ThinkWithPrompt by SynotticThinkWithPrompt

Extract and track the predictions in an article or talk

Pulls every forecast from content into a table with timeframe, confidence, and how to check later whether it came true.

Research & SummariesRTFStandardWritingRewrite & EditResearchFill-in Template

The prompt

Extract every prediction from the content below so I can track whether it comes true. Content: [PASTE ARTICLE, TRANSCRIPT, REPORT, OR THREAD] Who made the predictions: [AUTHOR / SPEAKER and when it was published] What I'll use this for: [e.g. "accountability tracker", "investment research notes", "fact-check a pundit"] Rules: - Include only forward-looking claims about what will happen. Skip opinions, recommendations, and descriptions of the present. - Quote or closely paraphrase; do not strengthen or soften the claim. If a prediction is vague ("AI will change everything"), still include it and mark it Not testable. - Use the timeframe and confidence only if the source states or clearly implies them. Otherwise write "Not stated." Never invent dates or probabilities. Return: 1. A table: # | Prediction (max 15 words) | Speaker | Timeframe | Stated confidence | Testable? (Yes / Partly / No) | How to verify (the specific data, event, or source that would settle it) | Check-back date. 2. For each "Partly" or "No," a rewritten version that would be testable, clearly labeled as my rewording, not the source's. 3. Hedges and escape hatches: phrases that would let the speaker claim success whatever happens. 4. A short note on any predictions that contradict each other.

Highlighted [brackets] are blanks to replace with your own details.

Fill in the blanks and use this promptMore Research & Summaries prompts

What to fill in

  • PASTE ARTICLE, TRANSCRIPT, REPORT, OR THREAD
  • AUTHOR / SPEAKER and when it was published

How this prompt is built

This prompt follows the RTF structure (Role, Task, Format). RTF tells the model who to be, what to do and what shape the answer should take. It is the quickest structure that still gives noticeably better results than a bare request.

Related prompts

Find more prompts like this