generate-realistic-person

What this skill does

# Generate Realistic Person

You generate a single fictional present-day person with rich, internally consistent life detail **and** an explicit behavioral layer that makes the persona suitable for simulation, synthetic populations, agent-based modeling, and event-response analysis. Output is a complete Markdown profile in English.

## Hard rules

- **Always fictional.** Never produce a profile that matches a recognizable real person.
- **One person per call.** Refuse requests for families, teams, groups, or non-person entities.
- **Present day only.** Refuse other historical periods or future settings.
- **English output.** Always.
- **Internal consistency required.** Every dimension must fit every other dimension.
- **Behavioral layer always present.** All 18 behavioral sections must be populated and traceable to the narrative profile.

## Input handling

Input is fully optional. Parse anything the user provided into the narrative dimensions below. Fill everything else realistically. Behavioral-layer fields are always derived by you — they are not accepted as direct input.

### Narrative dimensions

- **Demographics**: country/region, urbanization, area type, age, gender, ethnicity
- **Work**: profession, job title, seniority, industry, real company (optional), work setup, education
- **Income & status**: income bracket, socioeconomic status
- **Family**: relationship status, partner, children, living situation, pets
- **Social life**: close friends, friend circles, contact pattern
- **Health & lifestyle**: general health, physical activity, diet, sleep, diagnoses/neurodivergence, disabilities/chronic conditions
- **Personality & values**: traits, core values, religion & beliefs, motivations & drives, what brings them down, political leaning (optional)
- **Psychographics**: lifestyle segment, tribes/subcultures
- **Needs**: jobs-to-be-done, core needs
- **Tech & media**: digital literacy, media consumption
- **Attitudes**: sustainability, health, privacy, technology, etc.
- **Interests**: hobbies, passions
- **Constraints / scenarios**: life events, stressors or themes

### Pre-checks before generating

1. **Real-person match**: if provided attributes uniquely identify a real, recognizable person — adapt to a clearly fictional profile and add a Notes entry.
2. **Internal contradiction**: if input contains an impossible combination — adapt to the most logical resolution and add a Notes entry.
3. **Out-of-scope**: refuse briefly if the request is for multiple persons, a non-person entity, a different time period, or an explicitly real person.

## Generation rules

### Phase A — Narrative profile

- **Fill every narrative dimension.** No section empty unless it genuinely doesn't apply.
- **Honor user input verbatim** unless it triggers a pre-check adjustment.
- **Internal consistency**: income ↔ profession ↔ region ↔ seniority; routine ↔ work setup ↔ commute ↔ family load; hobbies ↔ budget ↔ time ↔ life stage; social pattern ↔ personality ↔ family load ↔ work; diet/activity/sleep plausible for age and situation; religion + values + motivations + attitudes form a coherent worldview.
- **Real company**: safe assumptions first, fall back to publicly known facts only when needed. Label assumptions in Notes.
- **Background story**: 2-4 paragraphs covering childhood, formative experiences, education, career path, key relationships, and path to current situation.
- **Dark or difficult sides** may be included when they fit and stay realistic — never as caricature.
- **Diagnoses / neurodivergence** must visibly shape routine, social pattern, stress response, sensory preferences, and information processing.

### Phase B — Behavioral layer (always generate all 18 sections)

Every behavioral field must **trace back** to something in the narrative profile. If it cannot be traced, either derive it from the narrative profile first or mark it `inferred`/`speculative` in Certainty markers.

Required behavioral sections:

1. **Decision style** — speed (fast ↔ deliberate), mode (analytical ↔ intuitive), autonomy (independent ↔ confirmation-seeking), risk posture (risk-averse ↔ experimental), orientation (principle-driven ↔ pragmatic)
2. **Behavioral thresholds** — concrete conditions for: disengaging, resisting, accepting higher cost, changing mind, breaking routine
3. **Scenario response patterns** — specific likely response for each of: financial setback, health issue, work-pressure/deadline escalation, government policy change, privacy incident, social unrest, climate event, interpersonal conflict (family/neighborhood/work)
4. **Stress response** — primary reaction, effect on decision-making, effect on social interaction, visible signals, recovery behavior
5. **Trust model** — government, experts, employer, media, partner/friends/colleagues/neighborhood, formal vs informal sources
6. **Social influence** — most influential people, sensitivity to group norms, sensitivity to peer judgement, tendency to adopt network behavior, preference for expert vs peer validation
7. **Value-behavior tensions** — ideals vs convenience, principles vs time pressure, privacy vs practicality, sustainability vs family logistics, social desire vs social energy. **At least one tension must surface a realistic inconsistency.**
8. **Action readiness** — level (low/medium/high), pattern (orient first ↔ act directly), mode (solo ↔ only in group), trigger (urgency/conviction/social pressure)
9. **Adaptability** — openness to information, openness to behavior change, conditions for change, speed of adaptation, regression risk
10. **Routine dependency** — planning need, disruption tolerance, impact of unexpected changes, fixed rhythms vs flexibility
11. **Priority logic** — priorities normal, priorities in crisis, first to be sacrificed, protected at all costs
12. **Certainty markers** — label key attributes with `confirmed` (user-given), `likely` (strongly implied), `inferred` (reasonable derivation), `speculative` (weakly supported)
13. **Stable vs. dynamic traits** — stable (core traits, values, diagnoses) / semi-stable (routines, moods) / dynamic (opinions on specific issues, context-bound reactions)
14. **Triggers** — time loss, financial harm, administrative complexity, unfairness, safety risk, loss of autonomy, schedule disruption, risk to children/family/reputation
15. **Inhibitors** — fatigue, lack of time, conflict avoidance, uncertainty, lack of trust, social awkwardness, practical/logistical barriers
16. **Information processing style** — deep reading vs scanning, source-checking habits, nuance vs headlines, sensitivity to emotional framing, multiple sources vs one trusted source
17. **Likely first response** — pick exactly one (do nothing / consult partner / check official information / search online / message friend or colleague / act immediately / postpone) and ground it in decision style + trust model + action readiness
18. **Recovery pattern** — recovery speed, need for rest/control/social contact, tendency toward structural adjustment, duration of aftermath

Behavioral-layer generation rules:

- **Specificity**: state concrete conditions, not "sometimes" or "varies"
- **Derivability**: behavior follows from the narrative profile
- **Coherence**: decision style ↔ stress response ↔ action readiness ↔ trust model ↔ information processing form one pattern
- **Distinguishability**: triggers, inhibitors, thresholds must differentiate this person from a generic persona
- **Realistic inconsistency**: value-behavior tensions must include at least one real mismatch

### Phase C — Daily life samples

- **Weekday** hour-by-hour — anchored to work setup, commute, family load, routines, and consistent with behavioral layer (e.g., high routine dependency shows clear rhythm)
- **Weekend** hour-by-hour — hobbies, social contact, family activities, rest patterns, consistent with adaptability and social-contact patterns
- Be specific: reference named activities, places, people in the person's li