AI Limits-First Prompt Pack

# Copy/paste prompt — Illness & Disability OS Router

---

I am using Illness & Disability OS and I don't know which system to use.

Do not diagnose me.
Do not solve the medical situation.
Do not give treatment, medication, legal, benefits, workplace, school, or accommodation-entitlement advice.
Do not make me compare a long list.
Ask at most one question.

Use a few optional words from me to choose one system only:

01 Limits-First Planning
02 Flare Day Mode
03 Energy Budget + Overextension Check
04 Brain Fog Holding Table
05 Medical Admin Sorter
06 Appointment Prep
07 Accommodation Request Draft
08 Help + Explanation Boundary Scripts
09 Capacity Change / Grief + Identity Support
10 Rest as Access + Close Loop

If multiple fit, choose the lowest-demand useful option.
If you cannot tell, choose 01 Limits-First Planning.
If I say "I don't know," provide a brief low-load response instead of asking me to explain.
If what I describe may involve immediate danger, symptoms needing medical judgment, medication/treatment decisions, or another urgent medical concern, do not route me into a planning system. Tell me to use the appropriate healthcare/emergency pathway.

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 begin the selected system.
Stop after routing.

# Limits-First Planning

# Copy/paste prompt — Limits-First Planning

---

I am planning with chronic illness, disability, or real body limits.

Do not give me a productivity plan. Do not tell me to push through. Do not tell me to be more disciplined. Do not give medical, legal, benefits, workplace, school. Do not create a full routine. Do not add new responsibilities.

Your job is only:

1. Acknowledge in 2 sentences max: my body/capacity is real data, not a mindset problem.

2. Sort what I paste into:
   — Must · — Maybe · — Not today · — Needs help · — Rest / protect capacity · — Medical / official review · — Unknown / unclear

   Do not put uncertain symptoms, medication, treatment, medical-safety questions, or clinical instructions into Must / Maybe / Not Today / Rest. Those go to Medical / official review, and you tell me to use the official instructions or appropriate person.

3. If movement is useful and my capacity allows it, offer one bounded nonclinical step. Otherwise say: "Stop here for now."

4. End with permission to stop.

Rules: Short bullets. No "you should." No productivity shame. No cure language. No giant plan. No pushing through.

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, and task complexity 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.
Do not treat my illness or disability itself as a state to fix — my limits are real data, not a mindset problem.

Work with fragments. Do not require more disclosure than I provide.

Do not diagnose, prescribe, interpret symptoms as medical conclusions, interpret tests/labs/imaging, recommend medication changes, give treatment advice, or determine medical urgency.
Do not decide that a medical instruction, symptom, medication, treatment, or safety issue can wait.
Do not determine accommodation, legal, benefits, workplace, or school eligibility or entitlement.
If part of what I describe needs medical or professional judgment, route that part to the appropriate human or official pathway.
If I describe immediate danger or crisis, stop the normal exercise and direct me toward appropriate human/emergency support.

One cycle only. Do not automatically offer another task or follow-up question. When done, tell me I can stop here.

Privacy: general, non-identifying details only. No full names, addresses, record numbers, or details I would not want in a third-party service — that service's own storage, retention, training, and privacy settings still apply.

Context — optional:

[A FEW WORDS OR SKIP]

**Stop when:** The day matches my body better — not perfectly.

# Flare Day Mode

# Copy/paste prompt — Flare Day Mode

---

I am in a flare or very limited-capacity day.

Do not give me a routine. Do not give me a full plan. Do not tell me to be positive. Do not tell me to push through. Do not give medical, legal, benefits, workplace, school. Do not ask me to explain my symptoms in detail.

Your job is only:

1. Acknowledge in 2 sentences max: flare days are not consistency failures; today is for reducing demand.

2. Help me sort:
   — Minimum necessary day · — Known nonclinical things that can wait · — What needs help · — What needs an appropriate support pathway · — Medical / official review · — One message I can send, if needed · — Rest / protect capacity

   Anything involving symptom, medication, treatment, or medical-safety uncertainty goes to Medical / official review — never "can wait."

3. Give me one short message draft if I need to cancel, delay, or ask for help.

4. End with: "Today is for reducing demand."

Rules: Short bullets. No diagnosis. No treatment advice. No full plan. No guilt. No "keep going."

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, and task complexity 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.
Do not treat my illness or disability itself as a state to fix — my limits are real data, not a mindset problem.

Work with fragments. Do not require more disclosure than I provide.

Do not diagnose, prescribe, interpret symptoms as medical conclusions, interpret tests/labs/imaging, recommend medication changes, give treatment advice, or determine medical urgency.
Do not decide that a medical instruction, symptom, medication, treatment, or safety issue can wait.
Do not determine accommodation, legal, benefits, workplace, or school eligibility or entitlement.
If part of what I describe needs medical or professional judgment, route that part to the appropriate human or official pathway.
If I describe immediate danger or crisis, stop the normal exercise and direct me toward appropriate human/emergency support.

One cycle only. Do not automatically offer another task or follow-up question. When done, tell me I can stop here.

Privacy: general, non-identifying details only. No full names, addresses, record numbers, or details I would not want in a third-party service — that service's own storage, retention, training, and privacy settings still apply.

Context — optional:

[A FEW WORDS OR SKIP]

**Stop when:** Stop whenever enough. The flare does not have to feel better for this to count.

# Energy Budget + Overextension Check

# Copy/paste prompt — Energy Budget + Overextension Check

---

I need help checking today's energy budget against real body limits.

Do not maximize my output. Do not tell me to push through. Do not tell me to use a good day to catch up. Do not give medical, legal, benefits, workplace, school. Do not create a productivity schedule. Do not promise prevention — help me check for overextension against what I already know about my own patterns.

Your job is only:

1. Acknowledge briefly: a good day is not proof that limits disappeared.

2. Sort what I paste into:
   — Must · — Maybe · — Not today · — Needs help · — May cost more capacity later · — Could be smaller · — Rest / protect capacity

   Anything involving symptom, medication, treatment, or medical-safety uncertainty goes to Medical / official review.

3. If useful, suggest one optional smaller version of one task — only if I ask for reduction ideas.

4. Name one stop point.

5. End with permission to stop while I still have a little capacity left.

Rules: Short bullets. No hustle language. No "catch up while you can." No shame. No full routine. No new tasks.

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, and task complexity 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.
Do not treat my illness or disability itself as a state to fix — my limits are real data, not a mindset problem.

Work with fragments. Do not require more disclosure than I provide.

Do not diagnose, prescribe, interpret symptoms as medical conclusions, interpret tests/labs/imaging, recommend medication changes, give treatment advice, or determine medical urgency.
Do not decide that a medical instruction, symptom, medication, treatment, or safety issue can wait.
Do not determine accommodation, legal, benefits, workplace, or school eligibility or entitlement.
If part of what I describe needs medical or professional judgment, route that part to the appropriate human or official pathway.
If I describe immediate danger or crisis, stop the normal exercise and direct me toward appropriate human/emergency support.

One cycle only. Do not automatically offer another task or follow-up question. When done, tell me I can stop here.

Privacy: general, non-identifying details only. No full names, addresses, record numbers, or details I would not want in a third-party service — that service's own storage, retention, training, and privacy settings still apply.

Context — optional:

[A FEW WORDS OR SKIP]

**Stop when:** I have a stop point that fits what I know about my own patterns.

# Brain Fog Holding Table

# Copy/paste prompt — Brain Fog Holding Table

---

I have brain fog and need help reducing cognitive load.

Do not give me a full plan. Do not give me many choices. Do not make me explain everything. Do not imply I am stupid or lazy. Do not give medical, legal, benefits, workplace, school.

Your job is only:

1. Say in 2 sentences max: brain fog can reduce cognitive capacity — it is not a character flaw.

2. Help me create a simple holding table with:
   — One sentence summary · — One document/tab/item involved · — One decision, if any · — One thing to save for later · — One tiny next touch — only if useful · — Stop point

3. If I paste messy context, simplify it without adding new tasks.

4. End with permission to stop.

Rules: Short lines. No long explanation. No more than one tiny touch — and none is allowed. No productivity pressure. No shame.

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, and task complexity 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.
Do not treat my illness or disability itself as a state to fix — my limits are real data, not a mindset problem.

Work with fragments. Do not require more disclosure than I provide.

Do not diagnose, prescribe, interpret symptoms as medical conclusions, interpret tests/labs/imaging, recommend medication changes, give treatment advice, or determine medical urgency.
Do not decide that a medical instruction, symptom, medication, treatment, or safety issue can wait.
Do not determine accommodation, legal, benefits, workplace, or school eligibility or entitlement.
If part of what I describe needs medical or professional judgment, route that part to the appropriate human or official pathway.
If I describe immediate danger or crisis, stop the normal exercise and direct me toward appropriate human/emergency support.

One cycle only. Do not automatically offer another task or follow-up question. When done, tell me I can stop here.

Privacy: general, non-identifying details only. No full names, addresses, record numbers, or details I would not want in a third-party service — that service's own storage, retention, training, and privacy settings still apply.

Context — optional:

[A FEW WORDS OR SKIP]

**Stop when:** My brain is holding less — or whenever I stop.

# Medical Admin Sorter

# Copy/paste prompt — Medical Admin Sorter

---

I am overwhelmed by medical or disability admin.

Do not give medical advice. Do not give legal advice. Do not give benefits advice. Do not give financial advice. Do not give workplace or school advice. Do not tell me what official action to take. Do not create a full admin plan. Do not classify the medical urgency of anything.

Use only general information.

Your job is only:

1. Acknowledge in 2 sentences max: medical/disability admin is real labor and can overload capacity.

2. Sort what I paste into:
   — Appointments · — Forms · — Portals / messages · — Referrals / follow-ups · — Insurance / benefits · — Bills · — Supplies · — Transportation · — Waiting on someone · — Known nonclinical admin that can wait · — Needs support · — Medical / official review · — Unknown / unclear

   If anything I paste concerns symptoms, medication timing, treatment, test interpretation, medical instructions, or urgency — it goes to Medical / official review. Never into "can wait." Tell me to use the official instructions or appropriate person for that item.

3. Name one tiny admin touch — only if useful: open the portal, write the document name, find a phone number, make a question list, ask someone for help, or stop.

4. End with permission to stop.

Rules: Short bullets. No official advice. No invented deadlines. No shame. No full plan.

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, and task complexity 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.
Do not treat my illness or disability itself as a state to fix — my limits are real data, not a mindset problem.

Work with fragments. Do not require more disclosure than I provide.

Do not diagnose, prescribe, interpret symptoms as medical conclusions, interpret tests/labs/imaging, recommend medication changes, give treatment advice, or determine medical urgency.
Do not decide that a medical instruction, symptom, medication, treatment, or safety issue can wait.
Do not determine accommodation, legal, benefits, workplace, or school eligibility or entitlement.
If part of what I describe needs medical or professional judgment, route that part to the appropriate human or official pathway.
If I describe immediate danger or crisis, stop the normal exercise and direct me toward appropriate human/emergency support.

One cycle only. Do not automatically offer another task or follow-up question. When done, tell me I can stop here.

Privacy: general, non-identifying details only. No full names, addresses, record numbers, or details I would not want in a third-party service — that service's own storage, retention, training, and privacy settings still apply.

Context — optional:

[A FEW WORDS OR SKIP — general, non-identifying only]

**Stop when:** The admin load feels nameable, not solved.

# Appointment Prep

# Copy/paste prompt — Appointment Prep

---

I need help preparing for an appointment without overwhelming myself.

Do not give medical advice. Do not diagnose me. Do not suggest treatment. Do not tell me what medication, test, or diagnosis to ask for. Do not give legal, benefits, workplace, school.

Your job is only to help me organize my own notes.

Organize only information I give you. Do not generate additional symptoms, tests, treatments, medications, diagnoses, or medical recommendations — unless I explicitly ask for non-medical wording help.

Create a simple appointment prep note with whichever of these I can supply — every field is optional:

1. Appointment purpose · 2. Top concern · 3. One question I want to ask · 4. One example from daily life · 5. One access/accommodation need, if relevant · 6. What I may need after the appointment · 7. One sentence I can read if I get overwhelmed

Rules: Short bullets. No medical recommendations. No official advice. No long script. No shame. Do not force completion of any field.

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, and task complexity 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.
Do not treat my illness or disability itself as a state to fix — my limits are real data, not a mindset problem.

Work with fragments. Do not require more disclosure than I provide.

Do not diagnose, prescribe, interpret symptoms as medical conclusions, interpret tests/labs/imaging, recommend medication changes, give treatment advice, or determine medical urgency.
Do not decide that a medical instruction, symptom, medication, treatment, or safety issue can wait.
Do not determine accommodation, legal, benefits, workplace, or school eligibility or entitlement.
If part of what I describe needs medical or professional judgment, route that part to the appropriate human or official pathway.
If I describe immediate danger or crisis, stop the normal exercise and direct me toward appropriate human/emergency support.

One cycle only. Do not automatically offer another task or follow-up question. When done, tell me I can stop here.

Privacy: general, non-identifying details only. No full names, addresses, record numbers, or details I would not want in a third-party service — that service's own storage, retention, training, and privacy settings still apply.

Context — optional:

[A FEW WORDS OR SKIP]

**Stop when:** I have a simple appointment note I can edit.

# Accommodation Request Draft

# Copy/paste prompt — Accommodation Request Draft

---

I need help drafting a simple accommodation or access request.

Do not give legal advice. Do not determine what I am legally entitled to. Do not determine eligibility. Do not tell me what official process applies. Do not diagnose me. Do not overexplain my medical history. Do not give workplace, school, benefits, medical.

Your job is only to help me draft clear language I can edit.

Create one concise draft with:

1. The barrier or access need · 2. The task or environment affected · 3. The adjustment requested · 4. How it may support access, participation, or completion · 5. What I choose not to disclose · 6. A neutral closing line

Tone: neutral, clear, not apologetic, not oversharing, not aggressive.

One draft only — do not give me versions to compare unless I explicitly ask for alternatives.

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, and task complexity 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.
Do not treat my illness or disability itself as a state to fix — my limits are real data, not a mindset problem.

Work with fragments. Do not require more disclosure than I provide.

Do not diagnose, prescribe, interpret symptoms as medical conclusions, interpret tests/labs/imaging, recommend medication changes, give treatment advice, or determine medical urgency.
Do not decide that a medical instruction, symptom, medication, treatment, or safety issue can wait.
Do not determine accommodation, legal, benefits, workplace, or school eligibility or entitlement.
If part of what I describe needs medical or professional judgment, route that part to the appropriate human or official pathway.
If I describe immediate danger or crisis, stop the normal exercise and direct me toward appropriate human/emergency support.

One cycle only. Do not automatically offer another task or follow-up question. When done, tell me I can stop here.

Privacy: general, non-identifying details only. No full names, addresses, record numbers, or details I would not want in a third-party service — that service's own storage, retention, training, and privacy settings still apply.

Context — optional:

[A FEW WORDS OR SKIP — general, non-identifying only]

**Stop when:** I have a draft I can edit or take to the right support pathway.

# Help + Explanation Boundary Scripts

# Copy/paste prompt — Help + Explanation Boundary Scripts

---

I need help writing a short script about my illness/disability/capacity limits.

Do not make me overexplain. Do not make me prove my limit. Do not make me disclose private medical details. Do not shame me. Do not give legal, medical, workplace, school, benefits.

Choose the ONE situation that best fits my context and write one short script for it:

1. Asking for help with one specific thing · 2. Explaining that my capacity changed · 3. Declining because something is not accessible today · 4. Requesting a lower-demand option · 5. Ending advice I did not ask for · 6. Saying I am not discussing whether my limit is real

If my context is unclear, write one neutral, low-disclosure script — do not ask more than one question.

Tone: clear, calm, not apologetic, not harsh unless needed, low-energy.

Rules: One script only. No long explanation. No medical details. No "sorry" unless I request it. No debate language.

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, and task complexity 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.
Do not treat my illness or disability itself as a state to fix — my limits are real data, not a mindset problem.

Work with fragments. Do not require more disclosure than I provide.

Do not diagnose, prescribe, interpret symptoms as medical conclusions, interpret tests/labs/imaging, recommend medication changes, give treatment advice, or determine medical urgency.
Do not decide that a medical instruction, symptom, medication, treatment, or safety issue can wait.
Do not determine accommodation, legal, benefits, workplace, or school eligibility or entitlement.
If part of what I describe needs medical or professional judgment, route that part to the appropriate human or official pathway.
If I describe immediate danger or crisis, stop the normal exercise and direct me toward appropriate human/emergency support.

One cycle only. Do not automatically offer another task or follow-up question. When done, tell me I can stop here.

Privacy: general, non-identifying details only. No full names, addresses, record numbers, or details I would not want in a third-party service — that service's own storage, retention, training, and privacy settings still apply.

Context — optional:

[A FEW WORDS OR SKIP]

**Stop when:** I have one script I can use, edit, or save.

# Capacity Change / Grief + Identity Support

# Copy/paste prompt — Capacity Change / Grief + Identity Support

---

Something about my body, health, disability, or capacity has changed and it needs somewhere to go.

Do not fix it. Do not reframe it too quickly. Do not tell me to be grateful. Do not tell me everything happens for a reason. Do not make meaning from it unless I ask. Do not turn my illness/disability into inspiration. Do not assume I am grieving — if grief or loss is part of this, it can be real and worth naming. Do not give medical, legal, benefits, workplace, school.

Your job is only:

1. Acknowledge in 3 short sentences max: a change in capacity can be real and worth naming.

2. Help me name — only what I offer:
   — What I miss · — What changed · — What I am angry or sad about · — What I do not want minimized · — What still feels like me, if anything · — One grounded sentence reflecting what I said without minimizing it

3. End with permission to stop — no next steps unless I ask.

Rules: No fixing. No silver lining. No productivity. No "at least." No forced acceptance. No forced hope.

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, and task complexity 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.
Do not treat my illness or disability itself as a state to fix — my limits are real data, not a mindset problem.

Work with fragments. Do not require more disclosure than I provide.

Do not diagnose, prescribe, interpret symptoms as medical conclusions, interpret tests/labs/imaging, recommend medication changes, give treatment advice, or determine medical urgency.
Do not decide that a medical instruction, symptom, medication, treatment, or safety issue can wait.
Do not determine accommodation, legal, benefits, workplace, or school eligibility or entitlement.
If part of what I describe needs medical or professional judgment, route that part to the appropriate human or official pathway.
If I describe immediate danger or crisis, stop the normal exercise and direct me toward appropriate human/emergency support.

One cycle only. Do not automatically offer another task or follow-up question. When done, tell me I can stop here.

Privacy: general, non-identifying details only. No full names, addresses, record numbers, or details I would not want in a third-party service — that service's own storage, retention, training, and privacy settings still apply.

Context — optional:

[A FEW WORDS OR SKIP]

**Stop when:** This feels witnessed, not solved — or whenever I stop.

# Rest as Access + Close Loop

# Copy/paste prompt — Rest as Access + Close Loop

---

I need to rest or close the loop with illness/disability limits.

Do not give me another task. Do not encourage me to keep going. Do not say "since you already started." Do not turn rest into something I have to earn. Do not give medical, legal, workplace, school, benefits. Do not decide whether anything medical can wait.

Your job is only:

1. Acknowledge in 2 sentences max: rest is access, and stopping can protect capacity.

2. Help me create a close-loop note:
   — What is held here now · — What I already know can wait · — What still needs medical / official review · — What rest protects · — Where I can return later · — What I do not need to explain · — One sentence for future me

3. Give one short closing permission.

4. End with: "You did enough for this round."

Rules: No next task. No momentum language. No shame. No streaks. No "get back on track." No follow-up question.

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, and task complexity 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.
Do not treat my illness or disability itself as a state to fix — my limits are real data, not a mindset problem.

Work with fragments. Do not require more disclosure than I provide.

Do not diagnose, prescribe, interpret symptoms as medical conclusions, interpret tests/labs/imaging, recommend medication changes, give treatment advice, or determine medical urgency.
Do not decide that a medical instruction, symptom, medication, treatment, or safety issue can wait.
Do not determine accommodation, legal, benefits, workplace, or school eligibility or entitlement.
If part of what I describe needs medical or professional judgment, route that part to the appropriate human or official pathway.
If I describe immediate danger or crisis, stop the normal exercise and direct me toward appropriate human/emergency support.

One cycle only. Do not automatically offer another task or follow-up question. When done, tell me I can stop here.

Privacy: general, non-identifying details only. No full names, addresses, record numbers, or details I would not want in a third-party service — that service's own storage, retention, training, and privacy settings still apply.

Context — optional:

[A FEW WORDS OR SKIP]

**Stop when:** Stop whenever enough. Rest does not have to feel easy for this to count.