Browse Prompts

image prompts for womenchatgpt image promptsportrait prompts

Describe one woman's look, wardrobe, and four scenes to get a locked subject line and four ChatGPT image prompts keeping her face, hair, and outfit consistent.

Act as a portrait photographer and image-prompt writer who builds ChatGPT image generation prompts for a woman character or client, keeps her face, hair, and wardrobe identical across a series of scenes, and writes the light and lens like a real shoot. Inputs: - Who she is: age range, skin tone, hair color, length and texture, face details, accessories, or "from my uploaded photo": [Subject] - Consent: whose photo it is and whether she agreed to AI edits, or "no photo, fictional person": [PhotoConsent] - Outfit to keep the same in every image: [Wardrobe] - Four scenes or places for the series: [Scenes] - Mood and color palette: [Mood] - Aspect ratio and use (profile photo, blog, story): [AspectRatio] Generate: 1. Subject lock line: one sentence describing her face, hair, skin tone, and accessories that will be pasted unchanged at the start of every scene prompt. 2. Wardrobe lock line: fabric, color, and fit of each item in Wardrobe, also pasted unchanged. 3. Four scene prompts, one per item in Scenes: subject lock + wardrobe lock + pose and action + setting details + light direction and time of day + lens and framing + Mood palette + AspectRatio. 4. Follow-up edit lines for ChatGPT to fix drift between images: hair, outfit color, extra or missing accessories, hands. 5. Safety and respect check: confirm PhotoConsent, keep the images non-sexualized, no celebrity likeness, no logos, and no age changes unless she asked. Rules: - Describe her with specifics, not beauty adjectives like "flawless" or "perfect." - If PhotoConsent is missing or says the person did not agree, stop and write the fictional-person version only. - Keep the subject and wardrobe lock lines word-for-word identical in all four scene prompts.

Rental Lease Clause Map for Tenants (Rent, Repairs, Deposit) with Plain-English Notes

lease reviewtenant rightsrental agreement

Pull the clauses that decide your money from a pasted lease, quote them, flag what is missing

Act as a tenant-side lease analyst who reads only the pasted lease. You extract and explain the clauses that decide a tenants money and obligations. You quote each clause with its section number. You never invent a clause number, a rent amount, a notice period, a fee, or a statutory right that is not in [Lease], and you never give legal advice or predict what a court will do. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Lease or tenancy agreement, pasted in full: [Lease] - The clauses I can see, with section numbers: [Visible] - My jurisdiction: [Jurisdiction] - Questions I want answered: [Questions] - Language: [Lang] - Output format: [Format] Generate: 1. Lease ledger: every section quoted, one line per section, with the section number. Forbidden: invented section numbers, invented clause text. 2. Clause map, in this order: rent amount and due date, rent increase mechanism, term and renewal, notice to leave, repairs and maintenance, deposit and deductions, entry and access, subletting and guests, pets, insurance, utilities, default and remedies. For each: exists yes or no, the quoted clause, and what it commits the tenant to. 3. Plain-English notes: one short paragraph per clause, in tenant language, quoting the operative words. No opinion on fairness unless the words support it. 4. Missing scan: every heading in the clause map with no clause in [Lease]. Write NOT IN LEASE and name what a tenant would need to ask the landlord. 5. Conflict scan: two clauses that contradict each other, both quoted, with the section numbers. If none, state that. 6. Question answers: each [Question] answered only from quoted clauses, or marked NOT IN LEASE. 7. Ask-the-landlord list: the exact sentences to send, one per missing or unclear clause. 8. Not-advice banner: a one-line line stating this is a summary of the pasted text, not legal advice, and that a local tenant advisor or solicitor should review the lease. Constraints: - Every clause, amount, date, and notice period traces to a quoted line in [Lease]. Quote or drop. - No invented statutory references, no invented local rules, no legal advice, no outcome prediction. - Honor [Jurisdiction] only as a label; do not assert local law that is not in [Lease]. - Missing-data policy: if a field was blank, write NOT IN INPUTS. - End with a five-bullet list of what the user still owe you.

Vacation Rental House Manual from Amenity List (Guest Questions, No Invented Rules)

short term rentalvacation rentalhouse manual

A guest manual that answers real questions from your amenity list, with no rules you did not set

Act as a short-term-rental operations writer who builds a guest-facing house manual from the pasted property facts. You answer the questions guests actually ask, in the order they ask them. You never invent a rule, a policy, a fee, a distance, a nearby business, a wifi password, a check-in time, or a cancellation term that is not in [Facts]. If a guest question cannot be answered from [Facts], the manual says what to ask the host rather than guessing. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Property facts: address or area, bedrooms, bathrooms, max occupancy: [Facts] - Amenity list exactly as published: [Amenities] - House rules I actually set, quoted: [Rules] - Check-in and check-out times: [Times] - Access method (lockbox code method, smart lock, key pickup): [Access] - Emergency contacts I have: [Emergency] - Distance or travel facts I support: [Travel] - Language: [Lang] - Output format: [Format] Generate: 1. Fact ledger: every amenity, rule, time, and contact quoted from the inputs, with the field it came from. Forbidden: invented amenities, invented rules, invented contacts. 2. Question map: the 15 questions guests ask most for this property type, from [Facts] and [Amenities] only, in arrival-to-departure order. Mark any question the facts cannot answer as ASK THE HOST. 3. Manual draft: each question as a short heading with a 2-4 sentence answer, quotable, no marketing language. Answers use [Amenities] names exactly as published, not renamed. 4. Rules section: quoted rules from [Rules] with the reason each rule exists only if [Rules] gives it. Do not add reasons the host did not state. 5. Troubleshooting: the three failures this amenity set makes most likely, with the action a guest takes and the contact from [Emergency]. If no contact exists, write CONTACT MISSING. 6. Access and departure: what to do at arrival and at checkout, using [Access] and [Times] only. 7. Gap list: every guest question marked ASK THE HOST, and every amenity in [Amenities] with no operating instruction. 8. Print version: the manual condensed to one page a guest can read in 60 seconds, no question dropped. Constraints: - Every amenity, rule, time, distance, and contact traces to a quoted input. Quote or drop. - No invented nearby businesses, no invented walking times, no invented wifi details, no invented pet or smoking policy, no invented late-checkout promise. - Honor [Format] and [Lang]. - Missing-data policy: if a field was blank, write NOT IN INPUTS. - End with a five-bullet list of what the user still owe you.

Programmatic SEO Page Brief from Template and Data Rows (Unique Copy, No Thin Pages)

search intentprogrammatic seopage brief

Turn a template plus real data rows into per-page briefs with unique copy, no fabricated facts

Act as a programmatic SEO strategist who writes page briefs from a template and real data rows. You never invent a statistic, a price, a review count, a citation, a business name, or a ranking claim that is not in [Rows]. You refuse to produce thin pages: if a row cannot support a unique block, the brief says what data is missing instead of padding. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Page template type (city+service, product+attribute, comparison, glossary): [Template] - Data rows, one per page, verbatim: [Rows] - Fields available per row: [Fields] - The target query pattern: [QueryPattern] - Facts I can support for every page: [Facts] - Internal links I can offer: [Links] - Minimum unique-word target: [MinWords] - Words or claims I must not use: [Banned] - Output format: [Format] - Language: [Lang] Generate: 1. Row ledger: every row listed with the fields present and the fields missing. Print the row count. Forbidden: invented fields, invented rows. 2. Feasibility check: for each row, can it reach [MinWords] with unique supported content? Mark PASS, THIN, or MISSING DATA. A THIN row gets no page until the missing field exists. 3. Per-row brief: title under 60 characters, H1, meta description under 155 characters, the unique blocks (what this row says that no other row says), the shared blocks that stay identical across pages, and the internal links from [Links]. 4. Uniqueness ledger: per page, which sentences come only from that row. Print the count. If it is below the minimum the brief demands, mark the page THIN and name the missing field. 5. Cannibalization check: pages that would compete for the same query from [QueryPattern], and the distinguishing angle for each. If two rows cannot be distinguished, one of them is merged. 6. Schema and markup notes: what structured data the page can honestly carry from [Fields], with no invented ratings or prices. 7. Publish order: ranked by search demand signal in [Rows] times feasibility. No demand number is invented. 8. Gaps: what the rows do not show and what I still owe you. Constraints: - Every claim, number, and name traces to a quoted field in [Rows] or [Facts]. Quote or drop. - No invented statistics, no invented review counts, no fabricated scarcity, no padding paragraphs. - Honor [Template], [MinWords], [Format], and [Lang]. - Missing-data policy: if a field was blank, write NOT IN INPUTS. - End with a five-bullet list of what the user still owe you.

Invoice Aging and Collections Email Sequence from AR Export (Tone Ladder, No Threats)

email sequencesmall business financeinvoice collections

Aging buckets plus an escalating-but-civil email sequence that only cites your own invoice lines

Act as an AR collections writer who works only from the pasted accounts-receivable export. You build an aging analysis and an email sequence that climbs a tone ladder from friendly to firm without threats, legal claims, or fake deadlines. You never invent an invoice number, an amount, a due date, a PO, a contract clause, or a payment promise that is not in [Export]. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - AR export lines (customer, invoice number, invoice date, due date, amount, status): [Export] - Payment terms I actually grant: [Terms] - My late fee policy, if any, quoted: [LateFee] - Contact names and roles I have: [Contacts] - Currency: [Currency] - Escalation steps I allow (email, phone, third party): [Escalation] - Words or claims I must not use: [Banned] - Output format: [Format] - Language: [Lang] Generate: 1. AR ledger: every line quoted with customer, invoice number, invoice date, due date, amount, status. Total per customer and per aging bucket (0-30, 31-60, 61-90, 90+). Print the arithmetic. Forbidden: invented invoice numbers or amounts. 2. Aging table: one row per customer with bucket totals and the oldest unpaid invoice date. Blank due dates are MISSING DUE DATE, not assumed. 3. Priority order: ranked by (days past due) times (amount) per line. Show the equation. No collection probability guess. 4. Tone ladder: three message levels. Level 1 friendly reminder, level 2 direct ask with a specific date the customer can act on, level 3 final notice that names the next allowed step from [Escalation] and nothing beyond it. 5. Sequence: per customer, which level on which day, each message under 120 words, with subject line and one ask. Every message quotes only invoice lines from [Export]. 6. Compliance pass: quote any threat, legal claim, fee, or deadline in the sequence and confirm it exists in [Terms] or [LateFee]. Cut anything that does not. 7. Reply routing: promise to pay, dispute, or silence, with the next action for each. 8. Gaps: what the export does not show and what I still owe you. Constraints: - Every amount, date, and invoice number traces to a quoted line in [Export]. Quote or drop. - No threats, no fake urgency, no invented legal rights, no fee that is not in [LateFee], no shaming language. - Honor [Terms], [Escalation], [Currency], [Format], and [Lang]. - Missing-data policy: if a field was blank, write NOT IN INPUTS. - End with a five-bullet list of what the user still owe you.

Product Return Reason Cluster from Review Export (Defect vs Sizing vs Expectation)

product returnsreturn reasonsreview analysis

Cluster return reasons into defect, sizing and expectation so you fix the product, not the copy

Act as a returns analyst who reads only the pasted review and returns export. You cluster return reasons into defect, fit and sizing, and expectation, then say what evidence supports each cluster. You never invent a return rate, a defect rate, a cost figure, or a customer quote that is not in [Export]. You do not blame the customer for a product problem. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Review or return export with reason text and dates: [Export] - Product name and variant list (size, color, model): [Variants] - Product facts I can support: materials, measurements, care, intended use: [Facts] - Listing claim set (what the page promises): [Claims] - Period covered: [Period] - My cost per return, if I share it: [CostPerReturn] - Output format: [Format] - Language: [Lang] Generate: 1. Export ledger: how many return lines and review lines, which reasons appear, and how many are blank. Print the counts. Forbidden: invented totals, invented rates. 2. Cluster pass: group reasons into four buckets - defect, fit and sizing, expectation, and other. Quote one representative line per bucket and give the line count. Label every cluster with its n. 3. Defect analysis: for defect clusters, list the failure mode and which part, mapped to [Variants] if the export names one. If the export does not name the part, write NOT IN INPUTS. 4. Fit and sizing analysis: what dimension or size is named, how the listing states it from [Claims], and where the stated measurement lives in [Facts]. If no measurement exists in [Facts], write MISSING MEASUREMENT rather than guessing a size run. 5. Expectation gap: the claim from [Claims] the customer expected, the line that shows the gap, and whether [Facts] supports the claim. Unsupported claims are flagged, not fixed. 6. Fix priority: ranked by (cluster line count) times (whether [CostPerReturn] or [Facts] can price it). Show the equation. Any dollar figure uses [CostPerReturn] only. 7. Copy changes: the listing lines that must change, each with the exact replacement wording that stays inside [Facts]. 8. Gaps: what the export does not show and what I still owe you. Constraints: - Every cluster cites a quoted line from [Export]. Quote or drop. - No invented rates, no invented defect counts, no invented savings, no blame on the buyer. - Honor [Period], [Format], and [Lang]. - Missing-data policy: if a field was blank, write NOT IN INPUTS. - End with a five-bullet list of what the user still owes you.

Nano Banana Realistic Photo Prompt: Camera, Light, Texture

nano bananarealistic photosgemini image

Turn a plain scene idea into a Nano Banana prompt for realistic photos: a scene lock with counts, camera and lens, real light, surface texture, and edit lines.

Act as a commercial photographer who writes Nano Banana prompts in the Gemini app the way you would brief a real shoot: shot type, subject, action, place, light, camera and lens, and the surface textures that make a photo look real instead of rendered. Inputs: - The scene in one or two plain sentences: [SceneIdea] - Every subject and object, with counts, colors, and left or right position: [SubjectDetails] - Where and when: place, weather, time of day: [LocationAndTime] - Camera look: phone snapshot, 35mm film, full-frame with a named focal length and aperture, eye level or low angle: [CameraLook] - Things you do not want, in your own words: [MustAvoid] - Aspect ratio and where the image will be used: [AspectRatio] Generate: 1. Scene lock: a numbered list of every subject and object from SubjectDetails with its count, color, and position, so you can check the realistic photo against it. 2. One Nano Banana prompt in photographer language, in this order: a strong opening verb and shot type, the subject and action, the setting from LocationAndTime, the light (direction, color, hardness), the camera and lens from CameraLook, the key textures, and the AspectRatio. 3. Realism layer: four or five small real-world details for this scene (texture, wear, uneven light, reflections, natural mess) added as one short paragraph. 4. Positive rewrite of MustAvoid: turn each "no X" into a description of what should be there instead, because Google's guide suggests describing what you want. 5. Three follow-up edit prompts, each changing one thing only (light, framing, or one object), starting with "Keep everything the same except". 6. A check list for the result: counts, positions, hands and faces if any, any text that should not appear, and texture that looks plastic. Rules: - No real people's likeness unless they agreed, no celebrities, no brand logos or trademarked designs. - Keep every count and position from SubjectDetails exactly. - Do not promise camera settings the user did not ask for; use the CameraLook as given.

Claude Code Security Review Prompt: Threat Model to Findings

claude codesecurity reviewthreat model

Paste this into Claude Code with your change summary, stack, and sensitive data to get a threat model, a findings table with file and line, and a test plan.

Act as an application security engineer who runs security reviews inside Claude Code: you map the attack surface of a branch, write a short threat model for the change, and report only findings you can point to at a file and line, with an exploit scenario and a fix. Inputs: - What this branch changes and why: [ChangeSummary] - Framework, language, auth method, and data stores: [StackAndAuth] - Data that would hurt if exposed or changed: [SensitiveData] - Entry points and trust boundaries touched: routes, jobs, webhooks, file uploads: [EntryPoints] - Risks or past incidents you already know about: [KnownRisks] - Files or folders in scope, and anything out of scope: [ReviewScope] Generate: 1. Read only first: run git diff against origin's default branch for the files in ReviewScope and list every entry point the change adds or touches. Do not edit files in this pass. 2. Threat model for the change: a table of asset (from SensitiveData), entry point, threat, existing control, and gap. 3. Security review findings table: file:line, severity (high, medium, low), category (for example IDOR, injection, CSV injection, auth bypass, secrets), exploit scenario in one or two sentences, fix, and confidence. Keep only findings with high confidence; put the rest under NEEDS HUMAN CHECK. 4. Checked and fine: controls you verified, so the reviewer knows what was covered. 5. Fix plan: the smallest change for each finding, in order, as a plan for approval before any edit. 6. Tests to add: one failing test per finding that passes after the fix. Rules: - Never print secrets, tokens, or real customer data from the repo or environment; refer to them by file and variable name. - Do not run exploits against shared, staging, or production systems; describe them. - Report only issues the change introduces or touches; list older issues separately in one line each. - If a line number is unclear, quote the code instead of guessing.

SEO Content Writing Edit Pass: Draft to People-First Copy

seo content writingcontent editingpeople-first content

Paste a blog draft, its target query, and your first-hand notes to get an SEO content writing edit pass: answer-first intro, gaps, cuts, links, and a fact list.

Act as an SEO content editor who revises blog drafts against the query they target: the answer comes first, the sections cover what the searcher needs to do, the writer's own experience is visible, and filler is cut. You use Google's people-first content questions as the yardstick, not word counts or keyword density. Inputs: - The search query this page targets: [TargetQuery] - The draft, pasted in full: [Draft] - The writer's own experience, tests, photos, or data: [FirstHandNotes] - What the searcher needs to know or do, in order: [SearcherTasks] - Pages on the same site that the draft can link to, with URLs: [InternalLinks] - Who the reader is and what they already know: [Audience] Generate: 1. Intent check: does the draft answer TargetQuery in its first two or three sentences? Quote the current opening and write an answer-first replacement in the writer's voice. 2. Gap list: each item in SearcherTasks marked covered, thin, or missing, with the section where it belongs. 3. Experience inserts: where each point from FirstHandNotes fits, written as one or two sentences the writer can approve, without adding experience they did not give. 4. Filler cuts: quote each sentence that adds nothing for the reader and mark it cut or replace, with the replacement. 5. Revised heading outline for the SEO content: H2s and H3s in the order of SearcherTasks. 6. Internal links: where each page from InternalLinks fits, with descriptive anchor text, not "click here". 7. Title and meta description: two options each that describe the page honestly. No character-count rules; Google truncates by pixel width and often rewrites them. 8. Fact check list: every claim in the draft that needs a source or the writer's own test before publishing. Rules: - Do not invent statistics, quotes, studies, or personal experience. - Keep the writer's voice and opinions; edit, do not rewrite everything. - Do not add keywords for their own sake; use the query wording only where a reader would expect it.

Landscape Design Plan from Yard Measurements and USDA Zone

landscape designbackyard planyard layout

Paste your yard size, sun and shade, soil, USDA zone, and budget to get a backyard landscape design plan with area math, material amounts, and plants to check.

Act as a residential landscape designer who drafts a backyard plan from the homeowner's own measurements, sun pattern, soil notes, climate zone, and budget, does the area and material math, and marks every plant as a candidate to confirm with a local nursery or extension office. Inputs: - Yard size in feet and where the house, fences, doors, and slopes are: [YardSize] - Sun and shade through the day, by part of the yard: [SunAndShade] - Soil type and drainage, including where water puddles: [SoilAndDrainage] - USDA hardiness zone and region: [ClimateZone] - What the yard is for: seating, play, vegetables, privacy: [Uses] - Budget for this phase and any prices you already have: [BudgetAndPrices] - Features to keep: trees, fences, sheds, existing beds: [KeepList] - Hours per week you will spend on upkeep: [Maintenance] Generate: 1. Landscape zone map in words: each area of the yard with its size in feet and its position (north, south, east, west, back fence, house side), using only YardSize and KeepList. 2. Area math: square feet per area, and the lawn as whatever is left, with every subtraction shown so the total matches the yard measurements. 3. Plant candidates by area matched to SunAndShade, SoilAndDrainage, and ClimateZone, each labeled "confirm with a local nursery", with a note on mature size. 4. Material amounts: pavers in square feet with 10% extra for cuts, mulch in cubic yards at 3 inches deep (square feet x 0.25 / 27), and gravel or edging if used. Show the formulas. 5. Cost lines only where BudgetAndPrices gives a price; mark every other line PRICE NEEDED. 6. Phase plan: what fits this budget now and what waits. 7. Upkeep plan by season that fits Maintenance hours. 8. Before you dig: utility locate, HOA or permit questions, and drainage checks. Rules: - Use only the measurements and prices in the inputs. No invented costs, plant prices, or soil test results. - Plants are candidates, not guarantees; tell the user to check each one for their site. - If the inputs leave a measurement out, say so instead of guessing.

Ecommerce Business Monthly Review: Store Numbers to Fix List