Zukko Blog
Guides to Claude Code, Cursor, Codex and the Claude and GPT APIs: setup, token savings and troubleshooting.
Home Assistant: a voice assistant on Claude via API
4 min · · автоматизация
Home Assistant's built-in voice assistant can run on any catalogue model through one API key. Configuration and the failure modes to expect.
Obsidian: setting up an AI plugin with your own API key
5 min · · инструменты
Obsidian AI plugins ask for three fields — endpoint, key, model. What goes in each, and why the first request sometimes fails with a 404.
Raycast: an extension using Claude and GPT on your own key
4 min · · IDE
Raycast AI bills by subscription seat; your own key bills by actual usage. Building a TypeScript extension that talks to the API directly, bypassing built-in AI.
Wiring an AI bot into Slack over API
4 min · · development
Slack expects a reply to a command within three seconds, and a model can take longer. How to work around it with response_url, with working Bolt code.
Gateway or direct API access: which to choose
3 min · · pricing
Each option has its own price and its own risks. Both sides honestly, including where a gateway is inconvenient.
Claude Code vs Cursor: How They Differ and What to Pick
3 min · · claude-code
Claude Code lives in the terminal, Cursor in the editor. Where each tool fits and how to connect both through one gateway.
How to Reduce Token Usage in Claude Code
3 min · · claude-code
The bill grows not from the number of tasks but from the context sent with each request. Here are practices that remove unnecessary tokens.
Nano Banana image generation through the API
2 min · · models
The image endpoint explained: three modes, request shape, and honest arithmetic on what an image costs.
Grok 4.6 through the API: setup and pricing
3 min · · models
Access through a shared key, a flat rate, a million-token window. Setup and what the model is worth comparing on.
Claude Code slash commands worth using
3 min · · claude-code
A walkthrough of the built-in Claude Code slash commands and how to package your own for the requests you type every day.
How to choose a model gateway: what to look at
3 min · · pricing
Six checks before you trust an intermediary with real traffic. Each takes a couple of minutes.
Where Qwen models fit in development work
3 min · · models
Qwen cost one and a half to three times less than flagships. Which tasks show no difference, and which do.
GPT-6 Astra through the API: setup and cost
3 min · · models
Setup takes one line. Here is the cost and a way to judge the model on your own work rather than on reviews.
Codex or Claude Code: comparing console agents
2 min · · codex
Both live in the terminal and edit code, but they are built differently. The differences that show up in use.
AI tools in the terminal: what exists and when to use what
2 min · · cli
Not everything called an agent is one. Three classes of tools and the jobs they were built for.
Astra, Grok and Claude: how to choose between them
3 min · · models
Comparing what can actually be verified: price, window, rate shape. Plus a method for choosing under your own load.
VS Code AI extensions: how they differ
2 min · · setup
There are many extensions; they differ in three things: how they connect a model, how they build context, how they ask permission.
Connecting Claude Code to an existing project
3 min · · claude-code
A first run of the agent in an unfamiliar repository: what it reads, how to set boundaries, and how a project instructions file helps.
503 errors — the model provider is unavailable
2 min · · errors
A 503 says the problem is not yours. How to survive it without losing work, and why streamed errors arrive with a 200.
context_length_exceeded — the context window is full
2 min · · errors
The model refused because your request does not fit. What makes up the context window and what to cut first.
429 Too Many Requests — what to do
2 min · · errors
A 429 is not a failure, it is a request to wait. How to retry properly and why agent workloads hit it soonest.
401 Invalid API key — why your key is rejected
2 min · · errors
You pasted the key and the server still answers 401. Here are the five causes, ordered by how often they actually happen.
Claude Code keeps asking to log in to Anthropic
2 min · · claude-code
Claude Code insists on signing in to Anthropic even though your key is configured. One environment variable is to blame.
The model's answer stops mid-sentence — why
2 min · · errors
The answer cuts off mid-word. There are four causes, and one field in the API response tells them apart.