droprompt/about v2026.07

Rough idea in. Production-ready out.

droPrompt is a Cairo-built platform that turns plain-language ideas into model-tuned prompts, AI agents, workflows, and automations — across 60+ AI models for image, video, chat, music, code, and UI/UX. Write rough, in English or Arabic. Get something ready to paste.

rough idea → draft → critique → refine → ready to paste
60+
AI models with tuned prompt specs — image, video, chat, code, UI/UX
5
Pipeline stages behind every generation — draft, critique, refine, tune, verify
11
Interface languages, natively reviewed — Arabic & English first-class
145
Production-tested prompts in the free library, across 10 categories
01
Why it exists

Most people don't want to learn prompting. They want a result.

Every AI model has its own quirks, syntax, and sweet-spot phrasing — and a new one launches every week. If your job is prompt engineering, that's craft. For everyone else, it's friction between an idea and its output.

droPrompt is the opinionated layer in between. You describe what you want in plain language — English or Arabic, rough is fine — and a multi-stage pipeline drafts, critiques, and refines a prompt tuned to the exact model you're targeting. The same engine now goes further than prompts: it builds AI agents on a visual canvas, composes multi-step workflows you can run live, and designs automations you can export straight to n8n, Make, Zapier, Pipedream, or Activepieces.

We're deliberately small. No framework bloat, no enterprise theatre, no sparkle-emoji AI branding. Just a sharp tool that ships fast and gets out of the way.

02
The platform

One engine, ten surfaces.

Everything runs on the same pipeline — pick the surface that matches what you're making.

01 Prompt Builder The core: rough sentence in, model-tuned prompt out — with refine loops and batch series. → 02 AI Agents New Design agents on a visual canvas, then prove they work in the built-in test bench before export. → 03 AI Workflows New Chain multi-step, multi-model workflows and run them live on the canvas — node by node. → 04 Automations Describe an automation in plain language; export it to n8n, Make, Zapier, Pipedream, or Activepieces. → 05 Community Feed New Publish and discover prompts, workflows, and automations — with source attribution built in. → 06 droPrompt for Coding Turn a project brief into a code-ready prompt for Claude Code, Cursor, Codex, and friends. → 07 Skill Builder Template-driven composer for repeatable, structured prompts and skill files. → 08 Prompt Library 145 production-tested prompts across 10 categories — free to browse and open in any builder. → 09 MCP Server Use droPrompt from inside Claude, Cursor, or ChatGPT via the Model Context Protocol. → 10 Apple Shortcut Generate prompts from anywhere on iOS and macOS with the native shortcut. →
03
Changelog

What's new — shipped, not promised.

We ship weekly. The highlights from the last stretch:

Jul 2026 New
Art Style extraction on the Feed
Any image post in the Community Feed now has an "Art style only" button — vision reads the post's actual image and captures its full aesthetic stack (lighting, palette, composition) without copying the subject.
Jul 2026 New
11 interface languages, natively reviewed
The whole interface now runs in 11 languages — each pass reviewed by native speakers, with a language picker right in the header.
Jul 2026 New
Agents got cheaper and easier to test
Agent builds now cost half the credits, extensions too — plus manual canvas zoom and a free test allowance so you can verify an agent before spending anything.
Jun 2026
Live workflow runs on the canvas
The Workflows Builder runs your chain node-by-node with live states, independent verification of each step, retries, and resume.
Jun 2026
Voice input, everywhere
Every builder accepts voice input — speak your idea instead of typing it, in English or Arabic.
Jun 2026
AI Agents joined the library
Agents became a top-level category with 50 ready agent prompts, "Open in Builder" buttons, and shareable feed permalinks.
04
How we build

Six rules we actually follow.

Product decisions are easy when your principles are strict. Every feature passes these filters before it ships.

01
Speed is the product.
Every UI choice gets judged on one axis: does it shorten time-to-result? If it doesn't, it's clutter — regardless of how good it looks in a screenshot.
02
Sharp, not decorated.
We skip the sparkle gradients and decorative chrome that plague AI tools. Tight spacing, clean hierarchy, no filler. The interface should feel like a blade.
03
Surprise through output.
The "wow" has to come from the generated result, not from UI tricks. When someone says "wait, that's actually perfect" — that's the whole product working.
04
Bilingual by design.
Arabic isn't a translation layer bolted on later — it's a first-class input language, with MENA-first defaults and UTC+3 resets.
05
Refund when we fail.
If the AI pipeline errors or returns nothing, the credits are refunded automatically. You never pay for a failed generation — not at the quota level, not at the billing level.
06
Ship weekly.
New models, bug fixes, UX tweaks — every week. Keep the surface small, keep the cadence high. Speed compounds trust.
05
Say hi

We read every message.

Feedback, bugs, feature ideas, partnerships — if it helps us ship better, send it over.

Email
[email protected]24-hour reply window
Support
Help centerGuides, FAQs & troubleshooting
HQ
Cairo, EgyptBuilt on Cairo time · UTC+3

Try it with an idea you're stuck on.

Rough sentence in, production-ready result out. That's the whole pitch.

See pricing Open the app →

Last updated: 2026-07-02