count_units
ChatGPTtext under UPI's published counting definitions and returns the count plus the counted units themselves as segments (character units return no segments). An unknown unit returns a structured INVALID_RULE result.By Sean Adams
Most prompt tools rewrite your prompt. Universal Prompt Intelligence compiles it. UPI takes a rough idea — or an existing prompt that already contains requirements you care about — and produces a model-ready prompt that protects the task contract: your constraints, methods, formats, limits, and intended results stay intact while the wording, structure, and conventions are adapted to the target model. Model-aware, not one-size-fits-all. Compile for GPT, Claude, or a universal format. Each adapter applies the target model's own conventions for structure, instruction placement, and formatting — so the same intent arrives correctly packaged for the model that will actually run it. Constraint-preserving by design. Numerical limits, non-negotiable requirements, "do not" rules, output schemas, and verification steps are treated as contract terms, not stylistic suggestions. When a requirement can't be verified, UPI says so instead of claiming compliance it didn't measure. Specialist workflows included. Dedicated references for technical and professional work — code modification, data and dashboard specification, and complex research prompts — where generic rewriting is most likely to drop the detail that made the request work. Use UPI when the prompt is complicated, the requirements are load-bearing, or the output has to survive contact with a different model than the one you wrote it on.
text under UPI's published counting definitions and returns the count plus the counted units themselves as segments (character units return no segments). An unknown unit returns a structured INVALID_RULE result.text under UPI's published counting definitions and reports pass/fail, the actual count, the signed delta from the nearest bound, and per-segment counts when per_segment is requested. Malformed rules return a structured INVALID_RULE result and malformed JSON_ITEM text returns MALFORMED; the tool never silently passes a bad rule.Click any prompt to copy it.
Add this badge to your site or README. It links to this page.
HTML
<a href="https://mcp-app-store.keenethics-labs.com/app/universal-prompt-intelligence"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>
Markdown
[](https://mcp-app-store.keenethics-labs.com/app/universal-prompt-intelligence)