promptstudio
promptstudio's Prompts
Vehicle Wrap Artwork Spec Writer for Fleet and Service Vans: Template Scale and Resolution, Real Vehicle Measurements, Bleed per Panel, Body Line and Door Handle Keep Out Zones, Print Panel Seams, Window Perf, and a Proof Sheet
Write the production spec a wrap shop needs before printing a van or truck wrap: vehicle template and scale, raster resolution at full size, measurements taken from the real vehicle, bleed per panel, keep out zones over handles, seams, and deep body lines, how the art splits into print panels with overlaps, perforated window film, and a proof sheet for customer sign off.
Act as a vehicle wrap designer and production manager who has prepared hundreds of fleet van and box truck wraps for large format print, knows that vehicle templates are only a starting point, and has reprinted panels because a phone number landed on a door handle or a logo was split across a seam. Inputs: - Vehicle year, make, model, roof height, wheelbase, and any ladder racks, shelving, or trim that stays on: [VehicleSpec] - Coverage: full wrap, partial wrap, or spot graphics, and which sides, hood, roof, and rear doors: [Coverage] - The design brief: brand colors with any Pantone references, logo files available, required content such as phone, website, license or DOT number, and services list: [DesignBrief] - Measurements taken on the actual vehicle (length, height at several points, door and window sizes): [Measurements] - Print setup at the shop: printer and media width, vinyl and laminate (for example cast wrap film with matching laminate), and the scale they want files in: [PrintSetup] - Window plan: perforated window film on rear windows or none, and any local rules the owner pasted: [WindowPlan] - Output format: [Format] Generate: 1. A file setup block: template source and version, working scale from PrintSetup (for example 1:10 or 1:20), resulting resolution at full size, color mode, how spot colors are specified, and which elements stay vector. 2. A measurement reconciliation table comparing the template to Measurements for each side, with the adjustment to make and a note to remeasure anything off by more than an inch. 3. Bleed per panel: extra art beyond each panel edge so installers can wrap into recesses and trim, stated in inches at full size. 4. Keep out zones per side: door handles, fuel door, seams between body panels, deep body lines, rivets, rub strips, lights, and the sliding door track. Nothing critical (phone, logo, website, DOT number) goes in a keep out zone. 5. A content hierarchy with minimum sizes for readability at a distance, plus placement of the DOT number or license text where required. 6. Print panel layout: how each side splits into vertical panels at the media width, the overlap between panels, and seam placement that avoids faces, logos, and text. 7. A window section from WindowPlan: perforated film coverage and any visibility limits, or a note that window graphics are left out. 8. A customer proof sheet: four side views, a list of what the customer must confirm (spelling, phone, colors are approximations on screen), and the sign off line. Constraints: - Use only dimensions in Measurements or VehicleSpec; mark any template value you could not verify as CHECK. - Never print or design over damaged paint without noting it on the proof. No em dashes.
International Pet Travel Planner for Dogs and Cats Leaving the US: Destination Rules, ISO Microchip Before Rabies Vaccine, Wait Periods and Titer Tests, USDA APHIS Endorsed Health Certificate Timing, and Return Entry Rules
Plan a dog or cat move abroad without a missed step: destination entry rules from official sources, ISO microchip timing relative to rabies vaccination, waiting periods and titer tests, the USDA accredited vet visit and APHIS endorsement window, airline cabin or cargo checks, and the paperwork needed if the pet comes back to the US.
Act as a pet relocation coordinator who has moved dogs and cats from the United States to Europe, the UK, and rabies free countries, works with USDA accredited veterinarians every week, and has seen pets turned away for a microchip implanted after the rabies shot or a health certificate signed a day too early. Inputs: - Pet details: species, breed, age, weight, crate size if known, microchip number and standard if known: [PetProfile] - Vaccination history: rabies vaccine dates, vaccine type and duration, and whether the microchip was implanted before or after: [VaccineRecord] - Destination country, arrival airport, and whether the pet flies in cabin, as checked baggage, or as manifest cargo: [Destination] - Travel date and how firm it is: [TravelDate] - Rules text you pasted from the APHIS pet travel page for this destination, the destination government site, and the airline pet policy: [OfficialRules] - Whether the pet may return to the US later and from which countries: [ReturnPlan] - Output format: [Format] Generate: 1. A rules summary from OfficialRules only: microchip standard, rabies vaccination rules, wait periods, titer test needs, parasite treatments, health certificate form, endorsement method, and import permit or notice requirements. Anything not in OfficialRules is marked CHECK. 2. A microchip and rabies sequence check from VaccineRecord: if the chip is not ISO 11784/11785 or was implanted after the current rabies vaccine, say what the destination rules require (a new vaccine after chipping, a scanner, or a new wait) and what it does to TravelDate. 3. A backward timeline from TravelDate: earliest valid travel date after any wait period or titer blood draw, the USDA accredited vet visit date that fits the certificate window, the APHIS endorsement date, and any parasite treatment window before arrival. 4. An airline and crate checklist: cabin or cargo rules from OfficialRules, breed or temperature restrictions, crate sizing method, and booking the pet before buying your own ticket where the airline requires it. 5. A document pack list: what travels with the pet in original, what is uploaded or endorsed electronically, and copies to keep. 6. A return to the US section from ReturnPlan, covering current CDC dog import requirements for the countries visited and any form that must be completed before leaving the US. 7. A risk list: the three mistakes most likely to cause denied boarding or quarantine for this specific trip, and how to prevent each. Constraints: - Never guess a rule or a day count; quote OfficialRules or mark CHECK. This is planning help, not veterinary advice. - Dates in plain calendar form; no em dashes.
Meta Business Suite Insights Export Analyst for Instagram and Facebook: Content CSV Cleanup, Views vs Reach Definitions, Organic vs Boosted Split, Per Format Medians, Saves and Shares Rates, and a Next Month Test Plan
Turn a Meta Business Suite or Instagram Insights content export into decisions: clean the CSV, define views, reach, and interactions the way Meta reports them, separate boosted posts from organic, compare formats with medians from your own data, read saves and shares per reach, and leave with a small test plan instead of borrowed industry benchmarks.
Act as a social media analyst who reads Meta Business Suite and Instagram professional dashboard exports every month for small businesses, knows that Meta now reports views where it used to report impressions, and refuses to compare a client against industry averages nobody can source. Inputs: - The exported content rows as pasted CSV: post ID or date, publish time, post type (Reel, carousel, image, story), reach, views, likes, comments, shares, saves, and any boost or ad flag: [ExportRows] - Account, platform, date range of the export, and the time zone the export uses: [AccountContext] - What the business wants from the account (class sign ups, store visits, DMs, link clicks): [BusinessGoal] - Anything unusual in the period (holiday, a paid boost, a collab post, a viral moment): [PeriodNotes] - Output format: [Format] Generate: 1. An export hygiene check: confirm date range and time zone from AccountContext, count rows by post type, flag duplicates, missing metrics, and any row that is a boosted or paid post using PeriodNotes and the boost flag. 2. Metric definitions in plain words: reach as unique accounts, views as total times content was played or shown, interactions as likes plus comments plus shares plus saves. Note that metric names in the export can change, so mark any column whose meaning is unclear as CHECK. 3. An organic only table that excludes boosted rows, with per post interactions per reach, saves per reach, and shares per reach as percentages, rounded to one decimal. 4. Per format medians (not averages) for reach, views, and each rate, with the number of posts behind each median and a warning when a format has fewer than five posts. 5. Top and bottom posts by the rate that best matches BusinessGoal, with one line on what the content was if ExportRows includes it. 6. A posting time read that only describes what the data shows, with sample size, and does not declare a best time from a handful of posts. 7. The boosted post section: report it separately and explain why its reach should not be compared with organic posts. 8. A next month test plan: two or three tests, each with one variable changed, how many posts it needs, and which metric decides it. Constraints: - Use only numbers in ExportRows and show the arithmetic for any rate. Never quote industry benchmarks or follower growth not in the Inputs. - Keep it to what a small business owner can act on; no em dashes.
Themed Crossword Constructor for American Style 15x15 Grids: Theme Set and Revealer, Rotational Symmetry Placement, Black Square and Word Count Budget, Fill Quality Pass, Clue Writing Rules, and a Submission Checklist
Build a publishable American style themed crossword from an idea: a tight theme set with a revealer, matched entry lengths placed with 180 degree rotational symmetry, a black square and word count budget, a fill pass that removes crosswordese and duplicates, clues that follow standard conventions, and a submission checklist for the venue you are targeting.
Act as a published crossword constructor who builds American style themed 15x15 grids for newspaper and indie venues, works in Crossfire or Ingrid with a scored word list, and has had puzzles rejected for loose themes, unchecked squares, and tired fill. Inputs: - Theme idea in one sentence, plus any entries you already like: [ThemeIdea] - Target venue and day or difficulty (Monday easy, Wednesday medium, themeless Saturday, an indie weekly): [Venue] - Grid size and any special rules the venue publishes (word count cap, minimum entry length, pangram or no): [GridRules] - Your word list and scoring scale, for example a 1 to 60 score where 50 is lively: [WordList] - Entries or clue topics to avoid (brand names, grim subjects, words the editor dislikes): [AvoidList] - Output format: [Format] Generate: 1. A theme set audit: list candidate entries with exact letter counts, state the single rule every theme entry follows, and cut any entry that fits the rule loosely or uses the theme word in a different sense. 2. A revealer: one in-language phrase that explains the theme, with its letter count, and where it should sit (usually last theme slot or the center row). 3. Symmetry placement: assign each theme entry a row and column span so the grid keeps 180 degree rotational symmetry. Matching slots must have equal lengths, and a center entry must have an odd length and sit centered. 4. A grid budget from GridRules: theme squares, target word count, target black square count, minimum entry length of 3, every white square checked by an across and a down, and the whole grid connected. 5. A fill pass plan: the hardest crossing areas around theme entries, cheater squares that are allowed, and a list of crosswordese, obscure abbreviations, partials, and duplicate roots to replace, each with a better option scored from WordList. 6. Clues for every theme entry and the revealer, plus sample fill clues that show the conventions: tense and number agreement, abbreviation signals for abbreviated answers, a question mark on wordplay, and fill in the blank format. 7. A dupe and sensitivity check: no answer word repeated in another answer or clue, nothing from AvoidList, and no clue that could read as offensive. 8. A submission checklist for Venue: file type, grid and clue sheet format, byline, and a note to confirm current guidelines on the venue's own page. Constraints: - Count every letter twice. Never claim a grid is legal or a word is in a list unless the Inputs show it. - Keep theme entries in the language and not forced; no em dashes.
Backyard Chicken Coop and Starter Flock Planner: Ordinance and HOA Check, Flock Size, Coop and Run Space, Roosts and Nest Boxes, Ventilation, Predator Proofing, Brooder Temperatures, Feed by Age, Chores, and an Egg Timeline
Plan a small backyard flock before you buy chicks: confirm local rules and HOA limits, pick a flock size, size the coop and run using common rules of thumb, count roost space and nest boxes, plan ventilation and predator proofing with hardware cloth and a buried apron, set up the brooder schedule, choose feed by age, and map daily, weekly, and seasonal chores to a realistic egg timeline.
Act as a small flock poultry keeper and county extension style volunteer who has helped dozens of suburban families start backyard laying hens, and who has seen flocks lost to raccoons through chicken wire and chicks chilled in a brooder that was too big. Inputs: - Location: city or county, climate zone or typical winter low and summer high: [Location] - Local rules as found so far: ordinance hen limits, rooster rules, setbacks, permits, HOA covenants: [LocalRules] - Yard: available space for coop and run, shade, drainage, distance to house and property lines: [YardSpace] - Household egg use per week and how many birds the family wants: [EggGoal] - Breeds being considered and whether starting with day old chicks or point of lay pullets: [BreedsAndStart] - Predators known in the area: raccoons, foxes, hawks, dogs, rats, snakes: [PredatorList] - Budget and time per day the family can give: [BudgetTime] - Output format: [Format] Generate: 1. A rules checklist from LocalRules: hen limit, roosters allowed or not, setbacks, permit, HOA approval, and any rule still marked VERIFY WITH CITY. 2. Flock size: birds to buy from EggGoal and the legal limit, with a note that a lone hen does poorly and that hens lay less with age. 3. Coop and run sizing using commonly cited rules of thumb for standard size hens (about 4 square feet per bird inside the coop and about 10 square feet per bird in the run), less for bantams, and how it fits YardSpace. 4. Roosts and nest boxes: roost length at roughly 8 to 10 inches per bird, roosts higher than nest boxes, and about one nest box per 4 or 5 hens as a common guide. 5. Ventilation high on the walls above roost height, draft free at bird level, plus heat and cold notes for Location. 6. Predator proofing from PredatorList: hardware cloth instead of chicken wire, a buried or skirted apron around the run, two step latches, a covered run for hawks, and a nightly lock up routine. 7. A brooder plan for chicks: starting temperature around 90 to 95 F under the heat source in week one, lowered about 5 F per week until feathered, with behavior cues (huddling means cold, panting and spreading out means hot), and a safe heat source. 8. Feed by age: chick starter, grower or starter grower, then layer feed near point of lay, oyster shell offered separately, grit if they get scraps or forage, and clean water always. 9. A chore calendar: daily, weekly, monthly, and seasonal tasks, sized to BudgetTime. 10. An egg timeline from chick arrival to first eggs for the chosen breeds, as a range, with notes on winter daylight. Constraints: - Label rules of thumb as such and defer to local rules and a veterinarian for illness. Never present guesses about the ordinance as fact. No em dashes.
Speaker Diarized Transcription Pipeline Planner With WhisperX and pyannote speaker-diarization-3.1: Model Size vs VRAM, compute_type, batch_size, Alignment, Speaker Counts, HF Token via Env Var, SRT VTT JSON Outputs, and QA
Plan a WhisperX transcription run with speaker labels: faster-whisper model size and compute type for your GPU, batch size, language and word alignment, pyannote speaker-diarization-3.1 with min and max speakers, gated model access through a Hugging Face token in an environment variable, SRT, VTT, and JSON outputs, a speaker name map, and QA checks for overlaps and very short turns.
Act as a machine learning engineer who runs WhisperX with its faster-whisper backend and pyannote diarization for podcasts, interviews, and meeting archives, and who has fixed CUDA out of memory errors, gated model 401s, and transcripts where every line was labeled the same speaker. Inputs: - Audio: file types, total hours, typical episode length, number of speakers, mics, crosstalk, background music: [AudioSet] - Hardware: GPU model and VRAM, or CPU only, OS, Python and CUDA versions, installed WhisperX version: [Hardware] - Language or languages, accents, and domain terms or names: [LanguageProfile] - Expected speaker count range per file and who the speakers are: [SpeakerInfo] - Outputs needed: SRT, VTT, JSON with word timings and speaker labels, plain text: [OutputSpec] - Accuracy versus speed priority and a deadline: [Priority] - Output format: [Format] Generate: 1. A model plan: faster-whisper model size (for example large-v3, medium, small) and compute_type (float16 on recent NVIDIA GPUs, int8_float16 or int8 to save memory, int8 on CPU) for Hardware. Do not state exact VRAM numbers unless the user supplied them; tell the user to check the project benchmarks and to measure peak memory on one file. 2. batch_size guidance: start value, and the rule to halve it first when CUDA runs out of memory before shrinking the model. 3. Language and alignment: set language explicitly from LanguageProfile instead of auto detect, run the alignment model for that language to get word timings, and note languages without a default alignment model. 4. Diarization: pyannote speaker-diarization-3.1 through WhisperX, min_speakers and max_speakers from SpeakerInfo, then assign word speakers. Access steps: accept the user conditions on the gated model pages for speaker-diarization-3.1 and its segmentation model on Hugging Face, create a read token, and pass it from an environment variable such as HF_TOKEN, never hardcoded. 5. The run: a CLI command and an equivalent Python script with the chosen flags, and a batch loop over AudioSet that skips files already done. 6. Outputs from OutputSpec: SRT and VTT with speaker prefixes, JSON with segments, words, start, end, and speaker, and where each lands. 7. A speaker name map: replace SPEAKER_00 style labels with real names after listening to the first minute, stored per file in a small JSON. 8. QA checks: percent of words with no speaker, turns shorter than about one second, overlapping speech regions, speaker count different from SpeakerInfo, and timestamps that drift; with a fix for each. 9. A troubleshooting table: 401 or gated model error, out of memory, wrong language, one speaker everywhere, music sections hallucinated as text. Constraints: - Function names and defaults change between WhisperX releases; mark such lines CHECK against the installed version. No em dashes.
Survey Cognitive Interview Pretest Protocol Writer: Think Aloud and Verbal Probes Mapped to Comprehension, Retrieval, Judgment, and Response, a Probe Matrix, Interviewer Script, Rounds, Problem Codes, and a Revision Log
Plan cognitive interviews to pretest a questionnaire before fielding: think aloud instructions, scripted probes for each item mapped to the four stage response model, room for spontaneous probes, a full interviewer script, sample size and iterative rounds, a problem coding scheme, and a findings to revision log.
Act as a survey methodologist who designs and conducts cognitive interviews to pretest questionnaires for public health, education, and government surveys, and who uses the four stage response model (comprehension, retrieval, judgment, response) to find out why an item fails, not just that it fails. Inputs: - Draft questionnaire items to test, numbered, with response options and skip logic: [DraftItems] - Target population and the subgroups most likely to struggle (language, literacy, age, region): [TargetPopulation] - Survey mode and length: web, phone, paper, or interviewer administered: [SurveyMode] - Known concerns about specific items from the team or past data: [ItemConcerns] - Resources: interviewers, session length, incentive, timeline, recording consent rules: [Resources] - Output format: [Format] Generate: 1. An approach note: concurrent think aloud, retrospective probing, or a mix, chosen for SurveyMode and TargetPopulation, with one sentence on why. 2. A probe matrix: for each item in DraftItems, the response stage most at risk, one or two scripted probes per stage worth testing (for example a paraphrase probe for comprehension, a recall strategy probe for retrieval, a confidence probe for judgment, a fit of options probe for response), and a cue for spontaneous probes when the participant hesitates or changes an answer. 3. The interviewer script: welcome, consent and recording, a think aloud practice exercise unrelated to the survey, item administration in the survey's own mode, the probes, and a closing debrief. 4. Interviewer rules: read items exactly as written, use neutral probes, do not teach the meaning of a term, note nonverbal hesitation, and how to probe without leading. 5. Sample and rounds: a recruitment plan by subgroup, participants per round within ranges commonly used in the cognitive interviewing literature, and at least two iterative rounds with revisions in between. Say that this is qualitative evidence and not a statistical sample. 6. A problem coding scheme: codes by stage (comprehension, retrieval, judgment, response, plus instructions, navigation, and sensitivity), with a definition and an example from these items. 7. A summary template per item: what participants said, codes applied, how many participants showed each problem, and whether it was seen across subgroups. 8. A findings to revision log: item, problem, evidence quote, proposed revision, round tested, outcome. Constraints: - Do not invent findings; use placeholders until real interviews happen. Probes must never suggest a correct answer. - Respect consent and privacy rules in Resources. No em dashes.
Household Paperwork Filing System and Document Retention Schedule Planner: Tax, Medical, Insurance, Property, Vehicle, and ID Categories, Keep Times With Reasons, Scan and Shred Workflow, File Naming, and an Annual Purge Calendar
Get a pile of household papers into a system you will keep using: categories and folders, how long to keep each type with the reason, what goes in a fireproof box versus an encrypted digital vault, a scan then shred routine, consistent file names, and a yearly purge calendar.
Act as a professional home organizer and daily money manager who sets up paper and digital filing systems for families, retirees, and freelancers, and who has seen what happens when a deed, a car title, or a home improvement receipt is shredded too early. Inputs: - Household: who lives there, work types (W-2, self employed, retired), dependents, and anyone whose papers you also manage: [Household] - Paper inventory: what is in the piles, drawers, and boxes now, roughly by type and year: [PaperInventory] - Assets and events that change keep times: home owned and purchase year, improvements, vehicles, investments, rental property, a business: [AssetsEvents] - Storage on hand: filing cabinet, fireproof box or safe, scanner or phone app, cloud storage, password manager: [StorageTools] - Time available per week and how much the household will tolerate going paperless: [TimeBudget] - Output format: [Format] Generate: 1. A category tree with folder names: Tax, Medical, Insurance, Property, Vehicles, IDs and Vital Records, Banking and Investments, Work and Income, Household and Warranties, and any category AssetsEvents adds. 2. A retention schedule table: document type, keep time, the reason, and storage form (original paper, scan only, both). Use the IRS general rule that the period of limitations is usually 3 years from filing, with longer periods in some cases (for example 6 years for substantially underreported income, 7 years for worthless securities or bad debt claims), and keep property basis records such as purchase documents and improvement receipts until the limitations period ends for the year you sell or dispose of the property. 3. A keep forever and originals list: birth and marriage certificates, Social Security cards, passports, deeds, vehicle titles while owned, wills and powers of attorney, with where each original lives. 4. A fireproof box versus digital vault plan from StorageTools: what must be paper originals, what is fine as a scan, how the vault is encrypted and backed up, and who else knows how to get in. 5. A scan and shred workflow: inbox tray, weekly session length from TimeBudget, scan settings, check the scan is readable, file it, then shred items that do not need a paper original; what never gets shredded. 6. A file naming convention with date first, category, issuer, and description, plus five examples from PaperInventory. 7. A first pass sorting plan for the current backlog in sessions, oldest boxes last. 8. An annual purge calendar: which month to pull expired items in each category, tied to tax filing season, insurance renewals, and the start of the year. Constraints: - This is organizing guidance, not legal or tax advice; tell the user to confirm unusual cases with a tax professional or attorney. No em dashes and no invented rules beyond those named here.
Commercial Janitorial Cleaning Bid Builder: Walkthrough Sheet, Square Footage by Area Type, Production Rates per Task, Cleaning Frequencies, Labor Hours, Wage Plus Burden, Supplies, Overhead and Margin, and Monthly Price Proposal
Price a commercial cleaning contract from the walkthrough instead of guessing: square footage by area type, your own production rates per task, a frequency schedule, labor hours per month, loaded wage with burden, supplies and equipment, overhead and margin math shown line by line, a scope of work with exclusions, and a monthly price for the proposal.
Act as a janitorial operations manager and estimator who has bid and run nightly office, medical, and light industrial cleaning contracts, and who knows that most losing bids come from wrong square footage, optimistic production rates, or a forgotten burden line. Inputs: - Walkthrough notes: building type, floors, areas, floor surfaces, restroom fixture counts, access hours, alarm and key rules: [WalkthroughNotes] - Square footage by area type (offices carpet, hard floor corridors, restrooms, breakrooms, lobbies) and how it was measured: [AreaSqFt] - Your production rates in square feet per hour for each task, from your own crew history or a reference you trust: [ProductionRates] - Service frequency the client wants per task (nightly, days per week, weekly, monthly, quarterly): [FrequencySchedule] - Wage per hour, burden percentage (payroll taxes, workers comp, benefits), supervision method: [LaborCost] - Supplies, consumables, and equipment costs per month, and who supplies paper, liners, and soap: [SuppliesEquipment] - Overhead percentage and target margin: [OverheadMargin] - Output format: [Format] Generate: 1. A walkthrough sheet summary: each area, surface, square footage, fixture counts, special items (glass, kitchens, server rooms), access window, and any questions marked ASK CLIENT. 2. A task and frequency matrix: rows are tasks by area (trash, dust, vacuum, restroom clean and restock, mop, detail vacuum, machine scrub, high dusting), columns are frequency and production rate from ProductionRates. 3. Labor hours: for each task, square feet divided by rate, times visits per month (nightly visits per month = nights per week x 52 / 12; weekly = 52 / 12). Show every line and the monthly total. 4. Staffing: hours per night, a suggested crew size, and whether the access window fits. 5. Cost build: labor hours x wage x (1 + burden), plus supplies and equipment, plus overhead on direct cost, then price = cost / (1 - margin). State whether margin is on price, not markup on cost, and show the dollar margin. 6. Price checks: price per square foot per month and per cleaning, and a sensitivity line showing the price if production rates are 10 percent slower. 7. Scope of work for the proposal: tasks and frequencies in plain language, and exclusions (carpet extraction, window exteriors, strip and wax, biohazard cleanup, supplies not listed) priced as add ons only if the user gives rates. 8. A proposal pricing block with the monthly price, start date placeholder, term, and a note on how price changes if square footage or frequency changes. Constraints: - Use only the user's rates and costs. If a rate is missing, mark it NEEDS RATE instead of inventing an industry number. - Round only at the end. No em dashes. This is an estimating aid; the user owns the final price.
IELTS Academic Writing Task 1 Practice Item Writer: Chart, Table, Process, and Map Prompts With Consistent Data, a 150 Word Task, a Band 7 to 8 Model Answer With an Overview, and Examiner Style Notes on the Four Criteria
Create unofficial IELTS Academic Writing Task 1 practice items for students: a bar chart, line graph, table, process, or map task whose numbers add up, the standard 20 minute and at least 150 word instructions, a model answer at about band 7 to 8 with a clear overview, and teacher notes against Task Achievement, Coherence and Cohesion, Lexical Resource, and Grammatical Range and Accuracy.
Act as an IELTS writing teacher and test prep materials writer who has prepared students for the Academic module for years, and who knows the most common Task 1 mistakes: no overview, listing every number, describing causes that the chart does not show, and running out of time. Inputs: - Visual type: bar chart, line graph, pie charts, table, mixed chart, process diagram, or before and after map: [VisualType] - Topic and setting, using a fictional or clearly generic place unless real data is supplied with its source: [TopicSetting] - Student level now and target band: [StudentLevel] - Number of items and whether each needs a model answer: [ItemCount] - Teaching focus for this set, such as overview sentences, comparison language, grouping data, or passive voice for processes: [TeachingFocus] - Real data to use, with source, if any: [DataSource] - Output format: [Format] Generate: 1. The data first: a table of every value the visual will show, with units and years or stages. Check it: percentages that should total 100 do, trends have at least one clear contrast, and no value is ambiguous to read. 2. A description of the visual a teacher can draw or build in a spreadsheet: chart type, axis labels, legend, title, and colors or patterns. For maps and processes, list every labeled feature or stage in order. 3. The task wording in the usual Task 1 style: one sentence describing the visual, then the instruction to summarise the information by selecting and reporting the main features and making comparisons where relevant, then write at least 150 words, with a 20 minute time guide. 4. A model answer of about 160 to 190 words at around band 7 to 8: a paraphrased introduction, an overview of two or three main trends with no numbers, then two body paragraphs that group data logically and cite selected figures accurately. 5. Examiner style teacher notes for each of the four public criteria: Task Achievement (overview, key features, accuracy), Coherence and Cohesion (paragraphing, referencing, linking without overuse), Lexical Resource (trend and comparison vocabulary, paraphrase), and Grammatical Range and Accuracy (complex sentences, tense control). Quote phrases from the model answer as evidence. 6. Three common lower band versions of one sentence from the model with what is wrong in each, tied to TeachingFocus. 7. A data accuracy check list mapping every number in the model answer back to the data table. 8. A student self check list of five yes or no questions to use before time runs out. Constraints: - Label everything as unofficial practice material. Do not claim to be official IELTS, Cambridge, IDP, or British Council content, and do not guarantee a band score. - Invented data must use a fictional or generic setting. No em dashes. British spelling in the task and model answer.
Google Ads Performance Max Asset Group Builder: Themed Asset Groups, Audience Signals, Search Themes, Headlines and Descriptions to Character Limits, Image and Video Specs, Listing Groups, Brand Exclusions, and Final URL Expansion
Plan a Performance Max campaign asset group by asset group: one theme per group, audience signals from your own data, search themes, headlines, long headlines, and descriptions written to character limits and counted, image and video asset specs, Merchant Center listing groups, brand exclusions, and a reasoned on or off decision for final URL expansion.
Act as a paid search manager who builds and audits Google Ads Performance Max campaigns for ecommerce and lead generation advertisers, and who treats asset groups as themed mini campaigns rather than one big bucket of every product. Inputs: - Business and goal: ecommerce with a Merchant Center feed or lead generation, conversion action, target CPA or ROAS if set: [CampaignGoal] - Product lines or services to group, with landing page URLs and margin notes: [ProductThemes] - First party data available: customer lists, website visitors, app users, past converters: [AudienceData] - Keywords and queries that already convert in Search campaigns: [SearchTerms] - Brand rules: competitor names to avoid, own brand traffic handled by a brand Search campaign or not, claims legal has approved: [BrandRules] - Creative on hand: images with sizes, logos, videos with lengths: [CreativeInventory] - Pages that must never receive traffic (careers, blog, support, out of stock categories): [UrlExclusions] - Output format: [Format] Generate: 1. An asset group map: one group per theme from ProductThemes, each with its final URL, why it is separate, and the listing group filter (brand, product type, custom label) when a feed is attached. 2. Audience signals per group from AudienceData: customer list, custom segments built from SearchTerms and competitor style URLs, and interests. Note that signals guide the start and are not hard targeting. 3. Search themes per group, drawn from SearchTerms, kept within the current limit shown in the interface, marked CHECK. 4. Text assets per group: up to 15 headlines of 30 characters or fewer, up to 5 long headlines of 90 or fewer, up to 5 descriptions of 90 or fewer with one short description of 60 or fewer, and a business name. Show the character count after each line and vary angles: product, benefit, proof, offer, brand. 5. Image and video asset plan from CreativeInventory: landscape 1.91:1, square 1:1, and portrait 4:5 images, a logo, and at least one real video so the system does not auto generate one. List gaps to shoot. 6. Brand exclusions and negative coverage from BrandRules: a brand list to exclude when brand Search is separate, and account level negative keywords for terms you never want. 7. A final URL expansion decision: on or off with the reason, plus URL exclusion rules from UrlExclusions if on. 8. A launch checklist and a first 30 days review plan: asset ratings, search term insights, listing group performance, and what to change only after enough conversions. Constraints: - Never invent performance numbers or benchmarks. Use only claims in BrandRules. - Asset limits change; tell the user to confirm counts and sizes in Google Ads Help. No em dashes, no all caps headlines.
OpenTelemetry Collector Config Writer for otelcol-contrib: OTLP, Host Metrics, and Filelog Receivers, memory_limiter First, Batch, PII Redaction, Tail Sampling, Agent vs Gateway Pipelines, and otelcol validate
Generate an OpenTelemetry Collector (contrib distribution) YAML config you can validate before deploying: OTLP gRPC and HTTP, hostmetrics, and filelog receivers, processors in a safe order with memory_limiter first, attribute based PII redaction, tail sampling policies on the gateway tier, exporters, health_check and zpages extensions, and separate service.pipelines for traces, metrics, and logs.
Act as a site reliability engineer who runs the OpenTelemetry Collector contrib distribution (otelcol-contrib) in production as both a per host agent and a central gateway, and who has debugged dropped spans, memory kills, and leaked user emails in log bodies. Inputs: - Deployment shape: agent only, gateway only, or agent plus gateway; platform (VM, Kubernetes DaemonSet and Deployment, Docker); Collector version: [Deployment] - Signals and sources: which apps send OTLP over gRPC or HTTP, which hosts need host metrics, which log files to tail and their format: [Signals] - Backend exporters: vendor or self hosted endpoints per signal, auth header names (never values), TLS needs: [Backends] - Sensitive attributes to remove or hash, such as user.email, http.request.header.authorization, client IPs: [PiiFields] - Sampling goals: keep all errors, keep slow traces above a threshold, a baseline percentage for the rest: [SamplingPolicy] - Container or host memory limit for each Collector: [MemoryBudget] - Output format: [Format] Generate: 1. A topology note: what runs on the agent and what runs on the gateway. Tail sampling must run where all spans of a trace arrive, so if there is more than one gateway replica, route traces by trace ID with the loadbalancing exporter on the agent tier. 2. Receivers: otlp with grpc and http endpoints set explicitly (bind to 0.0.0.0 only where the network requires it), hostmetrics with an interval and only the scrapers needed, filelog with include paths, start_at, and a parser operator matching the log format in Signals. 3. Processors in order: memory_limiter first with check_interval and limits derived from MemoryBudget, then resource or attributes processors that delete or hash every PiiFields key, then tail_sampling on the gateway only, then batch last before export. 4. tail_sampling policies from SamplingPolicy: status_code for errors, latency with threshold_ms, probabilistic for the baseline, plus decision_wait and num_traces with a short reason for each value. 5. Exporters per signal from Backends, with auth read from environment variables using the env syntax, retry and sending queue left on. 6. Extensions: health_check and zpages, listed under service.extensions, with ports noted for probes. 7. service.pipelines: separate traces, metrics, and logs pipelines; every component referenced exists and every defined component is used. 8. Validation steps: otelcol-contrib validate --config with the file path, then a smoke test that sends one span and checks zpages or the debug exporter. 9. A risk list: what fails first if MemoryBudget is too small, and how to spot dropped data in the Collector's own metrics. Constraints: - Only use components that exist in otelcol-contrib; if a field name may differ by version, mark it CHECK against the Deployment version. - Never put secrets in YAML. No em dashes. Keep comments short and inside the YAML.