All templates
⚡ Advanced Techniques·Works on: claude, chatgpt
AI Personality Forensics
Technique: Stylometric Inference
Paste 3 of your texts/posts/emails. Get a personality profile so accurate it's uncomfortable.
Advanced Viral#personality#viral#forensics#shareable
0/1
Fill the template
1 placeholder left.
~360tokens
Live preview· click any pill to jump to its field
Act as a forensic personality analyst. I will paste 3 writing samples I produced under different conditions. Read for signal, not content. Sample 1 (recent message to a friend): Sample 2 (recent professional email or post): Sample 3 (an opinion or rant I wrote when I was annoyed): Protocol — do NOT analyse what I said, analyse HOW I say things. Look for: - Sentence length distribution and rhythm. - Use of qualifiers, hedges, intensifiers. - Specificity vs. abstraction. - Where I cluster nouns vs. verbs. - How I handle disagreement (avoidance, engagement, deflection). - What I leave UNSAID — what someone with my topic would normally mention but I skip. Output, in this exact structure: 1. **Three things I clearly care about** (inferred from emphasis, not content). 2. **Three things I quietly worry about** (inferred from hedges, qualifications, what gets re-stated). 3. **My communication style in one sentence.** 4. **What I optimise for** — speed / clarity / being liked / being right / control / connection — pick the top 2 with the evidence. 5. **The personality trait I broadcast vs. the one I actually have** — if there's a gap, name it. 6. **The job / role I'd thrive in based on this signal** (not based on what I said). 7. **The single sentence that gave you the most signal.** Be specific. Generic horoscope-style readings are a failure of this prompt — quote my actual words as evidence.
Inputs0 of 1
›See the lazy version this template replaces
Before — the lazy prompt
What is my personality?
Why it works
- Reading for HOW vs. what is the actual skill of personality inference — and what makes the output feel uncomfortably accurate.
- Three samples across contexts (friend/professional/annoyed) triangulate signal that a single sample can't.
- The 'broadcast vs. actual' gap question is the one no horoscope reveals.
- Forcing the model to quote evidence kills generic 'you are creative and ambitious' output.
Make this one yours
Replace the bracketed placeholders, then paste into the Prompt Fixer to lint your customisation before hitting send.