Zukko Blog
Guides to Claude Code, Cursor, Codex and the Claude and GPT APIs: setup, token savings and troubleshooting.
Translating i18n Files with AI Without Breaking Placeholders
6 min · · development
A script that translates i18n files via API, keeps {name}, %s and ICU forms intact, and flags suspicious strings for manual review.
Few-shot prompting: examples in the request instead of long explanations
5 min · · development
Few-shot means a few examples of the right answer placed directly in the prompt. When it helps, how to format examples, and how to check whether they pay off.
10 prompts for programmers: from code review to database migrations
3 min · · development
Ten prompt templates for everyday developer tasks. Plug in your code and send it through the API: Python code, a cost estimate, and wording tips.
How to write Cursor rules in .cursor/rules without wasting tokens
4 min · · IDE
Cursor rules are persistent instructions for the agent. The .mdc format, four ways a rule gets applied, examples, and how rules affect the size of every request.
AGENTS.md for Codex CLI: what the file is and how to write it
5 min · · агенты
AGENTS.md is plain Markdown with instructions for the agent: build commands, tests, code style. How Codex discovers these files and what to put in them.
Claude Code on a remote server over SSH: install, API key and tmux
4 min · · development
Claude Code is a terminal agent and needs no GUI. Installing it on a server, authenticating with a Zukko key, working in tmux, and protecting the key.
Claude Code in CI/CD: non-interactive -p mode and GitHub Actions
5 min · · development
Claude Code can run without a UI: it takes a task via -p, does the work and exits. The documented flags and a working workflow for pull request reviews.
Claude Code on Windows: native or WSL install and Zukko setup
5 min · · development
Claude Code runs on Windows natively and in WSL. Official install methods, Git Bash, environment variables for Zukko, and the usual errors.
MCP servers in Claude Code: connect GitHub, databases and your own APIs
5 min · · development
MCP connects external tools to Claude Code: GitHub, databases, internal APIs. Commands, configuration scopes, and how MCP affects token usage.
Claude Code subagents: delegate tasks and keep your context clean
6 min · · агенты
A subagent handles a side task in its own context and returns only the summary. File format, ways to invoke, and model choice, based on the official docs.
Claude Code hooks: run scripts before and after agent actions
6 min · · development
Hooks run your scripts on Claude Code events: format code after edits, block risky commands, send notifications. A walkthrough of the configuration based on the official docs.
Tool use in Claude: the full tool-calling loop in Python
4 min · · development
Tool use lets Claude call functions in your code. The whole loop: tool definitions, parallel calls, errors and limits.
Migrating from OpenRouter to Zukko: what to change in code and config
5 min · · development
Moving from OpenRouter to Zukko's OpenAI-compatible API comes down to the address, key and model names. Every change, OpenRouter-specific parameters and a rollout plan.
Qwen Code CLI with Claude, GPT, and GLM via Zukko
4 min · · IDE
Qwen Code isn't limited to Qwen models: it supports OpenAI-compatible APIs, so through Zukko you can use Claude, GPT, and GLM in it.
Configure OpenClaw with Zukko — provider choice and config
4 min · · агенты
OpenClaw connects to Zukko as a custom provider. We cover the two API options, ready-made openclaw.json configs, and common errors.
Claude Code in VS Code — extension and integrated terminal
5 min · · IDE
Claude Code runs in VS Code both as an extension with a graphical panel and as a CLI in the integrated terminal. Both read one settings.json, so you set up Zukko once.
Classify customer inquiries with AI: Python example
3 min · · development
Support tickets, complaints, feature requests — some go to tech, some to product, some to the archive. Here's how an AI model routes them.
Generate SQL queries with AI: Python example
4 min · · development
The user asks a question in plain language, the model generates SQL. Feed the schema into the prompt and validate the query before running it.
LLM integration in Django: views, streaming, Celery and cost tracking
4 min · · development
Connecting a model to a Django project: OpenAI SDK client, a cached view, a streaming response, long jobs in Celery and a spend table in the admin. With common pitfalls explained.
Next.js Route Handler for AI: streaming responses in the App Router
4 min · · development
We write app/api/chat/route.ts that streams a model response through the OpenAI SDK, plus a client component that renders text as it is generated. The key stays on the server.
AI chat on Streamlit: streaming, history and settings
4 min · · development
Building a Streamlit chat app: history in session_state, streaming output with st.write_stream, model choice in the sidebar, the key in secrets.toml and spend from usage.
Streaming LLM responses in Node.js Express with the OpenAI SDK
4 min · · development
A working Express endpoint that receives a model stream through the OpenAI SDK and forwards it to the browser over SSE. Plus a fetch-based frontend, cancellation and error handling.
API key in Docker: passing it safely without baking it into the image
5 min · · development
A key in a Dockerfile stays in the image layers forever. We cover env_file, Compose secrets, build-time secrets, .dockerignore and image checks — with examples for a Zukko key.
How AI finds bugs in screenshots: UI analysis for QA
4 min · · development
Playwright captures the screen, a vision-capable model returns JSON with typos, overlaps and missing elements. Python code, a prompt template and an honest cost calculation.