Product · 15 tokens per run
Condolence Card Writer
Staring at a blank line above your signature on a group condolence card? Tell it who is grieving, how well you knew the person who died, one detail if you have one, and the tone. Get back two calibrated card messages side by side (a shorter restrained version and a longer more personal one) plus a short list of tempting phrases NOT to write. Uses the acknowledge-validate-connect etiquette arc, never invents facts, and refuses the phrases grief writers repeatedly call tone-deaf (better place, time heals, everything happens for a reason, let me know if you need anything).
About this tool
What does condolence card writer do?
Staring at a blank line above your signature on a group condolence card? Tell it who is grieving, how well you knew the person who died, one detail if you have one, and the tone. Get back two calibrated card messages side by side (a shorter restrained version and a longer more personal one) plus a short list of tempting phrases NOT to write. Uses the acknowledge-validate-connect etiquette arc, never invents facts, and refuses the phrases grief writers repeatedly call tone-deaf (better place, time heals, everything happens for a reason, let me know if you need anything).
How does Condolence Card Writer work?
Tell it who the card is for (the bereaved and how you know them), who they lost, the name of the person who died if you know it, how well you personally knew that person, one specific detail you have if any, how the loss happened, and the tone you want. One AI pass returns two calibrated card messages side by side: a short restrained version in 1 to 2 sentences (the safe default when you barely knew the person) and a longer more personal version in 3 to 4 sentences (warmer, weaves in your one detail). You also get a one-line explanation of the etiquette arc it used and a short list of 3 to 4 phrases you should NOT write, tailored to this specific loss.
How is this different from asking ChatGPT to write a condolence card?
ChatGPT returns one page-long paragraph in its default voice, often opens with a cliche it knows to avoid, and cheerfully invents shared memories it has no way to know. Condolence Card Writer runs a single structured pass with four hard rules: it never centers the writer (no 'I lost my mom too'), never invents facts you did not give it, honestly acknowledges when you barely knew the person instead of faking closeness, and expands your one supplied detail into exactly one sentence, no more. It returns two length-calibrated versions so you pick the one that fits your relationship, plus a red-flag panel of the exact phrases grief writers keep calling tone-deaf.
What if I barely knew the person who died?
That is the most common case (a coworker's spouse, a friend's parent, a neighbor's sibling) and the tool is built for it. When you mark closeness_to_deceased as 'not at all' or 'never met', it does NOT fake a relationship. Instead the longer version leans on an honest acknowledgement pattern that grief writers recommend: something like 'I never had the chance to meet Evan, but it is clear how much he meant to you.' You end up with something specific and warm, without pretending to memories you do not have.
Will it invent memories or details I did not give it?
No. The only facts it may state are the name of the person who died (if you supplied it), who they were to the bereaved (their spouse, their mother, their child), and the one detail you typed in yourself. If you leave the detail field blank, it does not fill the gap with invented memories. If you paste one specific thing (a running joke, a trait, a line from the obituary that stuck with you), it expands that into exactly one sentence in the longer version and never adds second-order invented detail on top.
What are the phrases it tells me to avoid, and why?
The 'don't write this' panel is tailored to your specific loss and relationship. For a sudden death it prioritizes 'everything happens for a reason' and 'God has a plan'. For a long illness it prioritizes 'at least she is not suffering anymore' and 'she lived a long life'. For a coworker context it always flags the vague open-ended 'let me know if you need anything' and the pressuring 'be strong'. And regardless of context it flags the universal grief-writer red flags: 'they are in a better place', 'time heals all wounds', 'I know how you feel', 'this too shall pass', 'find the better days ahead'. Each phrase comes with one line on why it lands badly for this specific bereavement, so you can recognise and skip it.
Why two versions instead of one?
Because the etiquette length norm scales with your relationship, and one calibrated default would be wrong for half the users. A 4-sentence card from someone who met the deceased once at a work happy hour reads as performance. A 1-sentence line from a close friend reads as cold. Two calibrated versions side by side let you pick without guessing (or re-running and re-spending tokens): short restrained is the safe default for thin relationships, longer personal is for when you actually have a memory to lean on or when you knew the deceased well.
How much does it cost, and do I need a subscription?
There is no subscription. Each run costs 15 produde tokens, a prepaid platform credit you buy once and spend across any produde tool, so you pay per card instead of a monthly fee. Most people write one to three condolence cards a year, so a small token top-up covers a long time. If a run fails, the full token cost is refunded to your wallet automatically.
Who is it for?
The person holding a group condolence card at work and staring at the blank line above their signature. Coworkers signing a card for someone whose parent or spouse died. Managers writing a note to a bereaved report. Neighbors, clients, distant relatives, anyone who genuinely wants to say something that lands but does not know what to write and is terrified of being tone-deaf. It is not for a eulogy or a delivered speech, that is Heartfelt Speech Writer; this is for the short written note you sign and hand over.
What does it cost?
15 tokens per run. 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.