AI Prompt Organizer — Save, Tag, and Reuse Your Prompts With Version History

Save your AI prompts with folders, tags, fill-in-the-blank variables, and full version history — free, stored only in your browser, export a backup any time.

If you write the same kinds of prompts over and over — for ChatGPT, Claude, or any other AI tool — this organizes them in one place: folders, tags, search, {{variable}} fill-ins, and a version history you can revert, all stored locally in your own browser.

How to use it

Click “+ New prompt” and write your prompt in the body field.
Optional: drop in {{tone}}, {{topic}}, or any {{variable}} — a fill-in field appears below the body, and the resolved preview updates as you type.
Tag it (comma-separated) and file it into a folder if you want to browse by category later.
Click “Copy resolved” to copy the filled-in version, or “Copy raw” to copy the template as-is.
Edit the prompt any time — each time you click away after a change, the previous version is saved to that prompt’s history, so you can revert if a rewrite doesn’t work out.
Export a backup as JSON (full data) or Markdown (readable) whenever you want a copy outside the browser.

Features

Folders and tags — file prompts by project or client, and filter by tag chips that build themselves from whatever tags you’ve used.
{{variable}} fill-ins — write a template once, fill in the specifics each time you reuse it, with a live resolved preview so you can see exactly what will get copied.
Version history — every meaningful edit is saved automatically when you click away from the body field, with a one-click revert; nothing is a one-way rewrite.
Duplicate — fork a prompt as a starting point for a variant without losing the original.
Status labels — mark a prompt Draft, Working, or Retired so a half-finished idea doesn’t get mixed in with the ones you actually use.
Search and sort — search across titles and bodies, and sort by recently updated, recently used, or title.
Rough token estimate — a quick characters ÷ 4 estimate next to the character count, useful for a sanity check against a model’s context window.
Export and import — back up everything as JSON, export a readable Markdown file, or copy any single prompt as Markdown. Importing always adds prompts as new entries — it never overwrites your existing library.
Nothing leaves your browser — no account, no server, no sign-in. Your prompts live in this browser’s local storage until you export them.

FAQ

Where are my prompts actually stored?

In this browser’s local storage, on this device. They are not uploaded anywhere. Clearing your browser’s site data for this page will delete them — export a JSON backup periodically if your library matters to you.

What happens if I revert to an old version?

The version you’re currently on is saved into history first, then the body is replaced with the older version. You can revert again later to undo the revert — nothing is discarded, up to the last 20 saved versions per prompt.

Does importing a backup overwrite my current prompts?

No. Imported prompts and folders are always added as new entries, even if you’re re-importing the same backup twice. If you want a clean re-import, delete the duplicates manually afterward.

Can I use the same variable name in more than one prompt?

Yes — variables are stored per prompt, not shared globally, so {{name}} in one prompt is independent of {{name}} in another.

Does the tool call any AI model to run my prompts?

No. It only stores and formats text. You copy the resolved prompt and paste it into whichever AI tool you’re using.

Is there a limit to how many prompts I can save?

Only your browser’s local storage limit, which is typically a few megabytes per site. The tool shows how much space your library is using, and warns you if a save fails so you can export a backup instead of losing the change.