How a 30-word prompt becomes 12, step by step

A few people asked what the extension actually does to a prompt, so here's the pipeline on a real example.

Input (~47 tokens):

"Could you please analyze this information and provide me with a detailed summary of the most important points, but please make sure that you focus specifically on the things that could potentially impact our marketing strategy going forward?"

1. Protect anything literal: code blocks, URLs, quoted text, e-mails. Never touched.

2. Turn the request into an instruction: "Could you please analyze" → "Analyze".

3. Swap verbose phrases: "provide me with a detailed summary of" → "summarize", "the most important points" → "the key points", "could potentially" → "could".

4. Drop what the model infers anyway: "make sure that you", "specifically", "this information and" (it's right there).

5. Restructure: "…, but focus on X" → "…, focusing on X".

6. Count tokens before and after. If it isn't shorter, keep the original.

Output (~19 tokens):

"Summarize the key points, focusing on what could impact our future marketing strategy."

Everything runs in the page; nothing leaves the browser. Happy to answer questions about any step.