AI Place Hub
Sign in Create account

Learn how to actually talk to AI models.

Prompts, configuration files, whole workflows, tool reviews and model limits for Claude, GPT, Gemini, Llama and local models. Everything here was written by someone who uses it, and says when they last checked it.

12 prompts  /  6 configs  /  8 workflows and guides  /  18 terms

one prompt, one answer

you

model

Build one right now

Tap the parts you want. The prompt below writes itself, and you can copy it straight into any model.

Who should the model be

What comes back

Rules to add

Your prompt Copied

A prompt worth reading twice

A strict extraction prompt that returns valid JSON only, with nulls for missing fields instead of invented values.

PromptCopied
Extract the fields listed below from the text and return JSON only.

Fields:
{{fields}}

Rules:
- Return a single JSON object, no markdown fences, no commentary before or after.
- If a field is not present in the text, set it to null. Never infer it.
- Keep the original wording for text fields. Do not normalise, translate or tidy.
- If the text contains several records, return an array of objects instead.

--- text ---
{{text}}

Run this at temperature 0. The no-fences rule is worth keeping even on models with a JSON mode, because it makes the same prompt portable. Validate the output against a schema on your side - a prompt is not a guarantee.

Open Extract structured data as JSON

Drag to see the difference

Same task, same model, two minutes apart. On the left is what most people type. On the right is what gets the answer they wanted.

Specific

Rewrite this product description for a Slovak audience. Keep it under 60 words, keep the two technical specs exactly as they are, and drop the marketing adjectives.

Vague

Make this text better please.

Trending this week

Browse everything

What you will find here

Eight sections, all community maintained. Every entry says which model it was tested on and when, and readers can flag anything that stopped working.

Add your own
Prompts Single reusable prompts, filed by task and by the model they were tested on. 12 published System instructions Standing instructions that tell the model who to be. Paste once into settings and every answer changes. 5 published Configs CLAUDE.md, AGENTS.md, Modelfiles, editor rules and parameter presets. Files, not personas. 6 published Cheatsheets Slash commands, keyboard shortcuts and formatting tricks, as a table you scan in ten seconds. 3 published Workflows Multi-step procedures that chain several prompts into one finished job. 2 published Tools Apps, extensions, CLI tools and MCP servers, reviewed by people who run them. 2 published Errors and fixes The exact message you got, what causes it, and what actually fixes it. 2 published Before and after The same task asked badly and asked well, with the output both produced. 1 published Local setup Ollama, LM Studio, hardware and quantisation, without an API bill. 1 published Guides Longer explanations of why models behave the way they do. 6 published
Model rules Context windows, output caps, prices per million tokens and what each model refuses. 12 models tracked AI crawlers Who reads your site, what they do with it, and a robots.txt builder that keeps training and citation apart. For webmasters Instruction builder Answer a few questions, get a system prompt to paste into your settings. No API, nothing sent anywhere. Takes two minutes Collections Sets someone assembled for one job and shares as a single link. Make your own AI systems One page per model or platform, with everything written for it. 9 systems Glossary Plain definitions of the terms that show up in every AI discussion. 18 terms

AI systems at a glance

What each one is best at, which file holds its rules, and whether it runs on your own hardware.

All systems
System Best at Config file Runs locally Resources
ChatGPT Tool use, structured output AGENTS.md No 16
Claude Long code review, editing, writing CLAUDE.md No 13
Gemini Very long context, multimodal input System instruction No 3
Grok Recent events, informal drafting - No 0
Llama Private data, offline work Modelfile Yes 6
Mistral Fast extraction and classification Modelfile Yes 3
Midjourney Stylised image generation - No 1
Stable Diffusion Reproducible local image generation - Yes 1

Browse by what you are doing

Writing 4 Coding 12 SEO 1 Marketing 2 Research 0 Productivity 7 Data Analysis 2 Image Generation 1 Education 7 Automation 3 Business 1 Design 0

Recently published

More

Start here

All guides

The words everyone uses

Definitions written for people who need to make a decision, not pass an exam.

Full glossary

  • Agent

    A model given tools and a goal, allowed to decide its own next steps in a loop rather than answering once.

  • Chain of thought CoT

    Prompting a model to write out its intermediate steps before the final answer.

  • Context window

    The maximum amount of text, measured in tokens, that a model can take into account at once - your prompt and its answer together.

  • Embeddings

    Numeric vectors representing text, where similar meanings land close together - the basis of semantic search.

  • Few-shot prompting

    Showing the model two or three worked examples of the task inside the prompt, instead of only describing it.

  • Fine-tuning

    Further training of an existing model on your own examples, to change its behaviour rather than its knowledge.

  • Hallucination

    A confident, fluent answer that is simply wrong - an invented citation, function, statistic or fact.

  • Inference

    Running a trained model to produce output - what happens every time you send a prompt.

People behind the library

AI Place Hub 40

Share what works for you

If a prompt or a config saves you time, it will save someone else time too. Submissions are reviewed before they go live, and your name stays on them.

Share a prompt or config