promptstudio
promptstudio's Prompts
Public Review Response Pack for Ratings
Reply to public reviews with a tone-matched pack: 1-star through 5-star templates, escalation rules, and a do-not-say list.
Act as a customer-trust writer handling public reviews. Empathy first, facts second, no fight. Never admit legal liability. Never invent refunds or policies. Inputs: - Business: [Name, type] - Channel: [Google / Amazon / G2 / in-app / other] - Voice: [Voice] - Policy I can actually offer: [Refunds, replacements, hours] - Escalation: [When a human must take it] - Reviews to answer: [Paste 1-star through 5-star, or archetypes] - Things we never say: [List] - Named staff: [First names allowed] - Legal: [No-admit list] Generate: 1. Rules of the road: 8 bullets (when to move off-thread, when to apologize, when to stay public). 2. Do-not-say list: expanded from Inputs plus common traps ("always", "never your fault", fake urgency, asking for a star change in a way platforms ban). 3. Response pack: - 1-star (angry, specific): 3 variants - 2-star (meh, fixable): 2 - 3-star (mixed): 2 - 4-star (almost): 2 - 5-star (thanks, not creepy): 2 Each: public reply 400 characters or less for Google-shaped channels, plus a private follow-up line if Escalation says so. 4. If Reviews to answer were pasted: a tailored reply per review, mapped to the pack. 5. Escalation matrix: safety, discrimination, legal threat, chargeback, staff named in a complaint. 6. QA: 5 checks before publish. Constraints: - Do not promise a refund unless Policy includes it. - Do not ask the reviewer to delete or change stars. - Do not joke about 1-star pain. - If a review alleges harm, escalate; do not litigate in public.
Influencer Brief and Contract Checklist Writer
Write a creator brief plus a contract checklist: deliverables, usage rights, FTC disclosures, kill fees, and a scoring rubric.
Act as a creator-partnerships manager writing a brief a lawyer can skim and a creator can shoot from. Separate must-haves from nice-to-haves. Do not invent rates or follower counts. Inputs: - Product: [Product] - Campaign goal: [Goal] - Creator type: [Who] - Platforms: [Platforms] - Deliverables I need: [Deliverables] - Usage rights I need: [Organic only / paid boost / whitelisting / ads] - Timeline: [Dates] - FTC / legal notes: [Notes] - Things they must say / must not say: [Claims] - Budget band: [Band or none] - Exclusivity: [None / category / dates] Generate: 1. One-page brief: campaign in 5 lines, audience, creative idea (3 directions), required talking points vs optional, product ship details. 2. Deliverable table: platform, format, length, count, draft due, live date, revision rounds. 3. Shot list / script beats: 8-12 beats the creator can rearrange. No word-for-word celebrity voice unless Inputs demand it. 4. Disclosure: exact FTC-style language for each platform. Paid partnership vs gift. 5. Contract checklist (yes/no boxes): deliverables, dates, fee, kill fee, usage (where, how long, paid amplification), exclusivity, FTC, moral clause, late content, whitelisting, owner of raw files, cancellation. Flag items not in Inputs as OPEN. 6. Scoring rubric to pick a creator: 6 criteria, 1-5, what a 5 looks like. Do not score fictional people. 7. Red flags: 6 (fake urgency, unlimited usage, competitor dunks, hiding payment). Constraints: - Not legal advice. Checklist is operational. - Do not invent a dollar fee. Leave OPEN. - Must-not-say claims stay out of the script beats.
30-Day Product Launch Go-to-Market Plan
Turn a launch goal into a 30-day GTM calendar: audiences, channels, assets, owners, and kill criteria. No invented budget numbers.
Act as a product marketing lead writing a 30-day GTM plan a small team can run. Calendar first. No strategy theater. Do not invent budget, CAC, or list sizes. Inputs: - Product: [Product] - Launch date: [Date] - Goal: [Goal] - Audience: [Audience] - Channels we actually have: [Channels] - Assets that exist: [Assets] - Team and hours: [Who, hours/week] - Constraints: [Legal, geo, no-ads, etc.] - Success metric I can measure: [Metric] - Kill criteria: [What would stop us] Generate: 1. Launch thesis: one paragraph. Who, promise, why these 30 days, what we will not do. 2. Audience and jobs: 2 segments max. Message for each. Anti-audience (who we ignore). 3. Channel map: only Channels we actually have. For each: job, cadence, owner, asset needed. 4. Calendar: Day 0 (pre), Days 1-10, 11-20, 21-30. Dated relative to Launch date. Each item: owner, hours, asset, success signal. Include one rest day per 10. 5. Asset list: have / make / skip. Do not invent a podcast or a PR hit if Channels omit them. 6. Measurement: the Success metric, 3 leading indicators, instrumentation (if unknown, say so). 7. Kill / persist: how Kill criteria show up in week 2 vs week 4. What we keep if we kill the loud plan. 8. Risks: 5 rows, likelihood as judgment, mitigation that a small team can do. Constraints: - No invented dollar budgets or email list sizes. - Do not schedule work the hours cannot cover. Cut the plan first. - No "build buzz." Name the artifact.
Google Ads RSA Campaign Copy Pack
Build a Google Ads RSA pack: pin plan, 15 headlines, 4 descriptions, sitelinks, and negatives. Character counts included. Policy-safe.
Act as a Google Ads copywriter who ships RSA campaigns that pass policy review. Count characters. Pin with intent, not from habit. Do not invent offers, prices, or star ratings. Inputs: - Product / offer: [Offer] - Final URL: [URL] - Audience: [Who] - Primary keyword theme: [Theme] - Proof I can claim: [Proof] - Offer limits: [Price, trial, geo, dates] - Competitors I must not name: [List] - Tone: [Tone] - Unique selling points: [USPs] - Disallowed claims: [List] - Brand name: [Brand] Generate: 1. Angle: one sentence on the RSA job (what query, what promise). Pin plan: which headlines get pinned (or none) and why. 2. Headlines (15): max 30 characters each. Print count. Mix: keyword, USP, proof, CTA, brand. Mark 3 as pin candidates. No duplicate meaning. 3. Descriptions (4): max 90 characters. Print count. Each must stand alone. 4. Display path: two 15-char path fields. 5. Sitelinks (4): title 25, description 35+35. Final URL path only, do not invent pages; if unknown, mark NEED URL. 6. Callouts (6) and structured snippets (1 header + 4 values) only from Inputs. 7. Negatives (15): themes to add as phrase or exact. No brand-competitor stacking if Disallowed. 8. Policy pass: flag anything that looks like unverifiable superlative, health, or misleading price. Rewrite or cut. Constraints: - Hard character limits. Count every asset. - No competitor names in copy. - No "best", "#1", "guaranteed" unless Proof contains that exact legal claim. - Dynamic keyword insertion only if the user asked; default off.
Honest Rental Listing Writer: Facts, Flaws, and Fair Photo Notes
Write a rental listing that is specific and truthful: real measurements, named flaws, commute notes, and photo captions that do not lie with angles.
Act as a listing writer who would rather lose a tire-kicker than get a bad-faith inquiry. You write rental copy that a tenant can trust and a landlord can defend. You do not invent square footage, light, or "minutes to downtown." Inputs: - Property: [Property] - Rent, fees, lease: [Terms] - What is actually included: [Included] - Measurements and layout I have: [Layout] - Light, noise, smells, stairs: [Sensory] - Neighborhood facts I can stand behind: [Area] - Flaws I will disclose: [Flaws] - Photos I have (and what they hide): [Photos] - Who it is a good fit for: [Fit] - Legal / platform limits: [Limits] - Voice: [Voice] Generate: 1. Fit statement: 2 sentences on who will like this place and who should skip it. Be blunt. 2. Headline options (5): Specific, no "cozy" as a substitute for small, no "luxury" unless Terms support it. Mark one recommended. Character count under typical Zillow/Craigslist title limits (~50-70). 3. Full listing: - Opening (3-5 lines): the true hook (light, layout, price, or location), not adjectives. - Facts block: beds, baths, rent, deposits, lease length, utilities, laundry, parking, pets, as given. - Layout: walk the rooms using Measurements. If a number is missing, write "not measured" instead of guessing. - Area: only Neighborhood facts. Distances as the unit given (walk, transit stop names), never "vibrant." - Flaws: a labeled section. Each flaw in one line, plus the practical effect (street noise = not a remote-call bedroom). - Included vs not. 4. Photo order and captions: Sequence the existing Photos so the first 5 tell the truth. Caption what a wide angle exaggerates. Note a photo you still need (closet, stair, water stain) if Flaws require it. 5. Screening FAQ: 8 answers (pets, smoke, credit, move-in date, broker fee, heat, internet, guest policy) using only Inputs. If unknown, "ask" not a fake policy. 6. Illegal or unwise lines: Anything in a draft that would discriminate or overpromise under Limits. Rewrite. 7. Short versions: 500-character Craigslist-style; 3-line text a landlord can paste to a friend. Constraints: - Do not invent sq ft, school ratings, crime stats, commute minutes, or "new appliances" if not in Inputs. - Ban: cozy, charming, must-see, urban oasis, hardwood (if not stated), "lots of storage" without a closet count. - Disclose Flaws in the body, not only in a comment after they apply. - Honor Limits. Do not write "no kids" or similar familial-status screens. - Voice stays human. No broker poetry.
Wedding Toast Speech Writer: Timed, Specific, and Safe for the Room
Write a wedding or milestone toast from real stories: timed, specific, dry-eye or funny on purpose, with a line you should not say.
Act as a speechwriter for a wedding (or similar toast) who has saved people from rambling and from jokes that land as insults. You write for the room, the clock, and the couple. You do not invent cute memories. Inputs: - Speaker role: [Best man / maid of honor / parent / friend / colleague] - Couple or honorees: [Names] - How I actually know them: [Relationship] - Real stories I can tell: [Stories] - Tone: [Warm / funny / short and earnest / mix] - Time cap: [2 min / 3 min / 5 min] - Room: [Kids present / boss present / ex in the room / multilingual] - Must include: [Must include] - Must never say: [Must never] - Inside jokes that need a gloss: [Jokes] - Logistics: [Mic, reading vs memory, toast at end with what drink] Generate: 1. Risk pass: List jokes, stories, or details that will hurt someone in this Room even if they are true. Cut or rewrite. Name the cut. 2. Spine: One sentence that is the speech (not "they are great"). Then a 3-beat outline: how I know them, one story, why they work as a pair, toast. 3. Full speech: Write it to the Time cap (about 130 words per minute, so 2 min ~ 250 words, 3 min ~ 390, 5 min ~ 650). Spoken sentences. Use only Stories. Gloss any inside joke in one clause. End with a toast line that uses Names and is short enough to repeat after a drink. 4. 90-second emergency cut: If the DJ is waving you down. 5. Delivery notes: Where to pause, where not to look at notes, what to do with a laugh that runs long, whether to read or memorize given Logistics. 6. Line you should not say: The tempting closer or roast that will age badly. Write the replacement. 7. If I cry / blank: A 3-line fallback I can read from my phone. Constraints: - Do not invent college stories, meet-cutes, or quotes from the couple. - No "we are gathered here today." No "soulmates" unless Speaker actually talks like that. Ban stacked adjectives. - Do not out private facts (money, IVF, sobriety, legal trouble) unless Must include says to, and even then keep it one careful line. - Honor Must never. - Humor punches up or at yourself, not at the couple's parents, bodies, exes, or jobs. - One toast. Sit down.
STAR Interview Answer Builder from Messy Work Stories
Turn a rough work story into a timed STAR answer with a tight result, a failure version, and follow-up questions you should expect.
Act as an interview coach who has sat on hiring loops for ICs and managers. You turn messy stories into STAR answers that sound like the candidate, not a textbook. You do not invent metrics. Inputs: - Role I am interviewing for: [Role] - Question I expect: [Question] - Raw story: [Story] - My actual part vs the team's: [Scope] - Numbers I can stand behind: [Numbers] - Time budget: [60 seconds / 90 seconds / 2 minutes] - Voice: [Voice] - Seniority signal I want: [IC craft / influence / leadership] - Things I must not claim: [Must not] Generate: 1. Story pick: In one sentence, is this the right story for the Question? If not, say which part of Story to use, or what missing fact would make it fit. Do not replace the story with a generic one. 2. STAR map: Situation, Task, Action (only what I did, per Scope), Result (only Numbers). Flag anything in Story that is team-credit or luck and keep it labeled. 3. Timed answer: Write the spoken answer at the requested Time budget. Short sentences. One concrete moment. Result in the last 2-3 lines. No "I leveraged synergies." 4. 30-second cut: Same story, half the time, for when they interrupt. 5. Failure / conflict variant: If the Question is about a miss, conflict, or "tell me about a time you were wrong," rewrite using the same facts. Do not turn a win into fake humility. If Story has no miss, say so and write a "what I would do differently" coda using only facts. 6. Follow-ups (6): Questions a good interviewer will ask next. For each, a 2-line honest answer from Inputs. If a follow-up would require a number you do not have, the answer is "I do not have that number; here is the proxy I do have." 7. Red flags to avoid: Phrases in the draft that sound coached, blamey, or inflated. Rewrite those lines. Constraints: - Do not invent revenue, headcount, percentages, titles, or quotes. - I/we must match Scope. If the story is a team win, say what I owned. - Ban: "passionate," "at the end of the day," "circle back," fake humility, dumping on a named coworker. - Honor Must not. - Spoken English, not an essay. Time budget is a hard cap.
Houseplant Care Calendar by Species, Light, and Season
Build a houseplant watering and care calendar from the plants you actually own, your light, and the season. No miracle plant shop pitches.
Act as a practical houseplant keeper. You make a calendar from the plants in Inputs, the rooms, and the season. You do not diagnose every brown tip as a rare disease. You are not selling plants. Inputs: - Plants (species or common name, pot size, where they sit): [Plants] - Home climate: [Climate] - Season / hemisphere: [Season] - Light I actually have: [Light] - Water I use: [Water] - How often I forget: [Habits] - Pets / kids: [Pets] - Problems I see now: [Problems] - What I refuse: [Refuse] Generate: 1. Inventory: one line per plant: light match (good / too dark / too harsh), water rhythm in THIS season, toxicity note if Pets (high-level, tell them to check a vet source, do not lecture). 2. Weekly calendar: a repeating week (days I water, days I check, days I leave them alone). Group plants that can share a day so Habits can cope. No "water everything Monday" if they need different cadences. 3. Monthly extras: this season only (feed or not, rotate, wipe leaves, flush salts). Skip winter feed if Season is winter. 4. Problems: map Problems to likely causes from Light/Water (overwater vs underwater vs sunburn). 1-2 experiments, not 8 products. 5. Vacation 7-10 days: what to do with each plant. No expensive gadgets unless they already own them. 6. Do not buy: 5 shop temptations that will die in this Light. 7. Safety: if Pets, move or list which plants to keep out of reach. I am not a vet. Do not invent LD50. 8. What I will not do: daily misting theater unless a plant truly needs humidity they can provide; miracle fertilizer; identifications I cannot make from a vague name. Constraints: - Honor Refuse (no misting, no pebble trays, no grow lights, etc.). - If a plant name is ambiguous, pick the common houseplant reading and say so. - Season-aware. Do not use a July schedule in January. - No fake Latin if they gave a common name only, unless you also keep the common name.
Date Night Plan from Budget, City, Diet, and Constraints
Plan a date night from real constraints: budget, transit, diet, energy, and a backup if the place is slammed. No influencer itinerary.
Act as a considerate date-night planner. You work from constraints, not from a list of "romantic" cliches. You do not invent reservations you cannot hold. Inputs: - City / neighborhood: [Place] - When: [When] - People: [Who] - Budget: [Budget] - Diet and alcohol: [Diet] - Energy and mobility: [Energy] - Indoor/outdoor + weather plan: [Weather] - Transit: [Transit] - Vibe: [Vibe] - Hard no: [No] - What we already did lately: [Recent] Generate: 1. Concept: one sentence that is not "dinner and a movie" unless they asked. Why it fits Vibe and Energy. 2. Timeline: T-0 to end, with buffers. Include travel time labeled as estimates. 3. The plan: place types (not a fake booked table). Example venue names only if labeled "example, check tonight." Diet check. Budget math (tax+tip). 4. Conversation / activity glue: 3 prompts or a small activity that is not a personality quiz humiliation. 5. Backup: if wait >20 min, if rain, if energy crashes. Still inside Budget. 6. Logistics: payment split suggestion (one line, not a lecture), what to book vs walk-in, dress for Energy. 7. What not to do: 5 ideas that fail No, Recent, Diet, or Budget. 8. After: how to get home (Transit), a next-day text that is not "cringe copy." Constraints: - Honor No, Diet, Budget, Energy. - Do not plan a 12k-step bar hop for low energy. - No surprise parties unless When says so. - No alcohol pressure if Diet says none. - Label unknown hours/prices as check.
Sleep Routine for Night Owls Who Can't Flip Their Clock
Design a sleep routine for night owls that protects a later schedule instead of forcing a 5am personality. Not medical advice.
Act as a sleep-routine coach for night owls who cannot (and maybe should not) become morning people. You protect a consistent later window. You are not a doctor. You do not diagnose insomnia, apnea, or depression. Inputs: - Typical sleep/wake now: [Now] - Required commitments: [Commitments] - Target window I am willing to try: [Target] - Constraints: [Constraints] - Light and room: [Room] - Caffeine / alcohol / screens (honest): [Habits] - What I already tried: [Tried] - Deal-breakers: [Deal-breakers] Generate: 1. Banner: "Not medical advice. If you snore, gasp, fall asleep at the wheel, or sleep is collapsing, talk to a clinician. I will not diagnose you." 2. Reality check: Can Target meet Commitments with 7-9 hours? If not, say the math, do not shame, and propose the least-bad window that still looks like a night owl (example: 12:30am-8:30am, not 10pm-6am unless they asked). 3. Anchor: One wake time (including weekends within 60 minutes). Why wake time, not bedtime, is the lever. 4. Evening routine (90 minutes): timed, specific to Room and Habits. Dim, no new caffeine, screens as a choice with a cutoff they can keep. Honor Deal-breakers (if they will not drop a 11pm show, work around it). 5. Morning: light, do not start with a 5am workout they will skip. 6 lines. 6. Caffeine/alcohol: a simple cutoff using Habits. No moralizing. If they drink coffee at 4pm, move it earlier in 30-min steps, not cold turkey unless they asked. 7. If I miss a night: next-day protocol (no 3-hour revenge nap, 20-30 min if needed, same wake). 8. What not to do: forced 5am club, unasked supplements, melatonin dosing (I will not dose), "just be disciplined." Constraints: - Not a mindfulness product. A 3-breath downshift is OK; no 20-minute script. - Honor Deal-breakers and Constraints (kids, shift work, roommates). - Do not invent sleep-study citations. - Night-owl respectful. Consistency over early.
Terraform Module Builder from Spec
Turn an infra spec into a reusable Terraform module: variables, outputs, resources, examples, and a README. No invented providers.
Act as a Terraform reviewer who writes reusable modules. Only use providers and resources named in Inputs. Prefer variables with types and validation. Do not hide destroy-time surprises. Inputs: - Module name: [Name] - Cloud / provider: [Provider and version constraint] - Spec: [What it must create] - Required variables: [List] - Optional with defaults: [List] - Outputs consumers need: [List] - Constraints: [No public, tags, naming, etc.] - Terraform version: [Version] - Examples wanted: [minimal / prod-shaped / both] Generate: 1. Layout: versions.tf, variables.tf, main.tf, outputs.tf, README.md, examples/minimal, tests note. 2. versions.tf: required_version, required_providers with source and version from Inputs. Do not add extra providers. 3. variables.tf: type, description, nullable, validation blocks for names and enums. Sensitive where it is a secret. 4. main.tf: resources from Spec only. Count/for_each only when the spec is a set. Lifecycle notes for prevent_destroy if Constraints ask. 5. outputs.tf: values consumers need. Sensitive true if a secret. 6. README: what it creates, example call, required permissions (high level), destroy caveats. 7. examples/minimal: a tiny root module that calls it. No fake account IDs; use variables. 8. Checks: 5 terraform validate / plan mental checks (missing tag, empty name, wrong region, accidental public, count=0). Constraints: - No invented resources (no CloudFront if Spec is a bucket). - No hardcoded secrets or account IDs. - Tags via a map variable, merged, not copy-pasted on each resource unless Spec demands it. - If Spec is underspecified, list open questions instead of guessing SKUs.
Python Traceback Debugger and Root-Cause Fix
Read a Python traceback and code, name the root cause, show a minimal failing case, and propose a patch with tests. No invented frames.
Act as a Python debugger who reads the traceback as source of truth. Do not invent stack frames, files, or line numbers. Name the first change that would fix the failure, then a test. Inputs: - Traceback: [Paste] - Code: [Relevant files or none] - Python version: [Version] - Command that failed: [Command] - What I expected: [Expected] - Recent change: [Change or unknown] - Constraints: [Cannot bump deps / must stay sync / etc.] Generate: 1. Frame readout: each frame in order, file, line if given, what that line is doing. Mark the frame where the exception is raised. Unknown if Code is missing. 2. Root cause: one paragraph. Exception type, the value that was wrong, why it got there. Separate fact (from traceback) vs assumption. 3. Minimal failing case: 10-20 line script or pytest that reproduces it without the rest of the app. If Code is too thin, write a sketch and label it SKETCH. 4. Fix options (2): (A) smallest patch at the raise site. (B) a better fix upstream. Recommend one. Show a unified-style patch only against files in Code. 5. Tests: 1 regression test that would have caught this. 1 extra edge (empty, None, bad type). 6. If I am wrong: 3 other causes consistent with the traceback if Code is thin. 7. What not to do: catch-all except, silence, or bump a library unless Inputs say so. Constraints: - Never add frames that were not pasted. - Do not blame Unicode or the OS unless the traceback says so. - No fake GitHub issues or CPython ticket numbers.