codebase in an evening: prompts for modules and entry points, checking answers against file references, and cost with and without cache.
MCP servers for Codex CLI: config.toml setup step by step
5 min · · development
Codex connects MCP servers through the codex mcp add command or tables in config.toml. Both ways, how to verify, and how many tokens tool descriptions use.
opencode or Codex CLI: which terminal agent to choose
6 min · · агенты
Both agents are open source, both read AGENTS.md and both work with your own API. The differences are in providers, modes and cache billing. We compare and price a session.
Cursor CLI: the Cursor agent in the terminal, install and setup
5 min · · агенты
The Cursor agent works in the terminal as well as the editor, with the agent command. Install it, configure it, run it in scripts and compare it with Claude Code and Codex.
Cline in Russia: installation and model setup without a foreign card
6 min · · IDE
The Cline extension runs locally, so the only question is where model requests go and how you pay for them. Step-by-step setup with one key and a task cost estimate.
An MCP server for Bitrix24: CRM deals and tasks in an agent's hands
6 min · · агенты
The official Bitrix24 MCP gives an agent documentation, not portal data. How to write your own server for deals and tasks and keep the agent on a short leash.
Function calling or MCP: what is the difference and what to choose
5 min · · агенты
These are not two rival methods but two levels of the same task. Who describes a tool, who runs it, and when you need MCP.
MCP in opencode: connecting local and remote servers
5 min · · агенты
Step by step MCP setup for opencode: the mcp block in config, headers and OAuth, disabling servers for specific agents, and what extra tools cost.
LangChain and MCP: connecting MCP tools to a LangGraph agent
5 min · · агенты
MCP servers give an agent ready-made tools, and adapters turn them into LangChain tools. The current approach, running on models through an OpenAI-compatible gateway.
Agentic coding: what it is and how it differs from autocomplete
5 min · · агенты
An agent does not suggest the next line, it runs a task in a loop: read, edit, run, check. How it works, which tools exist and what an hour costs via API.
How to use Claude Code: a working loop on a real task
5 min · · development
One full pass with Claude Code, from writing the task to the commit. How to read the diff, undo edits and keep token spend under control.
Cline Rules: how to write project rules in .clinerules
5 min · · IDE
Rules in .clinerules stop you from repeating the same requests in every task. Structure, Python and TypeScript examples, and the price of extra lines.
LangChain Deep Agents: what it is and how to run it on your own model
5 min · · агенты
Deep Agents is a ready-made harness on top of LangChain: files, subagents, context summarization. What is inside, how to run it via a gateway and what it costs.
Cline workflows: repeatable scenarios in one command
5 min · · автоматизация
Release, PR review, a daily changelog: any repeating sequence can be written in markdown and launched from the Cline chat. Format and cost per run.
Claude Code locally: what actually works with a local model
6 min · · агенты
Claude Code can run on a model that lives on your own computer. We cover the setup, how it differs from cloud models, and when you should count dollars instead of hardware.
AI for Java code: agent, tests and Spring without extra edits
5 min · · development
In a Java project an agent is valuable while it leaves your architecture and dependencies alone. Rules, build commands, test generation and a cost estimate.
The Codex extension for VS Code: features and your own provider
6 min · · IDE
The Codex extension lives in the VS Code sidebar and uses the same config.toml as the CLI. Features, connecting a gateway and the limits.
AI for C++ code: CMake builds, compiler errors and sanitizers
5 min · · development
C++ forgives a model a lot at compile time and punishes it later. How to build a build, test and sanitizer loop for an agent and what it costs on different models.
Claude Code in Russian: replies, comments and commit messages
6 min · · агенты
Claude Code understands Russian but answers in whatever language your instructions set. Where to put the language rule, what not to translate and what it costs in tokens.
Roo Code Ollama: local models in Cline and Aider, an honest look
6 min · · агенты
We connect Ollama to Roo Code, Cline, Aider and OpenCode per each tool's documentation, cover the limits and show when it is simpler to return to a cloud API.
Skills in Claude Code: what they are and how to write your own
7 min · · агенты
A skill is a folder with a SKILL.md file that Claude Code loads only when needed. We compare it with CLAUDE.md and write one of our own.
Vibe coding apps: from idea to a working MVP with an agent
6 min · · development
A practical route for building a small web app with an agent, from the idea to deployment. Where a human is required, and what it costs.
Vibe coding downsides: a risk checklist and how to reduce them
5 min · · агенты
Vibe coding speeds up prototypes, but it has a price. A checklist of nine risks with a practical measure for each and a calculation of possible losses.