Frequently Asked Questions

No experience needed. Frompting handles the framework structure for you — you just describe what you want to achieve. The Prompt Engineering Guide and Framework Library are both available free if you want to learn the theory behind the templates.

Any of them. Frompting generates text prompts — you copy and paste into whichever AI you use:

• ChatGPT (GPT-4o, GPT-4, GPT-3.5)
• Claude (Claude Sonnet, Claude Haiku)
• Gemini (Google)
• Grok (xAI)
• Llama, Mistral and other open-source models
• Any LLM that accepts text input

Frameworks are model-agnostic — the structured approach improves results regardless of which AI you use.

A few ways to use your prompts faster: the "Open in AI" buttons send a generated prompt straight to ChatGPT, Claude, Gemini & more in one click; our Chrome extension lets you enhance prompts right inside those tools and carry a whole conversation from one AI to another with Context Snapshots; and on Premium/Team you can run the prompt in-app with your own API key and get the answer without leaving Frompting.

It puts Frompting inside the AI tools you already use. A button appears beside the chat box on ChatGPT, Claude, Gemini, Perplexity, Grok and 13 more sites: type your rough idea, click it, and a structured prompt built on a real framework replaces what you typed. It also refines prompts in place, keeps a searchable history, saves prompts to your account, and — the part no chat site does for you — saves a whole conversation as a Context Snapshot so you can continue it anywhere. Type f/ in any chat box for the command menu.

Every feature with a screenshot: the complete extension guide.

A file holding an entire conversation — every message in order, with the documents that were attached converted to text and the pictures carried along — saved on your device. Type f/save in the chat you want to keep. Then open any chat, anywhere — a fresh one on the same site, a different AI, a different account, next week — type f/load, pick the snapshot, add your question and send. The new session has every decision, every draft and every file from the old one.

f/save is built and verified on ChatGPT, Claude, Gemini, Perplexity and Grok; f/load, f/prompt and f/enhance work on every site the extension runs on. Saving on other sites is not offered yet. Nothing is uploaded to Frompting; the file stays on your machine.

• Your free allowance runs out mid-project. Save, open another AI you have, load, and carry on. You lose nothing but the tab.
• The chat got too long and the AI has forgotten what you agreed early on. Save, open a fresh chat, load: the whole history in one clean file, read with a fresh context window.
• Draft in one AI, review in another. ChatGPT wrote it; you want Claude's read. Load the snapshot into Claude and ask what it would change.
• A reusable context pack. One conversation establishing your company, product and voice; load it at the start of every new chat.
• Hand a conversation to a colleague as a .md file, or keep a record of the chats that matter, with the attached documents inside.
• Before an upload expires. ChatGPT removes uploaded files after about an hour; snapshot while they are fresh and the document's text is kept for good.

Saving and loading are free for everyone, on every plan, and the snapshots are complete — documents, pictures, all of it. Premium adds the snapshot library: a page where you browse, search inside, read and download every snapshot as .md, and set a sync folder on your machine that Chrome, Edge and Brave all share.

Type f/ at the end of what is in the chat box and a menu appears — arrow keys to move, Enter to choose. Seven commands: f/save saves this conversation as a snapshot; f/load attaches a saved snapshot to this chat; f/switch moves this chat to another AI and says what it should do there; f/search searches inside your snapshots; f/folder switches or reconnects your snapshots folder; f/prompt inserts one of your saved or favourite prompts; f/enhance turns what you have typed into a better prompt (put your text first, then the command). Settings — the gear beside the Frompting button — lists them all with full descriptions and examples.

It reads what you type in the chat box only when you press Enhance, a refine, or an f/ command — never on its own. A snapshot is the one feature that reads the conversation on the page, and only when you run f/save; the result is saved on your device and is never uploaded to us. When you f/load, the snapshot goes to the AI site you are on as a file you attach, the same as dragging a file in yourself. The only server the extension talks to is Frompting's, to write prompts and to check your plan. Details in the privacy policy.

Refresh the tab once — the browser only injects extensions into pages opened after the install. If the site is not one of the 18 built-in ones, click the Frompting icon in the browser toolbar and press Enable here. If you dragged the buttons somewhere odd, double-click the dotted handle to put them back.

A prompt framework is a structured template that ensures your AI prompt includes everything the model needs: the right role, clear context, a specific goal, and the expected output format. Frameworks like RACE (Role-Action-Context-Expectation) and CO-STAR (Context-Objective-Style-Tone-Audience-Response) consistently outperform unstructured prompts because they eliminate ambiguity.

Browse the full Framework Library to see all 60+ with explanations and templates.

Frompting includes 60+ frameworks across 7 domains:

• AI & Prompt Engineering — CO-STAR, Chain of Thought, Tree of Thoughts, RTF, and more
• Communication & Storytelling — APE, CRISP, PREP, CLEAR, and more
• Persuasion & Conversion — AIDA, PAS, BAB, SPIN, PASTOR, and more
• Strategy & Planning — SCQA, MECE, SWOT, SMART, OODA, and more
• People & Teams — STAR, SOAR, RISE, ROSES, and more
• Product & Growth — RACE, FAB, BINGO, FOCUS, and more
• Reasoning & Analysis — COT, REACT, SCOPE

6 frameworks are free; 54 are on paid plans.

It depends on your goal:

• Writing copy or marketing content? → AIDA, PAS, or BAB
• Complex AI instructions? → CO-STAR or RTF
• Interview prep or performance reviews? → STAR or SOAR
• Business analysis or strategy? → SCQA, MECE, or SWOT
• Multi-step reasoning problems? → Chain of Thought or Tree of Thoughts
• General ChatGPT tasks? → RACE or CRISP

Not sure? Frompting can suggest the right framework based on your use case.

Yes — the free plan gives you access to 6 frameworks with a daily prompt limit. No credit card required to sign up. The free tier is designed for light use or trying Frompting before committing to a plan.

Two things: the AI models behind your prompts, and the features you unlock.

• Framework access — the big one. Free includes 6 frameworks; Premium unlocks all 60+, automatically matched to what you're writing. The framework is what shapes a prompt into something a model can actually follow, so this is the difference you'll feel most.
• Same AI model, either way. We don't hold back model quality on the free tier — free and guest prompts run on the same premium model our paying customers get, so what you see is our real quality. At busy times a free generation may take a few seconds longer, because paid plans get priority capacity and never queue.
• More volume. Guests get 15 prompts a week (up to 3 a day) and free accounts get 30 a week (up to 6 a day) — Premium raises this to 1,500 a day.
• Premium-only features. Image & video prompts, Brand Kit (on-brand voice + consistent characters), Saved Prompts, one-click refine, and run-in-app with your own API key.

In short: free is perfect for trying Frompting and light daily use; Premium is for people who want the best possible prompts and the full toolkit. See the Pricing page for a full comparison.

The Pro plan unlocks all 60+ frameworks and unlimited text, image & video prompts, plus several power features:

• Brand Kit — save your brand voice and a visual character/style so every prompt stays on-brand
• Run in-app (BYOK) — run a prompt with your own API key and get the answer inside Frompting, then edit & export it
• Saved presets — reuse your favourite generator settings in one click
• Prompt history & saving, analytics, priority AI models & support

See the Pricing page for current rates and a full plan comparison.

Yes. You can cancel your subscription at any time from your dashboard. Your Pro access continues until the end of the billing period — no prorated charges.

Yes. The Team plan includes everything in Pro plus multi-seat access, shared prompt history, team usage analytics, and higher concurrent session limits. It's designed for agencies, consultancies, and AI-forward teams. See the Pricing page for details.

LLMs perform best when given role, context, goal, and output format — explicitly. Most people skip most of these. A plain prompt like "write me an email" leaves the AI to guess audience, tone, length, and purpose. A RACE-structured prompt specifying Role, Action, Context, and Expectation gives the model everything it needs, producing a first draft that needs far less editing.

Try three things: (1) hit Refine under the generated prompt to auto-improve it — you can add a steer like "make it more specific" or "shorter"; (2) add more detail to your input — the more context you give Frompting, the more tailored the output; (3) try a different framework for the same goal — AIDA and PAS both work for persuasive copy but take different angles. You can also edit the generated prompt directly before pasting it into your AI tool.

Start with the Prompt Engineering Guide — it covers the fundamentals from first principles. The Framework Library has individual pages for all 60+ frameworks with step-by-step breakdowns. The ChatGPT Prompts page has 60+ ready-to-use prompts by category.