Product
Caregiver Sibling Update
For the adult child who is the primary caregiver for an aging parent and dreads writing the weekly update to siblings who live elsewhere. You are the one who took Mom to the specialist Tuesday, called the pharmacy about the med change Wednesday, sat with Dad through a bad sleep night Friday, and now on Sunday you are staring at a blank group text or a blank email draft with 200 half-remembered facts and one very tired resentment. Paste your rough dictated notes plus the sibling roster and get back a calm, factual six-section weekly update: this-week-at-a-glance, health and function, meds and medical care, daily life and safety, what we are watching next week, and requests and decisions. Every bullet anchored line by line to something you actually noted, with a used_notes and unused_notes receipt per section so you can see what was kept and what was dropped. Never invents medical facts, never interprets symptoms, never names a diagnosis, never invents siblings or asks. Strips the resentment leakage every caregiver-forum flags ('as usual I was the one who had to', 'nobody bothered to call', 'again I did it alone') into neutral factual phrasing, and boundaries the diary spiral ('I am so tired, I cannot do this') into one clear time-bound ask under Requests and decisions. The bluntness dial lets you set the register from soft (family-facing gentle, still names the safety events) through direct (neutral factual) to very-direct (names decline starkly, still no diagnostic labels). Deep tier adds full per-sibling calibration keyed to what you pasted about each sibling (lives-out-of-state, has-kids, works-nights, money-one, medical-one), full cross-section anti-repetition so no two bullets in the update share an opener stem or a closing clause, and full ai-tells strip of the caregiver-update cliche lexicon ('just wanted to update everyone', 'thanks in advance for your understanding', 'hanging in there', 'I am doing my best', 'let me know if you can help').
About this tool
What does caregiver sibling update do?
For the adult child who is the primary caregiver for an aging parent and dreads writing the weekly update to siblings who live elsewhere. You are the one who took Mom to the specialist Tuesday, called the pharmacy about the med change Wednesday, sat with Dad through a bad sleep night Friday, and now on Sunday you are staring at a blank group text or a blank email draft with 200 half-remembered facts and one very tired resentment. Paste your rough dictated notes plus the sibling roster and get back a calm, factual six-section weekly update: this-week-at-a-glance, health and function, meds and medical care, daily life and safety, what we are watching next week, and requests and decisions. Every bullet anchored line by line to something you actually noted, with a used_notes and unused_notes receipt per section so you can see what was kept and what was dropped. Never invents medical facts, never interprets symptoms, never names a diagnosis, never invents siblings or asks. Strips the resentment leakage every caregiver-forum flags ('as usual I was the one who had to', 'nobody bothered to call', 'again I did it alone') into neutral factual phrasing, and boundaries the diary spiral ('I am so tired, I cannot do this') into one clear time-bound ask under Requests and decisions. The bluntness dial lets you set the register from soft (family-facing gentle, still names the safety events) through direct (neutral factual) to very-direct (names decline starkly, still no diagnostic labels). Deep tier adds full per-sibling calibration keyed to what you pasted about each sibling (lives-out-of-state, has-kids, works-nights, money-one, medical-one), full cross-section anti-repetition so no two bullets in the update share an opener stem or a closing clause, and full ai-tells strip of the caregiver-update cliche lexicon ('just wanted to update everyone', 'thanks in advance for your understanding', 'hanging in there', 'I am doing my best', 'let me know if you can help').
How does Caregiver Sibling Update work?
You paste seven inputs: your rough dictated notes from the week caring for your aging parent, the parent's first name, the caregiver relationship label (daughter, son, in-law), the list of siblings you are writing to with any one-line context for each (lives out of state, works nights, has kids at home, is the medical one, is the money one), a bluntness dial (soft, direct, very-direct), any topics to keep out of this week's update, and the tier. One structured pass returns a six-section weekly update: This week at a glance, Health and function, Meds and medical care, Daily life and safety, What we are watching next week, Requests and decisions. Each section stays grounded in the notes you pasted, and receipts show used_notes, unused_notes, and every safety_flag that fired during rewrite.
How is this different from ChatGPT or a caregiver notes app?
A freeform ChatGPT prompt will happily infer that mom has early dementia because you mentioned she forgot the stove twice, invent a sibling ask you never wrote, and leave your caregiver frustration bleeding into the update where your siblings can read it. A generic caregiver notes app stores your dictation but does not write anything back for the family group text. Caregiver Sibling Update runs one structured pass with three wedge behaviors: it refuses to interpret symptoms or name conditions, it neutralizes the resentment phrasing that leaks into caregiver dictation, and it turns each sibling's one-line context into a specific coordination ask. Every rewrite is receipted in safety_flags so you can audit exactly what changed before you paste.
Will it invent medical facts, siblings, or asks I did not paste?
No. The no-invention guard refuses to add a medication, a dosage, a vital, a fall, an appointment date, a diagnosis, or a family circumstance that does not trace back to your pasted notes. The medical-inference refusal is stricter still: the tool will not interpret symptoms into conditions (it will not say 'this sounds like early dementia' or 'that could be a UTI' or 'she may have had a small stroke' or 'she seems depressed'), it will not name a clinical condition your notes did not name, and it will not suggest treatments or medication changes. Each medical inference it catches itself reaching for is refused and logged in safety_flags with kind medical_inference_refused. Same rule holds for siblings: if you did not paste context for a sibling, the tool will not fabricate an ask for them and logs sibling_detail_refused instead.
Will it write anything my siblings shouldn't read?
The tool assumes your siblings will read every word of the update, so it neutralizes the two leaks that most often torpedo family group texts. First, the resentment-leak neutralization: phrases like 'nobody else has helped this month', 'I am doing this alone', 'as usual it fell on me' get rewritten into observable factual phrasing about what actually happened this week, and each rewrite is receipted in safety_flags with kind resentment_rewritten. Second, the diary-boundary rewrite: private caregiver processing that belongs in a journal (your own exhaustion, your resentment of a specific sibling, financial fear framed as blame) gets moved out of the shared update, receipted as diary_boundaried. The bluntness dial (soft, direct, very-direct) lets you set how frank the health and requests sections read without ever letting the tone tip into blame.
How does the per-sibling asks block work?
The Requests and decisions section closes with one concrete, time-bound ask per sibling, calibrated to what you actually pasted about each of them. If you noted that one sibling lives out of state, the ask is remote-suitable (a scheduled call with mom on Tuesday evening, a portal login to review the labs). If a sibling has young kids at home, the ask is small and off-hours. If a sibling is the medical one in the family, the ask leans into medication review or a call with the primary care doctor. If a sibling is the money one, the ask leans into a specific bill or benefit form. Every ask is grounded in your used_notes for that sibling. If you did not paste enough context for a sibling to calibrate an ask honestly, the tool refuses to fabricate one and logs sibling_detail_refused instead of inventing.
Which tier should I pick, Quick or Deep?
Quick (15 tokens, Gemini 3.1 Flash Lite, up to 2 siblings) is the fast defensible pass: all six canonical sections, the no-invention and medical-inference guards, the resentment-leak neutralization, the diary-boundary rewrite, and best-effort anti-repetition across siblings. It is the right pick when you are writing to one or two siblings and want to move quickly on a Sunday night. Deep (40 tokens, Gemini 3.5 Flash, up to 4 siblings) adds the full cross-sibling anti-repetition so no two per-sibling asks share phrasing, full tone calibration to each sibling's one-line context, and the full safety_flag receipt set (medical_inference_refused, resentment_rewritten, diary_boundaried, sibling_detail_refused). Pick Deep when you are writing to three or four siblings and the update needs to sit right in a group text where everyone will compare notes.
How much does it cost, and do I need a subscription?
No subscription. You spend produde tokens per update: 15 tokens for a Quick run, 40 tokens for a Deep run. Tokens are a prepaid platform credit you buy once on the buy page and spend across any produde tool, so a caregiver pays per weekly update instead of a monthly SaaS bill that keeps charging during hospice weeks when you cannot bring yourself to write anything at all. If a run fails at any stage, the full token cost is refunded to your wallet automatically, so a partial pipeline error never costs you tokens on an update you did not actually receive.
What if my week's notes are just a few bullets or a voice memo transcript?
Thin notes are fine and expected. Most caregivers dictate three bullets in a car park and hit send. The tool refuses to pad: if a section has no grounding in your notes (say, you wrote nothing about meds this week), that section stays short and honest instead of getting filled with generic caregiver-blog prose, and the thin-notes case is surfaced in safety_flags so you can decide whether to add a line before pasting. Voice memo transcripts with filler words, false starts, and out-of-order thoughts work: the structured pass reorders your material into the six canonical sections without inventing anything the transcript did not contain, and unused fragments come back to you in unused_notes so nothing you dictated gets silently dropped.
What does it cost?
From 15 tokens. Pick a tier on the form above. Produde tokens are a prepaid platform credit; buy them once and spend on any tool.
Do I pay if a run doesn't complete?
No. You only pay for results. If a run can't finish, its full token cost stays in your wallet automatically.
Is my data stored?
Your input and the run output are saved to your account so you can browse history. They are not sold, shared with advertisers, or used to train shared models. See privacy for details.