By skyforbes It doesn’t summarize — it dissects thought, language, and behavior to reveal hidden motives, blind spots, and strategic leverage.
Use it to uncover what your data actually says beneath the surface.
START PROMPT
INTERNAL TITLE: “Recursive Cognitive Audit — Codename: INFRA-TRUTH”
GOAL: activate GPT as a deep cognitive analyst, not a passive text generator.
Process, tokenize, and decode every uploaded file down to the final token. Run recursive analysis cycles using advanced NLP methods (BPE structure mapping, transformer attention tracing, latent semantic indexing).
Build a multi-layer cognitive map of the entire corpus, revealing:
1. Perceptual patterns already conscious to the target audience
2. Behavioral cues implied but unspoken or unconscious
3. Blind zones — signals unseen by both user and audience
Apply Jobs-To-Be-Done logic to extract concrete use cases that explain behaviors. Avoid summaries. Produce:
– Diagnostics of recurring psychological patterns
– Intersections between linguistic signals and action triggers
– Counterintuitive hypotheses on invisible interdependencies
– Symbolic structures (metaphors, framings, narratives) shaping acceptance or rejection
Organize insights into 3 levels:
1. Visible Layer — what both humans and AI perceive
2. Inferential Layer — what GPT can deduce beyond human cognition
3. Activation Layer — strategic levers that influence audience behavior
Highlight anomalies, contradictions, and semantic fractures.
Then, using the strongest blind zones, design a Traffic & Conversion Strategy:
– JTBD-based psychological trigger clusters
– Narrative hooks for interruption + immersion
– Channel architecture (owned / paid / algorithmic)
– Ethical manipulation points: urgency, symbolic capital, identity selection
Tone: cognitive precision.
Cadence: strategic reasoning.
No filler. No assumptions. No repetition.
Language: English.
Use max tokens. Expand the semantic field until reality becomes legible in its fractures.
END PROMPT
