thinkhard.SI
Sign in Add API key

Get better answers from AI.

Free tools for working with ChatGPT, Claude and Gemini. Fix a weak prompt or start from one of 669 templates, then make the draft sound like you, check its facts and keep private details out of the chat.

Check a prompt Browse templates

Free, with no account to make. Your prompts stay in your browser. New to this? Start here.

Your prompt’s score inside ChatGPT and Claude, with Prompt Score for Chrome

37 words Example
5 A startout of 10

Read on this page as you type; nothing is sent anywhere. On a phone, share text here from any app.

What do you need to do?

Pick the job. Each tool does one, and most of them work without an API key.

Every tool, by what you want to do, with a finder that takes your own words. The newest tool, the Prompt Improver, rebuilds a rough prompt in one step.

For building with AI: Model Compare, Prompt Tester, Chatbot Safety Test, Token Counter, AI Cost Calculator, Version Compare, Prompt Linter, Library Audit, Agent Brief Check, My Prompts.

Start from a prompt that works

669 expert prompts, grouped by the work they do. Each has blanks for your own facts and a note on how to check the answer, and works in ChatGPT, Claude, Gemini and Copilot.

The same request, before and after

Before

Write a launch email for our new invoicing feature. Make it good and not too long.

1/10 Needs work

  • Has: What to make
  • Missing: Who it's for
  • Missing: Background facts
  • Missing: Length
  • Missing: Format
  • Missing: What to avoid
  • Missing: What good looks like
  • Missing: An example
  • Missing: If unsure
  • Missing: Layout

After, with what was added marked

<context> We are launching [FEATURE NAME]. What it does: [WHAT IT DOES]. The email goes to existing customers who give it about ten seconds. </context> <task> Write the launch email. Its one aim is a click on Try it, so lead with the customer's problem. </task> <constraints> - Body of 90 to 130 words, because most readers will not scroll. - Use only the facts above. Where a detail is missing, leave a note in brackets. </constraints>

6/10 Good

  • Has: What to make
  • Has: Who it's for
  • Has: Background facts
  • Has: Length
  • Missing: Format
  • Has: What to avoid
  • Missing: What good looks like
  • Missing: An example
  • Missing: If unsure
  • Has: Layout

Scored with the Prompt Grader's ten checks. The strong version leaves blanks for the facts only you know.

Pick your work

Packs for the job at hand: a presentation, an assistant’s week, report cards and performance reviews.

All templates By team and industry The prompt course

Your words go only where you send them

Most AI apps pass your words through their own server and keep a copy. Here, most tools run on your device: thinkhard.si sends the page, and when you run a prompt your browser talks to the AI company directly, with your own key. The engine tools and Pro’s included AI are the exceptions you choose: they send your text to us only when you press their button, and we read it to answer you and keep none of it.

A typical AI appChat apps and AI writing tools
thinkhard.siMost tools, with your own key
An engine toolOnly when you press its button
Youyour device
Writes it
Writes it, and keeps the results
Writes it, then presses the button
The app’s serverthe company behind the app
Reads every prompt keeps a copy
Never sees it. thinkhard.si sends the page, then steps aside
Reads it to answer keeps nothing
The AI companyOpenAI, Anthropic, Google…
Reads it, on the app’s account
Reads it, straight from your browser with your own key
Not involved: our engine answers on its own
Copies the app keeps
Every prompt
None
None
Who sees your prompt. In a typical AI app, you write it, the app’s server reads every prompt and keeps a copy, and the AI company reads it on the app’s account. With thinkhard.si, the page is sent once and your browser then talks to the AI company directly with your own key, so the prompt never reaches thinkhard.si and no copy is kept. With an engine tool, pressing its button sends the text to our engine, which reads it to answer and keeps nothing; no AI company is involved.
What exactly reaches us, and what stays with you

Kept in your browser

  • Your API keys, sent only to the provider they belong to. Turn off “Remember” to keep a key for this tab only.
  • Prompts, test suites, runs and grades.
  • Drafts and settings. Export everything as one file, or delete it, from Settings.

What reaches thinkhard.si

  • The request for each page, as with any website.
  • Anonymous daily totals of page views and tool runs. What is counted.
  • Text you send with an engine tool’s button or Pro’s included AI: read to answer you, then forgotten. What is sent.
  • Your saved work, only if you have Pro and turn on sync.
  • Whatever you type into the waitlist, contact or repository unlock form.

No ads and no trackers, and no scripts from other companies except Paddle’s, on the checkout page. An account, and its cookies, only if you sign in. Free keys from Google AI Studio, OpenRouter and Groq make every tool free to use.

Free stays free.

Pro adds included AI runs and sync across your devices for $8 a month when it opens. Team is next: one library of approved prompts, one click away in everyone’s assistant.

See the plans

Fair questions

Do I need to be technical to use this?

No. If you can write an email, you can use the templates, the Prompt Grader’s score and the AI Cost Calculator. Underlined words open a plain explanation, and the Start here page points you to the right tool in a minute.

What is an API key, and do I need one?

It is a code from an AI company, such as Google, Anthropic or OpenAI, that lets these tools send your prompt to their model, billed to your own account, usually a fraction of a cent per run. You only need one to run prompts on real models. Google AI Studio, OpenRouter and Groq give free ones. The full explanation.

Is it really free?

Yes. Most tools run in your browser, and the engine behind the rest is light to run, so serving them costs very little. Model calls are billed by your provider on your own key. Google AI Studio, OpenRouter and Groq all give free keys with free models, so you can use every tool without paying anyone.

What does the score out of 10 mean?

A count, not an opinion. A prompt meets ten checks: what to make, who it is for, background facts, length, format, what to avoid, what good looks like, an example, what to do if unsure, and a layout the AI reads well. Each one it covers adds a point. Each score has a thinking level, from Guess at 1, where the AI has to guess nearly everything, to Superintelligent at 10, where it has nothing left to guess.

How will this make money?

Through paid plans for what needs a server. Team, being shaped now, shares a team’s tested prompts inside the AI tools it already uses; a later plan for builders adds scheduled test runs, prompt CI and managed credits. The free tools stay free. See the plans.

Do you train on my prompts?

No. Most tools never send your prompts to us, and the few that do, when you press their button, read them only to answer you and keep nothing. What happens to your prompts at the model provider is set by that provider’s terms. Paid API traffic is generally not used for training, but some free tiers and free models may log or use what you send, so keep sensitive data off free keys and check your provider’s policy.