Poetry & Verse Prompt is a Claude AI skill — form-aware, occasion-calibrated verse generated from a three-question interview, with live research into what's resonating now.
You asked Claude to write a poem for your colleague's retirement. You got eight lines about "journeys" and "chapters" that end with something rhyming "day" and "way." It's not offensive. It's not memorable. It reads like it was written for anyone who has ever retired, which is to say it was written for no one in particular.
That's the specific failure of vanilla AI verse: it reaches for universality and lands on blandness. The rhymes arrive on schedule. The sentiment is broadly appropriate. And the person reading it can tell, immediately and viscerally, that no one thought about them when they wrote it.
The fix isn't asking Claude to "try harder" or "be more creative." It's giving the skill the context it needs before it writes a single word — and that's precisely what the Poetry & Verse Prompt skill does first.
What Generic Verse Actually Sounds Like
When you ask Claude to write a poem without any calibration, it draws on the statistical centre of poetry — the forms, rhythms, and images that appear most often across the training data. That means iambic-adjacent metre, ABAB or ABCB rhyme schemes, and metaphors built around light, paths, seasons, and time. The result is technically verse. It's just not verse for this.
The problem compounds with occasion. A wedding poem, a birthday poem, a poem for a child's first day of school — Claude treats these as variations on a single template rather than fundamentally different emotional tasks. A wedding poem needs to carry weight, because it's read aloud in front of everyone the couple loves. A child's first-day poem needs to be light and brave and honest about the morning being hard. These aren't the same poem. But without calibration, they often come out that way.
Generic AI poetry fails not because Claude lacks vocabulary or rhythm — it fails because it has no information about the specific person, occasion, or emotional register that would make the poem land.
The person who asked for help wanted something that felt personal. They got something that felt produced. That gap is what the skill closes.
Why Context Changes Everything in Verse
Poetry is a compression format. Every line has to carry more weight than prose because there are fewer of them. That means every choice — the form, the tone, the images, whether it rhymes or doesn't — needs to be right for this recipient, this occasion, this emotional note. A sonnet about grief lands differently than a free-verse piece. Both can be right. Neither is automatically correct.
The Poetry & Verse Prompt skill researches current form conventions and what kinds of verse are resonating for different occasions before it starts drafting. It isn't pulling from static training patterns. It's working with live signals about what's actually moving people — which wedding poems are being shared, which structures feel fresh versus tired, which imagery has been overused to the point of meaninglessness.
The difference between a poem that makes someone cry and one they politely thank you for is almost never the vocabulary — it's whether the form and register were chosen for them.
Then comes the three-question interview. Before a single line is drafted, the skill asks what it needs to know: who this is for, what the occasion is, and what emotional note it should carry. Is this funny or tender? Formal or intimate? Should it rhyme, or would rhyme make it feel forced for this particular person? The answers to those questions change the output more than any single writing instruction could.
What the Poetry & Verse Prompt Skill Actually Does
The skill is structured around one core principle: earn the right to write by understanding the situation first. Here's how that unfolds in practice.
Generic Output vs Calibrated Output
The gap between a generic AI poem and a calibrated one is easier to see than to describe. Here's the same request — a poem for a colleague leaving after ten years — handled both ways.
A journey filled with light and praise.
You've walked the path with grace and care,
And left your mark beyond compare.
As you move on to pastures new,
We'll cherish all the things you do.
Though time moves on and roads may part,
You'll always hold a special place in our heart.
which meetings end early, whose birthday cake
is worth staying late for. You became
the kind of colleague people mean
when they say they miss the place.
Take that with you. It's not small.
The next room gets the version of you
we made together.
The left version has all the marks of uncalibrated verse: abstract nouns, forced rhyme on a three-syllable stretch ("beyond compare"), and imagery that could apply to any departure, anywhere. The right version knows this is a workplace, knows the tone should be warm and a little dry, and earns its emotional landing through specificity — the good mugs, the birthday cake — rather than through stated sentiment. That shift comes entirely from the calibration questions the skill asks before it starts.
Who Gets the Most from Poetry & Verse Prompt
Creators and writers who need verse for real occasions — weddings, eulogies, celebrations, cards — and can't afford output that reads like it came from a template. Also: anyone who's been asked to "say a few words" in verse and felt completely out of their depth.
The primary user is someone who cares about getting this right but doesn't write poetry habitually. A best man who wants the speech to include something moving. A parent writing in their child's birthday book. A founder who wants the brand launch announcement to carry genuine weight rather than corporate warmth. These are people for whom the poem matters, but who don't have the form knowledge to know whether free verse or a structured stanza would serve the moment better.
The skill also works well for creators who produce verse regularly and want a structured starting point — a form recommendation, a register calibration, a quality check — before they do the final craft work themselves. It's not a replacement for a poet's judgment. It's the research and structure layer that frees them to focus on the writing.
The Output You Walk Away With
The skill produces multiple verse variants — typically two to three — each in a different form, with a brief note on what each form communicates and why it might be right for this occasion. A free-verse version that prioritises emotional specificity. A structured version that uses form to create a sense of ceremony. Sometimes a hybrid that rhymes selectively rather than relentlessly. You see the options and understand the trade-offs, so you can choose — or mix — with confidence.
From there, the path is direct: read the variants, pick the one that fits, make any personal adjustments, and use it. For most occasions, you're done in under ten minutes. For higher-stakes moments — a eulogy, a wedding reading — you have a strong structural foundation to work from, which is often the hardest part.
There's a reason the best poems at weddings and funerals and retirements make people cry. It's not because they're technically sophisticated. It's because they feel like they were written for that room, that person, that exact moment. That's not something generic AI can produce from a one-line instruction. It's something that requires knowing the context first — and then building every line from there.