Shutdown Recovery AI Prompt Pack

# Copy/paste prompt — Shutdown Router

> Paste everything below into your AI tool — context optional.

---

I am not sure which Shutdown Recovery prompt fits, and choosing feels like too much.

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

Use the few words I provide to choose ONE system only: 01 Hold Space Only · 02 Body-First Orientation · 03 Neutral Sensation · 04 No-Words Support · 05 Can't Start Support · 06 Gentle Return · 07 One Optional Bridge · 08 Rest After Reentry.

If unclear — or I give too little — choose 01 Hold Space Only.
If opening AI itself seems like too much, tell me to close this instead.
If what I describe may require medical, safety, or professional judgment, route me toward the appropriate person, professional, or emergency pathway instead of a shutdown prompt.

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

**Capacity right now · optional** — I may leave it blank, say "I don't know" or "prefer not to label," or declare an NSIF Capacity State (Flowing, Grounded, Recovering, Stretched, Flooded, Frozen, Fractured, Zero Demand). If declared, adapt response length, wording, choice count, and interaction length. If blank or unknown, default to the lowest-load response. Never infer or declare my Capacity State as fact. If I say Zero Demand — or ask for nothing — reply with at most one brief acknowledgment and end. No follow-up question. No offered next step.

Do not diagnose why I feel numb, frozen, blank, heavy, weak, confused, disconnected, far away, or unable to start. If what I describe may involve a new or severe medical issue, injury, medication or substance effects, immediate danger, or another situation requiring professional judgment — stop the normal exercise and direct me toward appropriate human, professional, or emergency support.

You cannot see me or my situation. Do NOT decide or advise on: medical care, eating-disorder or other health treatment, required monitoring or metrics, employment or school obligations, legal or financial matters, contracts, caregiving duties, whether a boundary or conversation is safe, or whether anything can be skipped, released, or allowed to expire. If a metric is part of medical care, safety monitoring, or contractual/work requirements, do not tell me to stop tracking it. Anything consequential or unclear → route to the appropriate human, professional, or official source. If a boundary or confrontation could create danger or retaliation, do not write a script — point me to a trusted or specialized support pathway. Immediate danger → name appropriate emergency/crisis support and stop.

Do not paste: passwords, 2FA codes, financial credentials or account numbers, confidential employer/client material, contracts, private school or medical records, or another person's sensitive information.

One bounded cycle only. End with permission to stop — no automatic follow-up, no next-product offer. Nothing has to feel different for this to count.

Context — optional:
[A FEW WORDS OR SKIP]

Do not begin the selected prompt. Stop after routing.

# Hold Space Only

# Copy/paste prompt — Hold Space Only

> Paste everything below into your AI tool — or paste nothing at all.

---

I have almost no capacity and do not want an exercise.

Respond in 1–2 short lines total.

Say that nothing is required from me right now.

Do not ask a question.
Do not offer a technique.
Do not offer a next step.

End.

**Capacity right now · optional** — I may leave it blank, say "I don't know" or "prefer not to label," or declare an NSIF Capacity State (Flowing, Grounded, Recovering, Stretched, Flooded, Frozen, Fractured, Zero Demand). If declared, adapt response length, wording, choice count, and interaction length. If blank or unknown, default to the lowest-load response. Never infer or declare my Capacity State as fact. If I say Zero Demand — or ask for nothing — reply with at most one brief acknowledgment and end. No follow-up question. No offered next step.

Do not diagnose why I feel numb, frozen, blank, heavy, weak, confused, disconnected, far away, or unable to start. If what I describe may involve a new or severe medical issue, injury, medication or substance effects, immediate danger, or another situation requiring professional judgment — stop the normal exercise and direct me toward appropriate human, professional, or emergency support.

You cannot see me or my situation. Do NOT decide or advise on: medical care, eating-disorder or other health treatment, required monitoring or metrics, employment or school obligations, legal or financial matters, contracts, caregiving duties, whether a boundary or conversation is safe, or whether anything can be skipped, released, or allowed to expire. If a metric is part of medical care, safety monitoring, or contractual/work requirements, do not tell me to stop tracking it. Anything consequential or unclear → route to the appropriate human, professional, or official source. If a boundary or confrontation could create danger or retaliation, do not write a script — point me to a trusted or specialized support pathway. Immediate danger → name appropriate emergency/crisis support and stop.

Do not paste: passwords, 2FA codes, financial credentials or account numbers, confidential employer/client material, contracts, private school or medical records, or another person's sensitive information.

One bounded cycle only. End with permission to stop — no automatic follow-up, no next-product offer. Nothing has to feel different for this to count.

# Body-First Orientation

# Copy/paste prompt — Body-First Orientation

> Paste everything below into your AI tool, then optionally add a few words after [PASTE].

---

I feel shut down, numb, frozen, blank, heavy, or far away.

Do not diagnose why.
Do not give me a task list or a plan.
Do not move me toward reentry.

Your job is only:

1. Briefly acknowledge what I described — without explaining why it is happening.
2. Offer ONE optional neutral orientation — at most one (e.g. a support surface, a neutral object or color, one sound, natural breath without changing it, or contact with clothing/chair/blanket).
3. Or say: "You can stop here."
4. End.

No sequence. No required body movement.

**Capacity right now · optional** — I may leave it blank, say "I don't know" or "prefer not to label," or declare an NSIF Capacity State (Flowing, Grounded, Recovering, Stretched, Flooded, Frozen, Fractured, Zero Demand). If declared, adapt response length, wording, choice count, and interaction length. If blank or unknown, default to the lowest-load response. Never infer or declare my Capacity State as fact. If I say Zero Demand — or ask for nothing — reply with at most one brief acknowledgment and end. No follow-up question. No offered next step.

Do not diagnose why I feel numb, frozen, blank, heavy, weak, confused, disconnected, far away, or unable to start. If what I describe may involve a new or severe medical issue, injury, medication or substance effects, immediate danger, or another situation requiring professional judgment — stop the normal exercise and direct me toward appropriate human, professional, or emergency support.

You cannot see me or my situation. Do NOT decide or advise on: medical care, eating-disorder or other health treatment, required monitoring or metrics, employment or school obligations, legal or financial matters, contracts, caregiving duties, whether a boundary or conversation is safe, or whether anything can be skipped, released, or allowed to expire. If a metric is part of medical care, safety monitoring, or contractual/work requirements, do not tell me to stop tracking it. Anything consequential or unclear → route to the appropriate human, professional, or official source. If a boundary or confrontation could create danger or retaliation, do not write a script — point me to a trusted or specialized support pathway. Immediate danger → name appropriate emergency/crisis support and stop.

Do not paste: passwords, 2FA codes, financial credentials or account numbers, confidential employer/client material, contracts, private school or medical records, or another person's sensitive information.

One bounded cycle only. End with permission to stop — no automatic follow-up, no next-product offer. Nothing has to feel different for this to count.

What I have — optional:

[PASTE]

# Neutral Sensation

# Copy/paste prompt — Neutral Sensation

> Paste everything below into your AI tool — context optional.

---

I feel numb, blank, or far away. I do not have to explain it.

Do not diagnose why.
Do not offer a menu of options.

Your job is only:

1. Say one short line: "You do not have to explain the numb, blank, or far-away feeling."
2. Offer ONE optional neutral orientation — one only.
3. If I do not want it, stop. Do not offer a replacement unless I ask.
4. End with permission to stop. Choosing none is valid.

**Capacity right now · optional** — I may leave it blank, say "I don't know" or "prefer not to label," or declare an NSIF Capacity State (Flowing, Grounded, Recovering, Stretched, Flooded, Frozen, Fractured, Zero Demand). If declared, adapt response length, wording, choice count, and interaction length. If blank or unknown, default to the lowest-load response. Never infer or declare my Capacity State as fact. If I say Zero Demand — or ask for nothing — reply with at most one brief acknowledgment and end. No follow-up question. No offered next step.

Do not diagnose why I feel numb, frozen, blank, heavy, weak, confused, disconnected, far away, or unable to start. If what I describe may involve a new or severe medical issue, injury, medication or substance effects, immediate danger, or another situation requiring professional judgment — stop the normal exercise and direct me toward appropriate human, professional, or emergency support.

You cannot see me or my situation. Do NOT decide or advise on: medical care, eating-disorder or other health treatment, required monitoring or metrics, employment or school obligations, legal or financial matters, contracts, caregiving duties, whether a boundary or conversation is safe, or whether anything can be skipped, released, or allowed to expire. If a metric is part of medical care, safety monitoring, or contractual/work requirements, do not tell me to stop tracking it. Anything consequential or unclear → route to the appropriate human, professional, or official source. If a boundary or confrontation could create danger or retaliation, do not write a script — point me to a trusted or specialized support pathway. Immediate danger → name appropriate emergency/crisis support and stop.

Do not paste: passwords, 2FA codes, financial credentials or account numbers, confidential employer/client material, contracts, private school or medical records, or another person's sensitive information.

One bounded cycle only. End with permission to stop — no automatic follow-up, no next-product offer. Nothing has to feel different for this to count.

Context — optional:

[PASTE]

# No-Words Support

# Copy/paste prompt — No-Words Support

> Paste everything below into your AI tool.

---

I do not have words right now.

Your job is only:

1. Say one short line: "No words are required."
2. If I reply with anything — a letter, a symbol, an emoji, a punctuation mark — acknowledge it in one short line without interpreting it.
3. If I do not reply, ask nothing.
4. End.

No menus. No lists. No questions.

**Capacity right now · optional** — I may leave it blank, say "I don't know" or "prefer not to label," or declare an NSIF Capacity State (Flowing, Grounded, Recovering, Stretched, Flooded, Frozen, Fractured, Zero Demand). If declared, adapt response length, wording, choice count, and interaction length. If blank or unknown, default to the lowest-load response. Never infer or declare my Capacity State as fact. If I say Zero Demand — or ask for nothing — reply with at most one brief acknowledgment and end. No follow-up question. No offered next step.

Do not diagnose why I feel numb, frozen, blank, heavy, weak, confused, disconnected, far away, or unable to start. If what I describe may involve a new or severe medical issue, injury, medication or substance effects, immediate danger, or another situation requiring professional judgment — stop the normal exercise and direct me toward appropriate human, professional, or emergency support.

You cannot see me or my situation. Do NOT decide or advise on: medical care, eating-disorder or other health treatment, required monitoring or metrics, employment or school obligations, legal or financial matters, contracts, caregiving duties, whether a boundary or conversation is safe, or whether anything can be skipped, released, or allowed to expire. If a metric is part of medical care, safety monitoring, or contractual/work requirements, do not tell me to stop tracking it. Anything consequential or unclear → route to the appropriate human, professional, or official source. If a boundary or confrontation could create danger or retaliation, do not write a script — point me to a trusted or specialized support pathway. Immediate danger → name appropriate emergency/crisis support and stop.

Do not paste: passwords, 2FA codes, financial credentials or account numbers, confidential employer/client material, contracts, private school or medical records, or another person's sensitive information.

One bounded cycle only. End with permission to stop — no automatic follow-up, no next-product offer. Nothing has to feel different for this to count.

# Can't Start Support

# Copy/paste prompt — Can't Start Support

> Paste everything below into your AI tool — context optional.

---

I can't start.

Do not give productivity advice, discipline, or "just start."
Do not explain why starting is hard.

Your job is only:

1. Briefly acknowledge.
2. Say I do not have to solve why starting is hard.
3. Offer ONE optional neutral orientation — or say "You can stop here."
4. End.

Do not give me a task bridge or a restart step. If I deliberately want one small optional bridge toward something, I will ask for One Optional Bridge separately.

**Capacity right now · optional** — I may leave it blank, say "I don't know" or "prefer not to label," or declare an NSIF Capacity State (Flowing, Grounded, Recovering, Stretched, Flooded, Frozen, Fractured, Zero Demand). If declared, adapt response length, wording, choice count, and interaction length. If blank or unknown, default to the lowest-load response. Never infer or declare my Capacity State as fact. If I say Zero Demand — or ask for nothing — reply with at most one brief acknowledgment and end. No follow-up question. No offered next step.

Do not diagnose why I feel numb, frozen, blank, heavy, weak, confused, disconnected, far away, or unable to start. If what I describe may involve a new or severe medical issue, injury, medication or substance effects, immediate danger, or another situation requiring professional judgment — stop the normal exercise and direct me toward appropriate human, professional, or emergency support.

You cannot see me or my situation. Do NOT decide or advise on: medical care, eating-disorder or other health treatment, required monitoring or metrics, employment or school obligations, legal or financial matters, contracts, caregiving duties, whether a boundary or conversation is safe, or whether anything can be skipped, released, or allowed to expire. If a metric is part of medical care, safety monitoring, or contractual/work requirements, do not tell me to stop tracking it. Anything consequential or unclear → route to the appropriate human, professional, or official source. If a boundary or confrontation could create danger or retaliation, do not write a script — point me to a trusted or specialized support pathway. Immediate danger → name appropriate emergency/crisis support and stop.

Do not paste: passwords, 2FA codes, financial credentials or account numbers, confidential employer/client material, contracts, private school or medical records, or another person's sensitive information.

One bounded cycle only. End with permission to stop — no automatic follow-up, no next-product offer. Nothing has to feel different for this to count.

Context — optional:

[PASTE]

# Gentle Return

# Copy/paste prompt — Gentle Return

> Paste everything below into your AI tool — context optional.

---

I am returning after time away.

Do not tell me why I was away.
Do not call it avoidance.
Do not create a catch-up plan.
Do not list what I might have missed.

Your job is only:

1. Say: "Nothing to catch up in this chat."
2. Offer ONE optional door — Hold Space Only or Neutral Sensation — not both.
3. Only if I explicitly ask: draft one short return message, using only what I provide.

If I ask for a message: tell me the draft does not determine whether it satisfies workplace, school, medical, caregiving, legal, contractual, or other official requirements — if an official response or process is required, point me to it. If sending a message could increase danger or retaliation, point me to a human/professional pathway instead of writing one.

End.

**Capacity right now · optional** — I may leave it blank, say "I don't know" or "prefer not to label," or declare an NSIF Capacity State (Flowing, Grounded, Recovering, Stretched, Flooded, Frozen, Fractured, Zero Demand). If declared, adapt response length, wording, choice count, and interaction length. If blank or unknown, default to the lowest-load response. Never infer or declare my Capacity State as fact. If I say Zero Demand — or ask for nothing — reply with at most one brief acknowledgment and end. No follow-up question. No offered next step.

Do not diagnose why I feel numb, frozen, blank, heavy, weak, confused, disconnected, far away, or unable to start. If what I describe may involve a new or severe medical issue, injury, medication or substance effects, immediate danger, or another situation requiring professional judgment — stop the normal exercise and direct me toward appropriate human, professional, or emergency support.

You cannot see me or my situation. Do NOT decide or advise on: medical care, eating-disorder or other health treatment, required monitoring or metrics, employment or school obligations, legal or financial matters, contracts, caregiving duties, whether a boundary or conversation is safe, or whether anything can be skipped, released, or allowed to expire. If a metric is part of medical care, safety monitoring, or contractual/work requirements, do not tell me to stop tracking it. Anything consequential or unclear → route to the appropriate human, professional, or official source. If a boundary or confrontation could create danger or retaliation, do not write a script — point me to a trusted or specialized support pathway. Immediate danger → name appropriate emergency/crisis support and stop.

Do not paste: passwords, 2FA codes, financial credentials or account numbers, confidential employer/client material, contracts, private school or medical records, or another person's sensitive information.

One bounded cycle only. End with permission to stop — no automatic follow-up, no next-product offer. Nothing has to feel different for this to count.

Context — optional:

[PASTE]

# One Optional Bridge

# Copy/paste prompt — One Optional Bridge

> Paste everything below into your AI tool — context optional.

---

I am deliberately asking for one small optional bridge toward something outside this exercise — or none.

Offer exactly ONE of these — or none:

- open something without acting on it
- look at one saved note
- place one object nearby
- draft a short request for more time — only if it fits my real obligations
- no bridge — stop here

Do not assume standing, walking, drinking water, breath exercises, or any body movement is right for me.
Do not explain why the option is "safe."
Do not add a task.

One cycle. End.

**Capacity right now · optional** — I may leave it blank, say "I don't know" or "prefer not to label," or declare an NSIF Capacity State (Flowing, Grounded, Recovering, Stretched, Flooded, Frozen, Fractured, Zero Demand). If declared, adapt response length, wording, choice count, and interaction length. If blank or unknown, default to the lowest-load response. Never infer or declare my Capacity State as fact. If I say Zero Demand — or ask for nothing — reply with at most one brief acknowledgment and end. No follow-up question. No offered next step.

Do not diagnose why I feel numb, frozen, blank, heavy, weak, confused, disconnected, far away, or unable to start. If what I describe may involve a new or severe medical issue, injury, medication or substance effects, immediate danger, or another situation requiring professional judgment — stop the normal exercise and direct me toward appropriate human, professional, or emergency support.

You cannot see me or my situation. Do NOT decide or advise on: medical care, eating-disorder or other health treatment, required monitoring or metrics, employment or school obligations, legal or financial matters, contracts, caregiving duties, whether a boundary or conversation is safe, or whether anything can be skipped, released, or allowed to expire. If a metric is part of medical care, safety monitoring, or contractual/work requirements, do not tell me to stop tracking it. Anything consequential or unclear → route to the appropriate human, professional, or official source. If a boundary or confrontation could create danger or retaliation, do not write a script — point me to a trusted or specialized support pathway. Immediate danger → name appropriate emergency/crisis support and stop.

Do not paste: passwords, 2FA codes, financial credentials or account numbers, confidential employer/client material, contracts, private school or medical records, or another person's sensitive information.

One bounded cycle only. End with permission to stop — no automatic follow-up, no next-product offer. Nothing has to feel different for this to count.

Context — optional:

[PASTE]

# Rest After Reentry

# Copy/paste prompt — Rest After Reentry

> Paste everything below into your AI tool.

---

I am stopping. This round is done.

Respond with:

1. One short acknowledgment — for example: "Stopping here is valid."
2. One optional closing line — nothing else is required from this chat.
3. End.

Do not inventory what I did.
Do not list permissions.
Do not suggest a next step or a next session.

**Capacity right now · optional** — I may leave it blank, say "I don't know" or "prefer not to label," or declare an NSIF Capacity State (Flowing, Grounded, Recovering, Stretched, Flooded, Frozen, Fractured, Zero Demand). If declared, adapt response length, wording, choice count, and interaction length. If blank or unknown, default to the lowest-load response. Never infer or declare my Capacity State as fact. If I say Zero Demand — or ask for nothing — reply with at most one brief acknowledgment and end. No follow-up question. No offered next step.

Do not diagnose why I feel numb, frozen, blank, heavy, weak, confused, disconnected, far away, or unable to start. If what I describe may involve a new or severe medical issue, injury, medication or substance effects, immediate danger, or another situation requiring professional judgment — stop the normal exercise and direct me toward appropriate human, professional, or emergency support.

You cannot see me or my situation. Do NOT decide or advise on: medical care, eating-disorder or other health treatment, required monitoring or metrics, employment or school obligations, legal or financial matters, contracts, caregiving duties, whether a boundary or conversation is safe, or whether anything can be skipped, released, or allowed to expire. If a metric is part of medical care, safety monitoring, or contractual/work requirements, do not tell me to stop tracking it. Anything consequential or unclear → route to the appropriate human, professional, or official source. If a boundary or confrontation could create danger or retaliation, do not write a script — point me to a trusted or specialized support pathway. Immediate danger → name appropriate emergency/crisis support and stop.

Do not paste: passwords, 2FA codes, financial credentials or account numbers, confidential employer/client material, contracts, private school or medical records, or another person's sensitive information.

One bounded cycle only. End with permission to stop — no automatic follow-up, no next-product offer. Nothing has to feel different for this to count.