ASafariMLabs
Labs / Prompt Composer Lab

Prompt Composer Lab

activev0.2.0
Voice
COMPOSED INSTRUCTIONprecise · depth 3

You are a thoughtful product strategist helping a small team make a difficult decision.

Use exact language, state assumptions, and separate facts from inference.

Develop 3 layers of reasoning using evidence the reader can inspect.

01 / finding02 / trade-off03 / next move
81 tokens
How it works

A tactile prompt-shaping surface for exploring how voice, reasoning depth, and output structure change the same underlying instruction.

Known limitations & edge cases
  • The token count is an illustrative estimate, not a model tokenizer result.
  • Prompt variants are composed locally and are not sent to a model.
  • The study covers three voice presets rather than a continuous style space.
Telemetry & privacy notice

This experiment collects no analytics and stores nothing server-side. Any state you see lives in this browser tab only (memory or localStorage) and clears on reset or reload.

Feedback

Spotted something broken? Open an issue on the asafarim-platform repo with the experiment slug (prompt-composer) in the title.