Find anything.

ToolEncoding

jwt decoder

Decode & inspect JSON Web Tokens.

ToolEncoding

base64 encode/decode

Encode or decode Base64, incl. URL-safe.

ToolTime & Cron

cron generator

Build & read crontab expressions, in plain English.

token counter

Count tokens with the real OpenAI tokenizers.

context window viz

See what fills a model's context budget.

context budget analyzer

Find which files eat your context window.

structured output builder

Schemas for OpenAI, Anthropic, Gemini & JSON Schema.

ai pattern explorer

Compare agent design patterns side by side.

rag chunking visualizer

See how a document splits into RAG chunks.

mcp config generator

Build & validate an MCP server config.

mcp inspector

Audit an MCP config: errors, secrets, broad access.

prompt linter

Catch vague, conflicting, or unclear prompts.

prompt diff

Compare two prompts and see what changed.

prompt simulator

See what a vague prompt forces AI to guess.

prompt builder

Answer 6 questions, get a clear prompt.

prompt optimizer

Score, fix & restructure any prompt, all-in-one.

ai coding prompt builder

Plan-or-build mode: signature, types, tests, stable libs, fresh docs.

good vs bad prompts

Before/after prompt rewrites, with the why.

learning path builder

Say a goal, get an ordered route through KB Cafe's real guides & tools.

llm vram calculator

Estimate VRAM to run a model locally.

llm cost calculator

Estimate API cost from token usage.

ai profit calculator

Price, users & usage in. Margin & break-even out.

agent memory simulator

See how AI memory keeps or loses facts.

ai pattern builder

Compose an agent flow, export scaffolding.

ai stack explainer

What building blocks an AI product uses.

tool call visualizer

Step through the function-calling loop.

prompt x-ray

See the structure of a prompt.

embedding explorer

Real embeddings + semantic search, in-browser.

claim & citation checker

Flag unsupported claims to verify.

scope sorter

MoSCoW your roadmap, find the MVP.

launch readiness rubric

Score your own idea across 6 dimensions.

ai architecture generator

Describe an app, get an architecture (BYO key).

ToolConverters

csv ⇄ json

Convert tabular data both ways, instantly.

ToolJSON

json → go

Generate Go structs from any JSON payload.

ToolApache

.htaccess gen

Redirects & rewrite rules, the easy way.

ToolConverters

curl → code

Turn a cURL command into fetch, axios, or Python.

ToolConverters

yaml ⇄ json

Swap between YAML and JSON cleanly.

ToolText

sql formatter

Beautify and standardize messy SQL.

ToolText

case converter

camelCase, snake_case, kebab, PascalCase.

rss validator

Check any feed against the spec.

opml converter

Import, view & convert OPML subscription lists.

rss to json

Turn any RSS or Atom feed into clean JSON.

feed preview

Read any feed as a clean, human-friendly list.

ToolEmail

deliverability checker

Check a domain's SPF, DKIM & DMARC, with a score & fix.

ToolNetwork

dns lookup

See any domain's A, MX, TXT, NS & more, live.

ToolNetwork

whois lookup

Registrar, dates, status & nameservers via RDAP.

ReferenceConcepts

Object-Oriented Programming concepts

Encapsulation, abstraction, inheritance, polymorphism, the four pillars, with examples.

ReferenceConcepts

Software design patterns

The classic Gang-of-Four patterns, creational, structural, behavioral, and when to use them.

ReferenceConcurrency

Multithreading & concurrency explained

Threads vs processes, race conditions, locks, deadlocks, and writing thread-safe code.

ReferenceMemory

Memory leaks explained

What a memory leak is, the common causes across languages, and how to find and prevent them.

ReferenceMemory

Smart pointers & RAII

Reference counting, ownership, and how unique/shared pointers stop leaks at the source.

ReferenceC++

Type casting in C++

static_cast, dynamic_cast, const_cast and reinterpret_cast, what each one actually does.

ReferenceC#

C# (C-sharp)

The language and .NET, the type system, and the features that define it: LINQ, async/await, generics, records.

ReferenceData formats

What is JSON?

JavaScript Object Notation: the syntax, the six data types, JSON vs XML, and why it runs every API and AI call.

ReferenceData formats

What is XML?

Elements, attributes, the tree, well-formed vs valid, namespaces, and the XML behind every RSS and Atom feed.

ReferenceFeeds

What is RSS?

Feeds, readers, the orange XML buttons, and how RSS and Atom differ, the basics, clearly.

What is DNS?

The internet's address book: how resolution works, every record type, TTL and propagation, and how to set records up.

What is WHOIS?

Domain registration records: WHOIS vs RDAP, the status codes, why registrant data is redacted, and keeping yours healthy.

HTTP explained

Methods, status codes, headers, and the request/response cycle behind every web call.

How SMTP works (sending email)

The protocol that moves email, the SMTP conversation, envelopes, and why mail gets blocked.

POP3 vs IMAP (reading email)

The two protocols for retrieving mail, how they differ, and which one you actually want.

Why your email goes to spam (deliverability)

The 4 reasons legit mail lands in spam, how SPF, DKIM, and DMARC work together, and how to diagnose and fix it.

DMARC explained

The umbrella over SPF and DKIM: alignment, the none/quarantine/reject policy progression, reports, and a safe rollout.

SPF explained

The record that says who may send for your domain: mechanisms, ~all vs -all, the forwarding break, and the 10-lookup limit.

DKIM explained

Cryptographic email signing: how the signature and selector work, what they protect, and why DKIM survives forwarding.

How ping & ICMP work

What ping really measures, the ICMP echo request/reply, TTL, and reading the results.

Port scanning explained

What a network port is, how a port scan works, the common scan types, and what they reveal.

ReferenceShell

Shell scripting explained

Bash basics, the shebang, variables and quoting, pipes and redirection, and the gotchas that bite.

ReferenceShell

Cron explained

Crontab's five fields, ranges and steps, named shortcuts, the day-of-month vs day-of-week OR gotcha, and why jobs fail.

ReferenceFront-end

How to start CSS

What CSS is, the three ways to add it, selectors, the box model, sane sizing, and a live editor to try it in.

ReferenceAgentic AI

A local-first command center where a coding agent works over your own notes and tools, with saved rules, routines, and verify loops. The real primitives (CLAUDE.md, slash commands, skills, subagents, MCP), the security you need, and a builder that writes a routine for you.

ReferenceVersion control

Git and version control

What version control is, the working-tree, staging, commit, and remote model, the everyday commands, branches, the gotchas (gitignore, secrets, conflicts), and why it is your safety net with AI coding.

ReferenceMonetization

Monetize a YouTube channel with AI

Use AI as your strategist to plan, package, script, and diversify a channel, grounded in the real 2026 Partner Program thresholds, the honest economics, and the AI-content policy that demonetizes 'slop'. With an eligibility checker and 8 rebuilt prompts.

What is a context window?

The model's working memory, tokens, what fills it, why it runs out, and how to manage it.

What are embeddings?

How content becomes a vector that captures meaning, the engine behind semantic search, RAG, and recommendations.

AIInfra

What is a vector database?

Where embeddings live, fast nearest-neighbour search over millions of vectors, and the backbone of RAG.

AIPattern

What is RAG (Retrieval-Augmented Generation)?

Grounding a model in your own documents at query time, embeddings, retrieval, and why it beats fine-tuning for facts.

What is agent memory?

Why an assistant 'remembers' or 'forgets', context vs short-term vs long-term memory, and how persistence is faked.

AIProtocol

What is MCP (Model Context Protocol)?

The open standard for connecting AI apps to tools and data, the 'USB-C for AI tools.'

AIProtocol

What are MCP servers?

The servers you actually plug in: files, databases, GitHub, the web, browsers. The categories that matter, real examples, and how to connect one.

What is tool calling (function calling)?

How a model decides to call a function, returns structured arguments, and your code runs it, the basis of agents.

AIAgents

How to build an AI agent

What actually turns a chatbot into an agent: a model plus tools plus a loop plus memory plus a stop. The five parts, the agent loop, the security you cannot skip, and a builder that writes your agent's spec.

What are structured outputs?

Making a model return JSON that matches a schema every time, vs JSON mode, and why it powers extraction and tool calls.

What is an AI hallucination?

Why models state false things with total confidence, where the risk is highest, why you cannot trust the tone, and the concrete ways to reduce it.

What are reasoning models?

Models that think step by step before answering, trading speed and cost for accuracy on hard problems. Chain of thought, what it is great at, where it is wasted, and the tradeoffs.

AIPractice

What are AI evals?

How to measure whether AI output is good enough: the dataset, the scorer, the metric, and the four ways to score (exact match, rubric, LLM-as-judge, human). The verify gate at scale.

AISecurity

Prompt injection and AI security

Why a model cannot tell your instructions from text it just read, the lethal trifecta that makes agents dangerous, and the architecture-level defenses that actually contain it.

AIPractice

How to prompt Claude Fable 5

Fable 5 is built for long, hard, autonomous work, and that shifts how you prompt it: effort as the main dial, saying less, grounding long runs, setting boundaries, and the migration gotcha that quietly falls back to Opus 4.8.

AITool

Context Window Visualizer

See how your system prompt, history, retrieved docs, and output fill a model's context budget.

AITool

Structured Output Builder

Define fields once; generate the schema for OpenAI, Anthropic, Gemini, and plain JSON Schema.

AIGuide

How to write code with AI

A practical, no-hype method: set expectations, work around the training cutoff, manage context, plan then dictate, test everything, and know when to take over.

AITool

AI Coding Prompt Builder

Assemble a precise coding prompt: pick plan-or-build mode, then toggle signature, types, tests, stable libraries, and fresh docs. Copy-paste ready.

What is workflow automation?

The trigger-to-action model behind every automation, no-code tools like Zapier and n8n, and where AI fits.

AI agents vs workflows vs RPA

Three ways to automate work and how they differ, who decides the steps: you, a UI script, or the model.

Webhooks vs polling

The two ways software learns that something happened, push vs ask, and the latency, cost, and reliability tradeoffs.

What is AI orchestration?

The layer that coordinates models, tools, and data steps into one reliable pipeline, vs a single prompt, workflow, or agent.

What are AI loops?

Giving a model a goal, a way to verify its own work, and a rule for when to stop, so it plans, does, checks, and iterates instead of answering one prompt at a time. The verify gate, when a loop is over-engineering, and a builder for your own.

Automation Blueprint Explorer

Five real automation blueprints broken into clickable steps, trigger, action, AI, human, and integration.

PatternQuality

Reflection pattern

The model critiques its own output, then improves it.

PatternControl flow

Router pattern

The model classifies the request and routes it to the right path.

PatternEvaluation

Judge pattern

One model evaluates another’s output against a rubric (LLM-as-judge).

PatternReliability

Verification pattern

Check the claims before responding.

PatternSafety

Human-in-the-Loop pattern

The agent requests human approval before it acts.

New on KB Cafe: Practice in Every Lesson, Capstones, and a Path Builder

Our biggest update yet: interactive practice interleaved into lessons (try it live in the post), build-this capstones on every path, a streak and XP system, a learning path builder that composes real vetted pages, and two new guides. All free, no account, in your browser.

BlogAI

What Actually Happens When Anthropic Goes Down?

When Claude has an outage, dozens of AI tools degrade at once while their own status pages stay green. What an upstream provider outage looks like, why it's so confusing to debug, and how to tell in seconds whether it's you, your tool, or the model underneath.

New on KB Cafe: Flashcards, TIL, and Free Resources

Three new ways to make AI knowledge stick: spaced-repetition flashcards built from the dictionary, a Today I Learned stream of bite-size snippets, and free downloadable templates with no email wall. All client-side, no account.

Learn to Code With AI, Hands-On

New: a Coding with AI track with an interactive prompt builder and a spot-the-hallucination exercise, an MCP servers lesson with a match-the-server quiz, and a CSS starter with a live editor. Built to be used, not just read.

Meet the Study Window: Read Two Pages at Once

A resizable second window that docks on the right so you can keep the knowledge map, a reference, or a tool open beside whatever you are reading. How it works, how to open it, and when it earns its keep.

New to AI? We Built a Beginner's Guide That Starts From Zero

A complete, no-jargon guide to AI for people with zero background, plus an easy way to share it with anyone who wants to learn the simplest way possible.

How an AI Agent Blanked 26 Files, and the One Question That Saved Me

A postmortem: an AI coding agent null-wrote 26 of my source files. Why git only saved half, how I rebuilt the rest from build output and verified every byte, and the rule I locked in.

How We Built the Prompt Optimizer: Six Tools Into One

We turned six focused prompt tools into one flagship, without deleting any of them. The real product decision behind composing tools into a mega-tool.

Introducing AI Business, and Building in Public

We've opened a new section, AI Business, to cover the half of the journey nobody teaches honestly: getting traffic and making money. Plus our first case study is live.

BlogOpen Web

The Blog That Used to Live at This Address

Before KB Cafe was an AI knowledge base, this domain ran one of the early web's RSS blogs. The story of iBLOGthere4iM, and where it lives on.

The Return of KB Cafe

kbcafe.com was a go-to developer resource in the early 2000s. Here's what it is now, and why we brought it back.

BlogOpen Web

RSS Didn't Die, The Open Web Just Got Quiet

Everyone declared RSS dead a decade ago. It's quietly running newsletters, podcasts, and the fediverse the whole time.

BlogAI

Feeds for Machines: RSS as the Structured Layer for AI

LLMs are starving for clean, structured, permissioned content. A 25-year-old format turns out to be a good answer.

Token

A token is a small chunk of text an AI reads at once, usually a short word or a piece of one.

Context window

A context window is how much text an AI can keep in mind at once, its short-term memory, measured in tokens.

Prompt

A prompt is simply what you ask an AI to do: the instructions or question you type in.

AI agent

An AI agent is an AI that can take actions on its own: not just answer, but use tools, run steps, and work toward a goal.

Embedding

An embedding turns text (or images) into a list of numbers that captures meaning, so a computer can compare ideas.

Vector database

A vector database stores embeddings so an AI can quickly find the most similar pieces of information by meaning.

DNS

DNS (Domain Name System) is the internet's phonebook: it turns a name like kbcafe.com into the numeric address computers use.

Domain

A domain is your website's human-friendly address, like kbcafe.com, that you rent and point at a server.

URL

A URL is the full web address of a specific page, like https://kbcafe.com/dictionary.

HTTP

HTTP is the language browsers and servers use to talk. Every page, image, and API call is an HTTP request and response.

SSL / TLS

SSL (now properly called TLS) is the encryption that puts the padlock in your browser and the "s" in https.

Browser

A browser is the app that turns web code into the pages you see, like Chrome, Safari, Firefox, or Edge.

Cookie

A cookie is a small note a website saves in your browser to remember you between visits, like staying logged in.

Model

A model is the trained AI itself, the thing that actually generates answers. GPT-5, Claude, and Gemini are all models.

LLM

An LLM (large language model) is an AI trained on huge amounts of text to predict the next word, which is how it writes and reasons.

Inference

Inference is the moment an AI model actually runs to produce an answer, as opposed to being trained.

Fine-tuning

Fine-tuning is taking a general model and training it a bit more on your own examples so it specializes.

RAG

RAG (retrieval-augmented generation) lets an AI look things up before answering, so it can use your documents instead of only its training.

MCP

MCP (Model Context Protocol) is a standard way to plug tools and data into an AI, so any model can use any connected tool.

Tool calling

Tool calling is when an AI decides to use an external tool (search, a calculator, your API) instead of answering from memory.

Structured outputs

Structured outputs force an AI to answer in a strict format like JSON, so your code can use the result reliably.

Hallucination

A hallucination is when an AI states something false with total confidence, inventing facts, sources, or quotes.

Conversational AI

Conversational AI is AI you talk to in plain language, back and forth, like ChatGPT or a voice assistant.

Temperature

Temperature is a setting that controls how random or focused an AI's answers are.

Multimodal

Multimodal AI can handle more than one kind of input or output: text, images, sound, even video.

Diffusion model

A diffusion model is the kind of AI that makes images. It starts from random noise and refines it into a picture based on your prompt.

API

An API is a way for two pieces of software to talk: one asks for something, the other answers, over the web.

Deploy

Deploy means to put your code live on the internet so other people can actually use it.

Repository

A repository (or "repo") is the folder that holds all your project's code and its full history of changes.

Git

Git is the tool that tracks every change to your code, so you can rewind, branch off, and work with others without chaos.

Commit

A commit is a saved checkpoint of your code: a snapshot, with a short note, that you can always return to.

Branch

A branch is a parallel copy of your code where you can try changes safely, without touching the working version.

Pull request

A pull request (PR) is a proposal to merge your changes in, so others can review them before they go live.

Environment variable

An environment variable is a setting stored outside your code, used for secrets and config like API keys.

CLI

A CLI (command-line interface) lets you control software by typing commands instead of clicking buttons.

Terminal

A terminal is the window where you type commands to your computer, the place a CLI lives.

Package

A package is a bundle of prewritten code you can drop into your project so you do not build everything yourself.

Dependency

A dependency is a package your project relies on to work. If it is missing, your app breaks.

Framework

A framework is a ready-made foundation for building apps, with structure and tools so you do not start from zero.

Frontend

The frontend is the part of an app you see and click: the screens, buttons, and layout in your browser.

Backend

The backend is the part of an app you do not see: the server, the database, and the logic behind the scenes.

Database

A database is where an app stores its information so it is saved, organized, and findable later.

MVP

An MVP (minimum viable product) is the smallest version of your idea that is still useful enough to launch and learn from.

SaaS

SaaS (software as a service) is software you pay for by subscription and use in your browser, instead of buying and installing it.

Freemium

Freemium is a model where the basic product is free and you pay to unlock more, the most common way software grows.

Conversion rate

A conversion rate is the percentage of people who take the action you want, like signing up or paying.

CAC

CAC (customer acquisition cost) is how much you spend, on average, to get one new paying customer.

Churn

Churn is the rate at which customers cancel and leave. High churn quietly drains a business.

ARR

ARR (annual recurring revenue) is the predictable subscription revenue a business earns in a year.

LTV

LTV (lifetime value) is how much money one customer brings in, on average, over the whole time they stay.

Product-market fit

Product-market fit is the moment a product clearly solves a real need, so people adopt it and tell others on their own.

Hosting

Hosting is renting space on an always-on computer so your website or app stays available on the internet.

Server

A server is a computer that runs all the time and answers requests from other devices over the internet.

CDN

A CDN (content delivery network) stores copies of your site around the world so it loads fast for everyone.

IP address

An IP address is the numeric address of a device on a network, like a phone number for computers.

Webhook

A webhook is an automatic message one app sends another the instant something happens, so nothing has to keep checking.

OAuth

OAuth is the "Log in with Google" system: it lets one app access another on your behalf without sharing your password.

Vibe coding

Vibe coding is building software mostly by describing what you want in plain English and letting an AI write the actual code.

Localhost

Localhost is a private web address that shows your app only on your own computer, so you can preview it before anyone else can.

Dev server

A dev server is a program on your computer that shows a live, instantly-updating preview of your app while you build it.

Build

A build is the step where the computer packages all your code and files into a finished version ready to put online.

Scaffolding

Scaffolding is the empty starter skeleton of an app, the folders and blank files, that the AI sets up first before any real features.

Boilerplate

Boilerplate is routine, repetitive code that nearly every app needs, so you generate or copy it rather than writing it fresh.

Refactor

To refactor is to clean up and reorganize existing code without changing what the app actually does.

API key

An API key is a secret password that lets your app use an outside service, like an AI model, proving the request comes from your account.

Port

A port is the numbered "door" on your computer (the 3000 in localhost:3000) that a running app uses to talk to your browser.

Console

The console is a running text feed where your app and browser print messages, warnings, and errors about what is happening underneath.

Stack trace

A stack trace is the wall of red text shown when an app crashes, listing the chain of steps that led to the error.

Authentication

Authentication (or "auth") is the system that checks who a user is when they log in, so the right person sees the right things.

Schema

A schema is the blueprint that defines what information your app stores and how it is organized into tables and fields.

Rate limit

A rate limit is a cap on how many requests you can make to a service in a short time before it tells you to slow down.

Context rot

Context rot is when an AI coding assistant slowly loses track of earlier details in a long session and starts contradicting itself or inventing things.