Simplify an English paragraph by replacing complex words and splitting long sentences, with Flesch before and after
Prompt Builder
Build a structured prompt with ready frameworks (CO-STAR, RTF, RISEN, Chain-of-Thought, few-shot) from role, context, task, constraints and output fields, as text, JSON messages or XML tags, with a token estimate and an ambiguity check.
Workspace
Prompt
How the result is calculated
The prompt is assembled from the fields by framework: CO-STAR (Context, Objective, Style, Tone, Audience, Response), RTF (Role, Task, Format), RISEN (Role, Instructions, Steps, End goal, Narrowing), CoT (adds “think step by step”) and few-shot (includes examples). Empty sections are dropped. The output is text with headings, system + user JSON messages or XML tags. Tokens are an estimate = characters ÷ 4. The ambiguity check is rule-based: empty or short task, vague words (something, stuff…), missing format.
Limits:
The token count is approximate (4 characters/token for English; Arabic and code differ) and not an actual count for any model. There is no guarantee the model responds as you expect; a framework is organisation, not magic. Nothing is sent to any service, and templates are not saved locally in this version.
FAQ
Which framework should I choose?
CO-STAR for writing and marketing, RTF for simple tasks, RISEN for multi-step tasks.
When to use few-shot?
When you want a specific output pattern; give two or three examples.
