AI Demand-Reduction Prompt Pack

# Copy/paste — Digital Detox OS Router

---

I don't know which Digital Detox OS prompt fits.

Do not solve my situation.
Do not give me a digital detox plan.
Do not make me compare a long list.
Ask at most one question.

Use the few words I provide to choose one system only:

01 Digital Noise Audit
02 Notification Boundary
03 Feed Spiral Interrupt
04 Bedtime Input Boundary
05 Calm Defaults
06 Reachable But Not Invaded
07 Lower-Input Bridge
08 Screen Reentry Plan

If several fit, choose the lowest-demand useful option.
If I say "I don't know" or give too little information, choose 01 Digital Noise Audit.
If opening or changing digital settings itself seems like too much right now, tell me I can stop instead.

Respond only with:
- system number + name
- one short reason
- "Use it, choose none, or stop here."

Context — optional:
[A FEW WORDS OR SKIP]

Do not start the selected exercise.
Stop after routing.

# Digital Noise Audit

# Copy/paste — Digital Noise Audit

---

I am overwhelmed by digital noise, notifications, feeds, and screen inputs.

Do not shame me. Do not give me a 30-day detox. Do not tell me to delete everything. Do not create a productivity plan.

Sort what I paste into:
1. Actually needed
2. Useful but too loud
3. Feels urgent, but I already know it can wait
4. Comparison / activation source
5. Bedtime input
6. Can be muted/hidden
7. Needs human / official judgment — work, school, care, security, financial, health/safety, or other real-world obligations. Do not decide these can wait; flag them for me.
8. One optional calmer default — if useful

Short bullets only.

Capacity right now — optional:
[leave blank / I don't know / prefer not to label / NSIF state]

If I name a Capacity State, adapt response length, language, number of choices, task complexity, and interaction length to it.
If I leave this blank or say "I don't know," default to a short, low-load response — do not ask me to explain further.
Do not infer or declare my Capacity State.

Rules: Acknowledge briefly. Work with fragments — do not require more disclosure than I provide. Do only the reduction or sorting I asked for; do not expand scope. No 30-day detox, no monk mode, no moral lecture, no shame. Do not diagnose me or act as a therapist.

Essential Access First: do not tell me to silence or remove emergency alerts, accessibility tools, alarms, security/2FA messages, care-related reminders, designated essential contacts, or required work/school communication. Flag those for me to decide instead.

Do not decide that real-world obligations — work, school, care, security, financial, or health/safety — can wait. Route those to "Needs human / official judgment."

If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support.

One cycle only — when done, tell me I can stop here; do not offer another task or ask "want me to...".

Privacy: Do not paste passwords, 2FA codes, private messages you do not have permission to share, account numbers, confidential work/school information, medical/private care information, or other people's sensitive information. Your AI provider's own privacy/storage settings apply.

Context — optional:
[A FEW WORDS OR SKIP]

**Stop whenever enough.**

# Notification Boundary

# Copy/paste — Notification Boundary

---

Help me create a notification boundary.

Do not shame me. Do not make me unreachable. Do not give me a giant settings plan.

Sort what I paste into:
1. User-designated essential / immediate — I decide what counts, not you
2. Scheduled check
3. Badge only
4. Manual check only
5. Mute
6. Remove
7. Needs human / official judgment

Rules: do not decide what counts as an emergency. Do not mute or remove emergency alerts, accessibility tools, alarms, security/2FA messages, care-related reminders, designated essential contacts, or required work/school communication — flag those for me to decide. Suggest at most one optional setting change. Stop may be the result.

Capacity right now — optional:
[leave blank / I don't know / prefer not to label / NSIF state]

If I name a Capacity State, adapt response length, language, number of choices, task complexity, and interaction length to it.
If I leave this blank or say "I don't know," default to a short, low-load response — do not ask me to explain further.
Do not infer or declare my Capacity State.

Rules: Acknowledge briefly. Work with fragments — do not require more disclosure than I provide. Do only the reduction or sorting I asked for; do not expand scope. No 30-day detox, no monk mode, no moral lecture, no shame. Do not diagnose me or act as a therapist.

Essential Access First: do not tell me to silence or remove emergency alerts, accessibility tools, alarms, security/2FA messages, care-related reminders, designated essential contacts, or required work/school communication. Flag those for me to decide instead.

Do not decide that real-world obligations — work, school, care, security, financial, or health/safety — can wait. Route those to "Needs human / official judgment."

If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support.

One cycle only — when done, tell me I can stop here; do not offer another task or ask "want me to...".

Privacy: Do not paste passwords, 2FA codes, private messages you do not have permission to share, account numbers, confidential work/school information, medical/private care information, or other people's sensitive information. Your AI provider's own privacy/storage settings apply.

Context — optional:
[A FEW WORDS OR SKIP]

**Stop whenever enough.**

# Feed Spiral Interrupt

# Copy/paste — Feed Spiral Interrupt

---

I keep getting pulled into scrolling or feed spirals.

Do not shame me. Do not tell me to just stop. Do not give a discipline plan.

Help me identify:
1. What I may be needing before I scroll
2. What the feed gives me
3. What the feed costs me
4. One optional lower-demand alternative matching the need I named
5. One optional reentry boundary — or stop

End with: "This is demand reduction. No character verdict needed."

Capacity right now — optional:
[leave blank / I don't know / prefer not to label / NSIF state]

If I name a Capacity State, adapt response length, language, number of choices, task complexity, and interaction length to it.
If I leave this blank or say "I don't know," default to a short, low-load response — do not ask me to explain further.
Do not infer or declare my Capacity State.

Rules: Acknowledge briefly. Work with fragments — do not require more disclosure than I provide. Do only the reduction or sorting I asked for; do not expand scope. No 30-day detox, no monk mode, no moral lecture, no shame. Do not diagnose me or act as a therapist.

Essential Access First: do not tell me to silence or remove emergency alerts, accessibility tools, alarms, security/2FA messages, care-related reminders, designated essential contacts, or required work/school communication. Flag those for me to decide instead.

Do not decide that real-world obligations — work, school, care, security, financial, or health/safety — can wait. Route those to "Needs human / official judgment."

If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support.

One cycle only — when done, tell me I can stop here; do not offer another task or ask "want me to...".

Privacy: Do not paste passwords, 2FA codes, private messages you do not have permission to share, account numbers, confidential work/school information, medical/private care information, or other people's sensitive information. Your AI provider's own privacy/storage settings apply.

Context — optional:
[A FEW WORDS OR SKIP]

**Stop whenever enough.**

# Bedtime Input Boundary

# Copy/paste — Bedtime Input Boundary

---

Bedtime gets eaten by my phone and screens.

Do not shame me. Do not give a rigid detox. One softer boundary only — or none.

Essential Access First — keep essential overnight access intact: alarms, accessibility tools, care/monitoring, designated essential contacts, safety/security. Keep the phone nearby if I need it for any of those. Phone outside the bedroom is one option, not the goal.

Help me name:
1. What reaches me before sleep
2. What it costs
3. What is allowed tonight
4. What is not allowed tonight
5. Phone location or display default — optional
6. One low-demand alternative — optional

Capacity right now — optional:
[leave blank / I don't know / prefer not to label / NSIF state]

If I name a Capacity State, adapt response length, language, number of choices, task complexity, and interaction length to it.
If I leave this blank or say "I don't know," default to a short, low-load response — do not ask me to explain further.
Do not infer or declare my Capacity State.

Rules: Acknowledge briefly. Work with fragments — do not require more disclosure than I provide. Do only the reduction or sorting I asked for; do not expand scope. No 30-day detox, no monk mode, no moral lecture, no shame. Do not diagnose me or act as a therapist.

Essential Access First: do not tell me to silence or remove emergency alerts, accessibility tools, alarms, security/2FA messages, care-related reminders, designated essential contacts, or required work/school communication. Flag those for me to decide instead.

Do not decide that real-world obligations — work, school, care, security, financial, or health/safety — can wait. Route those to "Needs human / official judgment."

If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support.

One cycle only — when done, tell me I can stop here; do not offer another task or ask "want me to...".

Privacy: Do not paste passwords, 2FA codes, private messages you do not have permission to share, account numbers, confidential work/school information, medical/private care information, or other people's sensitive information. Your AI provider's own privacy/storage settings apply.

Context — optional:
[A FEW WORDS OR SKIP]

**Stop whenever enough.**

# Calm Defaults

# Copy/paste — Calm Defaults

---

I need one calmer digital default — not a whole phone overhaul. Or no change today.

Do not give me a checklist of every app. Do not shame me.

From what I paste, suggest one optional default if useful (badge, mute, placement, focus mode, grayscale, charger location, feed boundary) — or say "no change today" if nothing clearly helps. Essential Access First — do not touch emergency alerts, accessibility tools, alarms, security/2FA, care reminders, essential contacts, or required work/school communication.

Say what I am not changing today. End with permission to stop.

Capacity right now — optional:
[leave blank / I don't know / prefer not to label / NSIF state]

If I name a Capacity State, adapt response length, language, number of choices, task complexity, and interaction length to it.
If I leave this blank or say "I don't know," default to a short, low-load response — do not ask me to explain further.
Do not infer or declare my Capacity State.

Rules: Acknowledge briefly. Work with fragments — do not require more disclosure than I provide. Do only the reduction or sorting I asked for; do not expand scope. No 30-day detox, no monk mode, no moral lecture, no shame. Do not diagnose me or act as a therapist.

Essential Access First: do not tell me to silence or remove emergency alerts, accessibility tools, alarms, security/2FA messages, care-related reminders, designated essential contacts, or required work/school communication. Flag those for me to decide instead.

Do not decide that real-world obligations — work, school, care, security, financial, or health/safety — can wait. Route those to "Needs human / official judgment."

If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support.

One cycle only — when done, tell me I can stop here; do not offer another task or ask "want me to...".

Privacy: Do not paste passwords, 2FA codes, private messages you do not have permission to share, account numbers, confidential work/school information, medical/private care information, or other people's sensitive information. Your AI provider's own privacy/storage settings apply.

Context — optional:
[A FEW WORDS OR SKIP]

**Stop whenever enough.**

# Reachable But Not Invaded

# Copy/paste — Reachable But Not Invaded

---

I want to be reachable for what matters but not invaded by everything else.

Do not tell me to go offline. Do not shame me.

Organize the access I tell you is essential — do not decide for me who or what needs real access.

Help me sort:
1. Who/what I already know needs reliable access
2. What does not deserve lock-screen access
3. What can move to manual check
4. Needs human / official judgment — work, school, care, security, health/safety, or other real-world obligations you cannot verify
5. One sentence: people who matter reach me through ___ ; everything else can be checked later — if that fits my real obligations

Do not decide that work, school, care, or safety communication can wait — flag those for me.

Capacity right now — optional:
[leave blank / I don't know / prefer not to label / NSIF state]

If I name a Capacity State, adapt response length, language, number of choices, task complexity, and interaction length to it.
If I leave this blank or say "I don't know," default to a short, low-load response — do not ask me to explain further.
Do not infer or declare my Capacity State.

Rules: Acknowledge briefly. Work with fragments — do not require more disclosure than I provide. Do only the reduction or sorting I asked for; do not expand scope. No 30-day detox, no monk mode, no moral lecture, no shame. Do not diagnose me or act as a therapist.

Essential Access First: do not tell me to silence or remove emergency alerts, accessibility tools, alarms, security/2FA messages, care-related reminders, designated essential contacts, or required work/school communication. Flag those for me to decide instead.

Do not decide that real-world obligations — work, school, care, security, financial, or health/safety — can wait. Route those to "Needs human / official judgment."

If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support.

One cycle only — when done, tell me I can stop here; do not offer another task or ask "want me to...".

Privacy: Do not paste passwords, 2FA codes, private messages you do not have permission to share, account numbers, confidential work/school information, medical/private care information, or other people's sensitive information. Your AI provider's own privacy/storage settings apply.

Context — optional:
[A FEW WORDS OR SKIP]

**Stop whenever enough.**

# Lower-Input Bridge

# Copy/paste — Lower-Input Bridge

---

Quiet or lower input feels uncomfortable after a lot of input.

Input can get quieter gradually — I do not have to move from high input to silence, and silence is not the goal.

Suggest one optional bridge only: lower volume, audio without video, a dim screen, a still image, shorter input not none, one lower-input option, phone face down, rest, or do nothing. All optional — my accessibility and safety needs override any suggestion.

Capacity right now — optional:
[leave blank / I don't know / prefer not to label / NSIF state]

If I name a Capacity State, adapt response length, language, number of choices, task complexity, and interaction length to it.
If I leave this blank or say "I don't know," default to a short, low-load response — do not ask me to explain further.
Do not infer or declare my Capacity State.

Rules: Acknowledge briefly. Work with fragments — do not require more disclosure than I provide. Do only the reduction or sorting I asked for; do not expand scope. No 30-day detox, no monk mode, no moral lecture, no shame. Do not diagnose me or act as a therapist.

Essential Access First: do not tell me to silence or remove emergency alerts, accessibility tools, alarms, security/2FA messages, care-related reminders, designated essential contacts, or required work/school communication. Flag those for me to decide instead.

Do not decide that real-world obligations — work, school, care, security, financial, or health/safety — can wait. Route those to "Needs human / official judgment."

If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support.

One cycle only — when done, tell me I can stop here; do not offer another task or ask "want me to...".

Privacy: Do not paste passwords, 2FA codes, private messages you do not have permission to share, account numbers, confidential work/school information, medical/private care information, or other people's sensitive information. Your AI provider's own privacy/storage settings apply.

Context — optional:
[A FEW WORDS OR SKIP]

**Stop whenever enough.**

# Screen Reentry Plan

# Copy/paste — Screen Reentry Plan

---

When I return to screens I fall through every app.

Do not shame me. Help me write:
1. What I am here for
2. What app/site
3. Do I want a stop cue? time · purpose · event · body/environment cue · or none
4. What I am not opening
5. Where I go after

Short. One session only.

Capacity right now — optional:
[leave blank / I don't know / prefer not to label / NSIF state]

If I name a Capacity State, adapt response length, language, number of choices, task complexity, and interaction length to it.
If I leave this blank or say "I don't know," default to a short, low-load response — do not ask me to explain further.
Do not infer or declare my Capacity State.

Rules: Acknowledge briefly. Work with fragments — do not require more disclosure than I provide. Do only the reduction or sorting I asked for; do not expand scope. No 30-day detox, no monk mode, no moral lecture, no shame. Do not diagnose me or act as a therapist.

Essential Access First: do not tell me to silence or remove emergency alerts, accessibility tools, alarms, security/2FA messages, care-related reminders, designated essential contacts, or required work/school communication. Flag those for me to decide instead.

Do not decide that real-world obligations — work, school, care, security, financial, or health/safety — can wait. Route those to "Needs human / official judgment."

If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support.

One cycle only — when done, tell me I can stop here; do not offer another task or ask "want me to...".

Privacy: Do not paste passwords, 2FA codes, private messages you do not have permission to share, account numbers, confidential work/school information, medical/private care information, or other people's sensitive information. Your AI provider's own privacy/storage settings apply.

Context — optional:
[A FEW WORDS OR SKIP]

**Stop whenever enough.**