Methodology

Wonder, with the wires labeled

MagicClaude is built for people who want to show what Claude can do without exaggerating, hiding assumptions, or making uncited product claims.

What we cover

MagicClaude covers one territory: impressive Claude demos that a reader can reproduce. A page belongs here when it has a prompt, setup notes, expected reveal, verification steps, and source links. The Stage at /stage/is the signature spotlight: one sourced wow-demo, not the whole cabinet.

What we do not cover

We do not publish weekly Claude news, generalized beginner guides, Claude Code engineering doctrine, memory systems, broad context-engineering theory, or edge-case experiments. Those belong elsewhere in the Claude Network.

We also do not benchmark. Demos here name the model they run on, because that is what makes a prompt reproducible, but MagicClaude publishes no scores, no rankings, and no measurements of its own. Model comparisons with methodology belong to Claude Reports, and the general question of which model to use belongs toClaude Central.

How we keep model names current

The fastest-decaying thing on a demo site is the model name. The current lineup is Claude Fable 5.1, Claude Opus 5, Claude Sonnet 5, and Claude Haiku 4.5, and it is recorded on the casting page with API IDs and a link to Anthropic's model overview. When a model ships or retires, the casting note on every affected demo is updated and the change is logged on the freshness page.

No page on this site reproduces a Claude transcript. Demo pages describe what the prompt should produce and how to check it, which is a claim a reader can falsify by running the prompt.

Claude Network links

This site is part of the Claude Network. Start at Claude Central, then use the sibling publications below when a topic leaves MagicClaude's demo lane.

Disclaimer

MagicClaude is an independent educational publication and is not affiliated with Anthropic. Product availability, limits, and model behavior can change. Always check official Anthropic sources before relying on a demo for training, procurement, or production workflows.

Primary Sources

These demos cite official Anthropic docs or help-center pages for capability claims. They do not claim identical output across accounts, plans, models, or dates.

Anthropic Docs

Claude model overview

The model overview is the canonical list of generally available Claude models, API IDs, context limits, and retirement dates.

Claude Platform Docs

Choosing a model

Anthropic guidance is to start with Claude Opus 5 for most work, move to Claude Fable 5.1 for demanding reasoning and long-horizon agentic tasks, use Sonnet 5 for the speed/intelligence balance, and Haiku 4.5 for the fastest and cheapest runs.

Claude Code Docs

Claude Code overview

Claude Code is an agentic coding tool available in terminal, IDE, desktop app, and browser surfaces.

Claude Help Center

Let Claude use your computer in Cowork

Computer use lets Claude interact with permitted apps and files, but Anthropic warns users to avoid sensitive apps and monitor behavior.