# Low Capacity OS — AI Prompt Pack · AI Prompt Pack *by Whelmish · Operational relief for whatever your bandwidth.* Capacity-safe operational infrastructure for overwhelmed minds. --- # Low Capacity OS ## AI Prompt Pack *Bounded AI support for overwhelmed minds.* --- Capacity-aware AI support for mental overload — not therapy, not a planner, not a hustle prompt list. **Eight bounded systems. Copy, paste, stop.** One prompt is a complete session. --- # For the days when reading a prompt list is already too much If you're opening this while overloaded, frozen, or behind — that's who this was made for. Each page is one short, contained AI prompt. You copy it, paste it, get one bounded answer, and stop. You can: - Use one system only - Stop mid-prompt - Return days later — nothing expires here - Skip anything that feels like too much **One prompt is a complete session.** --- # Nothing right now? If opening AI itself feels like more demand, close this pack. Nothing else is required. No prompt. No sorting. No decision. No explanation. Return whenever you want. --- # Need a person instead? This pack can wait. Depending on the issue, appropriate support may include: - a trusted person - a healthcare or mental-health professional - a caregiver or support person - workplace or school support - an appropriate legal, financial, or other professional resource - emergency or crisis support when immediate danger is involved You do not need to complete an AI prompt first. --- # How to use this pack with AI No third-party AI platform is certified by this pack — these prompts only shape what happens inside one conversation. ## Before you paste 1. Pick **one** system — or use the Low-Capacity Prompt Router if choosing feels like too much. 2. Copy the whole block — from **COPY FROM HERE** to **END PROMPT**. 3. Paste into your AI tool. ## During - Answer honestly. Fragments and mess are fine. - If the AI starts planning your whole life, **stop** — re-paste the prompt or close the chat. - You can say "stop" at any point. Stopping counts. ## After - You do **not** need another round. - Partial counts. Choosing not to act counts. - If something lands as too heavy, **Emotional Holding** (05) is the lightest place to put it. ## Privacy note This PDF has no account or tracking requirement — **your AI provider does**. Its storage, retention, training, and privacy settings still apply to anything you paste. Do not paste passwords, 2FA codes, account numbers, full medical/legal/financial records, confidential work or school data, or another person's sensitive information. ## What we never do here - Streaks or "don't break the chain" - Catch-up or "you're behind" - Clinical or therapy claims - Infinite chat — every system is designed to **end** --- # What this is not - **Not** mental health treatment, crisis support, or medical advice - **Not** a productivity system, habit tracker, or optimization tool - **Not** a test you can fail — no scores, streaks, or catch-up - **Not** "100 ChatGPT hacks" — eight bounded systems only - **Not** a replacement for a person or professional when the situation needs one These prompts do not decide whether anything real can wait. If something involves health, safety, legal, financial, caregiving, work, school, housing, benefits, or another consequential obligation and you are unsure — use the appropriate person or official source. If you are in danger or crisis, contact local emergency or crisis services — this pack cannot see you or intervene. --- # Why these prompts end Most AI tools keep expanding the conversation. These prompts are built the other way: **input → one bounded response → stop.** Each system is one cycle. One prompt can be the whole session — a successful use of this pack might last twenty seconds. The goal is lower demand and a clear stopping point — not a calmer feeling you have to reach, and never another prompt you have to run. Built around Capacity-Safe design principles: **the system adjusts to the human.** --- # Low-Capacity Prompt Router Choosing feels like too much? Use 01 · Brain Dump Sorter — or close the pack. One system is a complete session. ## 01 · Brain Dump Sorter My head is too loud — somewhere to put it Intro p. 9 · Prompt p. 10 ## 02 · Low-Capacity Orientation I feel shut down or frozen Intro p. 12 · Prompt p. 13 ## 03 · One Bounded Step I want one tiny move — or none Intro p. 15 · Prompt p. 16 ## 04 · One Thing Today I need today made smaller Intro p. 18 · Prompt p. 19 ## 05 · Emotional Holding Emotional pressure is loud Intro p. 21 · Prompt p. 22 ## 06 · Mental Load Splitter Invisible mental load is everywhere Intro p. 24 · Prompt p. 25 ## 07 · Low Capacity Situation Clarity The situation is tangled — make it smaller, not solved Intro p. 27 · Prompt p. 28 ## 08 · Gentle Return I'm returning after time away Intro p. 30 · Prompt p. 31 Low-Capacity Prompt Router · p. 33 No chains. One prompt, one cycle, then stop. --- # System 01 — Brain Dump Sorter # Copy/paste prompt — Brain Dump Sorter > Paste everything below into your AI tool. --- I'm going to paste a messy brain dump. My capacity is low. Do not coach me, optimize my life, or give me a long plan. Your job is only: 1. Acknowledge in one or two sentences — before any sorting. 2. Sort what I paste into these buckets (leave empty buckets blank): - **I already know needs attention soon** - **I already know can wait** - **Emotional load** — worries, shame loops, not tasks - **Needs human / professional judgment** - **Unclear — hold** 3. If anything could involve health, safety, legal, financial, caregiving, work or school obligations, housing, benefits, or another consequential matter — put it in "Needs human / professional judgment," even if I placed it somewhere else. You are not deciding urgency; only I am, or a person is. 4. Give me one short sentence on what looks heaviest — reflection, not advice. 5. End with permission to stop. Rules for you: - Maximum **8 items total** in the output. Group similar entries into one line. - If more remains, say only: "More is held here. You do not need to review it now." - Short bullets only. No essays. No "you should." - Do not add tasks I didn't mention. Capacity right now — optional: [leave blank / I don't know / prefer not to label / declared NSIF state] If I name a Capacity State, adapt your response: language, length, number of choices, task complexity, and interaction length. If I leave this blank or say "I don't know," default to the lowest-load response — the shortest output, one choice at most. Never infer or declare my Capacity State. If I say I have no capacity for this right now — or describe needing nothing — do not continue a normal exercise. A minimal acknowledgment, a human-support option if relevant, and stop. Rules: Acknowledge briefly first, before sorting or analyzing. Work with fragments — do not require more disclosure than I give. Do only what I asked; do not expand scope. Do not diagnose me, interpret my nervous system, or act as a therapist. Do NOT decide whether anything can wait — medical, safety, legal, financial, caregiving, workplace, school, housing, benefits, or any other consequential obligation. If I explicitly said something can wait, preserve that. Otherwise place it under "Needs human / professional / official judgment." Do not manufacture urgency or deferral. If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support — do not continue the exercise. 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, account numbers, full medical/legal/financial records, confidential work or school data, or another person's sensitive information. My AI provider's storage, retention, training, and privacy settings still apply. Here is my brain dump: ``` [PASTE HERE — MESSY IS FINE] ``` --- **Stop whenever enough.** You do not have to fill every bucket. --- # System 02 — Low-Capacity Orientation # Copy/paste prompt — Low-Capacity Orientation > Paste everything below into your AI tool. --- I feel shut down or frozen — or something close to that. Do not diagnose me or explain why I feel this way. Do not give me a task list, a life plan, or productivity advice. Your job is only: 1. Acknowledge in one or two sentences. No shame, no interpretation. 2. Offer one optional orientation choice — framed as a choice, never an instruction. Examples: notice a support surface under me; loosen one muscle only if that's comfortable; let one natural breath happen without changing it; notice one neutral sight or sound. 3. Offer one only — not a sequence. 4. If I say none of them work — or I don't respond — stop. Do not escalate into more techniques, more questions, or planning. 5. End with permission to stop or rest. Capacity right now — optional: [leave blank / I don't know / prefer not to label / declared NSIF state] If I name a Capacity State, adapt your response: language, length, number of choices, task complexity, and interaction length. If I leave this blank or say "I don't know," default to the lowest-load response — the shortest output, one choice at most. Never infer or declare my Capacity State. If I say I have no capacity for this right now — or describe needing nothing — do not continue a normal exercise. A minimal acknowledgment, a human-support option if relevant, and stop. Rules: Acknowledge briefly first, before sorting or analyzing. Work with fragments — do not require more disclosure than I give. Do only what I asked; do not expand scope. Do not diagnose me, interpret my nervous system, or act as a therapist. Do NOT decide whether anything can wait — medical, safety, legal, financial, caregiving, workplace, school, housing, benefits, or any other consequential obligation. If I explicitly said something can wait, preserve that. Otherwise place it under "Needs human / professional / official judgment." Do not manufacture urgency or deferral. If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support — do not continue the exercise. 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, account numbers, full medical/legal/financial records, confidential work or school data, or another person's sensitive information. My AI provider's storage, retention, training, and privacy settings still apply. How I'm arriving — optional: [A FEW WORDS OR SKIP] --- **Stop whenever enough.** Choosing none counts. --- # System 03 — One Bounded Step # Copy/paste prompt — One Bounded Step > Paste everything below into your AI tool. --- My capacity is low. I may want one bounded step — or none. Do not give me multiple options. Do not give me a plan. Do not optimize anything. Do not add new tasks. Your job is only: 1. Acknowledge briefly — one or two sentences. No shame. 2. If useful, offer exactly one small, reversible step — something that can be undone without cost. If a step is not right, return "Stop here for now." 3. Do not claim any action is "safe" or "safest." If anything involves health, safety, legal, financial, caregiving, work, school, or another consequential matter, say it belongs with the appropriate person — not this prompt. 4. End with permission to do it, skip it, or rest. Capacity right now — optional: [leave blank / I don't know / prefer not to label / declared NSIF state] If I name a Capacity State, adapt your response: language, length, number of choices, task complexity, and interaction length. If I leave this blank or say "I don't know," default to the lowest-load response — the shortest output, one choice at most. Never infer or declare my Capacity State. If I say I have no capacity for this right now — or describe needing nothing — do not continue a normal exercise. A minimal acknowledgment, a human-support option if relevant, and stop. Rules: Acknowledge briefly first, before sorting or analyzing. Work with fragments — do not require more disclosure than I give. Do only what I asked; do not expand scope. Do not diagnose me, interpret my nervous system, or act as a therapist. Do NOT decide whether anything can wait — medical, safety, legal, financial, caregiving, workplace, school, housing, benefits, or any other consequential obligation. If I explicitly said something can wait, preserve that. Otherwise place it under "Needs human / professional / official judgment." Do not manufacture urgency or deferral. If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support — do not continue the exercise. 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, account numbers, full medical/legal/financial records, confidential work or school data, or another person's sensitive information. My AI provider's storage, retention, training, and privacy settings still apply. My situation — optional: [DESCRIBE, OR SKIP] --- **Stop whenever enough.** Rest is a complete answer. --- # System 04 — One Thing Today # Copy/paste prompt — One Thing Today > Paste everything below into your AI tool. --- My capacity is low. I need today made smaller — not planned. I'm going to list what's on my mind for today, roughly. Do not optimize it, expand it, or turn it into a system. Your job is only: 1. Acknowledge in one or two sentences. 2. Return **one thing** — the one I already named as mattering, or the one you can reflect back as carrying the most weight in my own words. If nothing stands out, say: "Nothing has to be picked today." 3. Only if I explicitly said I have room for more, return up to three. 4. Everything else goes under **Held outside today's exercise** — do not label it "can wait" unless I already said that. 5. Anything involving health, safety, legal, financial, caregiving, work, school, housing, benefits, or another consequential matter goes under **Needs human / professional judgment** — not on the list at all. 6. End with permission to stop. Capacity right now — optional: [leave blank / I don't know / prefer not to label / declared NSIF state] If I name a Capacity State, adapt your response: language, length, number of choices, task complexity, and interaction length. If I leave this blank or say "I don't know," default to the lowest-load response — the shortest output, one choice at most. Never infer or declare my Capacity State. If I say I have no capacity for this right now — or describe needing nothing — do not continue a normal exercise. A minimal acknowledgment, a human-support option if relevant, and stop. Rules: Acknowledge briefly first, before sorting or analyzing. Work with fragments — do not require more disclosure than I give. Do only what I asked; do not expand scope. Do not diagnose me, interpret my nervous system, or act as a therapist. Do NOT decide whether anything can wait — medical, safety, legal, financial, caregiving, workplace, school, housing, benefits, or any other consequential obligation. If I explicitly said something can wait, preserve that. Otherwise place it under "Needs human / professional / official judgment." Do not manufacture urgency or deferral. If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support — do not continue the exercise. 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, account numbers, full medical/legal/financial records, confidential work or school data, or another person's sensitive information. My AI provider's storage, retention, training, and privacy settings still apply. What's on today — rough is fine: ``` [PASTE OR LIST — MESSY IS FINE] ``` --- **Stop whenever enough.** One thing — or none — is a complete answer. --- # System 05 — Emotional Holding # Copy/paste prompt — Emotional Holding > Paste everything below into your AI tool. --- Something feels emotionally heavy and I need it held — not fixed, not analyzed, not reframed. Do not tell me why I feel this way. Do not diagnose me or interpret my nervous system. Do not give me tasks, homework, or a plan. Your job is only: 1. Reflect what I said — briefly, in my own words. 2. One short, grounded acknowledgment. 3. Then either one low-demand question — or stop, whichever is lighter. 4. If useful, remind me I do not have to solve anything in this chat. 5. End. Capacity right now — optional: [leave blank / I don't know / prefer not to label / declared NSIF state] If I name a Capacity State, adapt your response: language, length, number of choices, task complexity, and interaction length. If I leave this blank or say "I don't know," default to the lowest-load response — the shortest output, one choice at most. Never infer or declare my Capacity State. If I say I have no capacity for this right now — or describe needing nothing — do not continue a normal exercise. A minimal acknowledgment, a human-support option if relevant, and stop. Rules: Acknowledge briefly first, before sorting or analyzing. Work with fragments — do not require more disclosure than I give. Do only what I asked; do not expand scope. Do not diagnose me, interpret my nervous system, or act as a therapist. Do NOT decide whether anything can wait — medical, safety, legal, financial, caregiving, workplace, school, housing, benefits, or any other consequential obligation. If I explicitly said something can wait, preserve that. Otherwise place it under "Needs human / professional / official judgment." Do not manufacture urgency or deferral. If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support — do not continue the exercise. 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, account numbers, full medical/legal/financial records, confidential work or school data, or another person's sensitive information. My AI provider's storage, retention, training, and privacy settings still apply. What's heavy — optional: [A FEW WORDS OR SKIP] --- **Stop whenever enough.** You do not have to feel better for this to count. --- # System 06 — Mental Load Splitter # Copy/paste prompt — Mental Load Splitter > Paste everything below into your AI tool. --- I'm carrying invisible mental load — the remembering, tracking, and holding nobody sees. I'm going to dump it roughly. Do not judge it, optimize it, or turn it into a delegation plan. Your job is only: 1. Acknowledge in one or two sentences — before any sorting. 2. Sort what I paste into these buckets (leave empty buckets blank): - **Invisible load / unsorted** - **I already know needs attention soon** - **I already know can wait** - **Emotional load** - **Needs human / professional judgment** - **Unclear** 3. If anything could involve health, safety, legal, financial, caregiving, work or school obligations, housing, benefits, or another consequential matter — put it in "Needs human / professional judgment," even if I placed it somewhere else. You are not deciding urgency; only I am, or a person is. 4. One sentence naming that this was a lot to hold — reflection, not advice. 5. End with permission to stop. Rules for you: - Maximum **8 items total** in the output. Group similar entries into one line. - If more remains, say only: "More is held here. You do not need to review it now." - Unsorted is a valid end state — do not push me to resolve it. - Do not add anything I didn't mention. Capacity right now — optional: [leave blank / I don't know / prefer not to label / declared NSIF state] If I name a Capacity State, adapt your response: language, length, number of choices, task complexity, and interaction length. If I leave this blank or say "I don't know," default to the lowest-load response — the shortest output, one choice at most. Never infer or declare my Capacity State. If I say I have no capacity for this right now — or describe needing nothing — do not continue a normal exercise. A minimal acknowledgment, a human-support option if relevant, and stop. Rules: Acknowledge briefly first, before sorting or analyzing. Work with fragments — do not require more disclosure than I give. Do only what I asked; do not expand scope. Do not diagnose me, interpret my nervous system, or act as a therapist. Do NOT decide whether anything can wait — medical, safety, legal, financial, caregiving, workplace, school, housing, benefits, or any other consequential obligation. If I explicitly said something can wait, preserve that. Otherwise place it under "Needs human / professional / official judgment." Do not manufacture urgency or deferral. If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support — do not continue the exercise. 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, account numbers, full medical/legal/financial records, confidential work or school data, or another person's sensitive information. My AI provider's storage, retention, training, and privacy settings still apply. The load — rough is fine: ``` [PASTE OR LIST — MESSY IS FINE] ``` --- **Stop whenever enough.** Unsorted counts. --- # System 07 — Low Capacity Situation Clarity # Copy/paste prompt — Low Capacity Situation Clarity > Paste everything below into your AI tool. --- Something feels tangled and I need it made smaller — not solved. I'll describe it roughly. Do not turn it into a project. Your job is only: 1. Brief acknowledgment first. 2. Reflect back at most two short points — in my words, simplified. 3. Name one core tension — only if it's actually clear. If it isn't, say you're not sure instead of guessing. 4. Then either one option — or rest. Only offer a second option if I explicitly said I have room for more. Never more than two. 5. One cycle only, then stop. Capacity right now — optional: [leave blank / I don't know / prefer not to label / declared NSIF state] If I name a Capacity State, adapt your response: language, length, number of choices, task complexity, and interaction length. If I leave this blank or say "I don't know," default to the lowest-load response — the shortest output, one choice at most. Never infer or declare my Capacity State. If I say I have no capacity for this right now — or describe needing nothing — do not continue a normal exercise. A minimal acknowledgment, a human-support option if relevant, and stop. Rules: Acknowledge briefly first, before sorting or analyzing. Work with fragments — do not require more disclosure than I give. Do only what I asked; do not expand scope. Do not diagnose me, interpret my nervous system, or act as a therapist. Do NOT decide whether anything can wait — medical, safety, legal, financial, caregiving, workplace, school, housing, benefits, or any other consequential obligation. If I explicitly said something can wait, preserve that. Otherwise place it under "Needs human / professional / official judgment." Do not manufacture urgency or deferral. If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support — do not continue the exercise. 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, account numbers, full medical/legal/financial records, confidential work or school data, or another person's sensitive information. My AI provider's storage, retention, training, and privacy settings still apply. The situation — rough is fine: ``` [DESCRIBE — MESSY IS FINE] ``` --- **Stop whenever enough.** It doesn't have to feel resolved. --- # System 08 — Gentle Return # Copy/paste prompt — Gentle Return > Paste everything below into your AI tool. --- I am returning after time away. Do not ask or guess why I was away. Do not call it avoidance or failure. Do not make a catch-up plan. Do not tell me to rebuild momentum. Your job is only: 1. Acknowledge in one sentence that returning counts on its own. 2. Offer a quiet landing — I pick one or none: - **Safety** — a quiet landing, nothing required - **Brain Dump** — somewhere to put it - **Rest** — nothing required 3. Be explicit: no catch-up. Nothing to make up for. 4. If even a landing feels like too much, say that stopping here is a complete session. 5. End. Capacity right now — optional: [leave blank / I don't know / prefer not to label / declared NSIF state] If I name a Capacity State, adapt your response: language, length, number of choices, task complexity, and interaction length. If I leave this blank or say "I don't know," default to the lowest-load response — the shortest output, one choice at most. Never infer or declare my Capacity State. If I say I have no capacity for this right now — or describe needing nothing — do not continue a normal exercise. A minimal acknowledgment, a human-support option if relevant, and stop. Rules: Acknowledge briefly first, before sorting or analyzing. Work with fragments — do not require more disclosure than I give. Do only what I asked; do not expand scope. Do not diagnose me, interpret my nervous system, or act as a therapist. Do NOT decide whether anything can wait — medical, safety, legal, financial, caregiving, workplace, school, housing, benefits, or any other consequential obligation. If I explicitly said something can wait, preserve that. Otherwise place it under "Needs human / professional / official judgment." Do not manufacture urgency or deferral. If I describe immediate danger or crisis, stop this exercise and direct me toward appropriate human or emergency support — do not continue the exercise. 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, account numbers, full medical/legal/financial records, confidential work or school data, or another person's sensitive information. My AI provider's storage, retention, training, and privacy settings still apply. Anything I want to say — optional: [A FEW WORDS OR SKIP] --- **Stop whenever enough.** Returning was the action. --- # Low-Capacity Prompt Router # Copy/paste — Low-Capacity Prompt Router > Paste everything below into your AI tool. --- I don't know which Low Capacity OS 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 long list. Ask at most one question. Use the few words I provide to choose one system only: 01 Brain Dump Sorter 02 Low-Capacity Orientation 03 One Bounded Step 04 One Thing Today 05 Emotional Holding 06 Mental Load Splitter 07 Low Capacity Situation Clarity 08 Gentle Return If several fit, choose the lowest-demand option. If I say "I don't know" or give too little information, choose 01 Brain Dump Sorter. If even a prompt seems like too much right now, tell me I can stop instead — that is a complete answer. 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. --- # That's the end of the pack You did enough for today. One prompt is a complete session. Closing this file is a complete session. This pack does not expire. Real-world obligations keep their actual timelines. Return whenever you want. Nothing to catch up. --- **Low Capacity OS · AI Prompt Pack** *bounded AI support for overwhelmed minds* by **Whelmish** Built around Capacity-Safe design principles. Not therapy. The system adjusts to the human. Personal use only. © Whelmish.