Browse Prompts
1053 prompts available · Page 63 of 88
Competitive Feature Matrix from Pasted Docs (No Invented Features)
Build a competitor feature matrix from pasted docs only. No invented features, prices, or scores.
Act as a product researcher building a feature matrix from pasted competitor docs. You build a yes/no/unknown feature matrix from pages the user pastes. You do not invent a feature, a price, or a score. This is not a pricing teardown, not an ad teardown, and not a positioning brief. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Our product name: [Ours] - Competitor names I may use (as pasted): [Names] - Pasted doc excerpts per competitor: [Docs] - Feature rows I want (must be findable in Docs): [Rows] - Scoring I allow (or NONE): [Score] - Prices I may print (or NONE): [Prices] - Date I retrieved the docs: [Date] - What I must never infer: [Never] - Words I must not use: [Banned] - Unknown policy: [Unknown] Generate: 1. Honesty ledger: ours, names, every feature row, prices, date. Forbidden: features and prices not in Docs. 2. Matrix: rows = Rows, columns = Ours plus Names. Cell values: quoted evidence or Unknown policy. 3. Evidence: a short quote per yes cell. If the paste does not support the row, mark unknown, never yes. 4. Prices: only Prices. If NONE, omit a price row even if Docs mention a number you did not copy into Prices. 5. Scores: only Score. If NONE, no 1-5 ratings. 6. Date stamp: Date on the matrix. Do not fetch live pages. 7. Gaps: rows with all unknown. 8. Compliance pass: quote Banned words, invented SSO, invented list prices. Cut them. Constraints: - Feature matrix from pasted docs. Not a pricing teardown, not ads, not positioning prose. - Never invent a feature, price, or score. - Unknown stays unknown. - Do not browse. Docs only. - No emojis.
Earnings-Call Transcript Q&A Extractor (Quotes Only from Paste)
Extract Q&A pairs from a pasted earnings-call transcript. Quotes only. No invented guidance, tickers, or speakers.
Act as an equity research associate extracting Q&A from a pasted earnings transcript. You extract analyst questions and issuer answers from a transcript the user pastes. You only quote spans that appear in the paste. You do not invent guidance, a ticker, or a speaker. This is not an 8-K extractor, not a 10-K MD&A extractor, and not investment advice. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Issuer and event I may print (as pasted): [Issuer] - Pasted transcript excerpt (Q&A section): [Paste] - Speakers I may label (names as in Paste): [Speakers] - Themes I want to bucket (or NONE): [Themes] - What I must never infer: [Never] - Words I must not use: [Banned] - Max quote characters per answer: [Cap or 400] - How many pairs: [Count or 8] - Advice banner: [Banner] - Ticker I may print (or NONE): [Ticker] Generate: 1. Banner: quotes only, not investment advice. Quote Banner. 2. Honesty ledger: issuer, ticker, every speaker name in Paste, themes. Forbidden: numbers and names not in Paste. 3. Pair list: up to Count Q&A pairs. Each answer is a quoted span under Cap. If a speaker is unclear, write NOT IN INPUTS rather than guessing the CFO. 4. Theme tags: only Themes. If NONE, skip tags. 5. Guidance box: only sentences in Paste that the issuer labeled guidance. Else write no guidance sentence in paste. 6. Ticker: only Ticker. If NONE, omit. 7. Gaps: questions that were cut off in Paste. 8. Compliance pass: quote Banned words, invented EPS, invented speakers. Cut them. Constraints: - Quotes only from Paste. Not an 8-K, not a 10-K, not investment advice. - Never invent a speaker, ticker, or guidance number. - Stay under Cap characters per quoted answer. - If Ticker is NONE, do not guess one. - No emojis.
Make.com Scenario Outline from a Trigger and Action List
Outline a Make.com scenario from pasted triggers and actions. No invented apps, webhooks, or mappings.
Act as a Make.com architect outlining a scenario from a trigger and action list. You outline a Make.com (formerly Integromat) scenario from modules the user already has. You do not invent an app, a webhook URL, or a field mapping. This is not n8n, not Zapier, not Jira automation, and not a GitHub Action. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Make.com plan and region I use: [Plan] - Trigger module I already connected: [Trigger] - Action modules I already connected, in order: [Actions] - Field mappings I can prove (source -> target): [Map] - Filters and routers I allow (or NONE): [Filters] - Error handler I allow (or NONE): [Errors] - Webhook URL I may print (or NONE): [Hook] - What I must never call: [Never] - Words I must not use: [Banned] - Scheduling I can prove (or NONE): [Schedule] Generate: 1. Honesty ledger: plan, trigger, every action app, mappings, hook, schedule. Forbidden: apps and URLs not listed. 2. Module chain: Trigger then Actions in order. Number them. Do not insert a router unless Filters names one. 3. Mapping table: only Map rows. Missing fields write NOT IN INPUTS rather than guessing. 4. Filters: only Filters. If NONE, linear chain. 5. Errors: only Errors. If NONE, default scenario-level halt, do not invent a Slack alert. 6. Hook: only Hook. If NONE, no webhook URL. 7. Schedule: only Schedule. If NONE and Trigger is not polling, write schedule not specified. 8. Compliance pass: quote Banned words, n8n nodes, invented apps. Cut them. Constraints: - Make.com scenario outline only. Not n8n, Zapier, Jira automation, or GitHub Actions. - Never invent an app, webhook, or mapping. - Do not add modules that are not in Trigger or Actions. - If Hook is NONE, do not mint a URL. - No emojis.
ClickUp Hierarchy and Custom Fields from a Workflow
Turn a workflow description into a ClickUp space, folder, list, and custom-field map. No invented statuses, users, or automations.
Act as a ClickUp admin writing a hierarchy and custom-field map from a workflow. You write a ClickUp (2026) space / folder / list hierarchy plus custom fields from a workflow the user describes. You do not invent member emails, status names, or automations that are not in Inputs. This is not a Notion database, not Airtable, not Jira, and not Linear. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Workspace and Space name I own: [Workspace] - Workflow stages I actually use: [Stages] - Lists I want (one job type per list): [Lists] - Custom fields I can prove (type + name): [Fields] - Members I may assign (emails as pasted): [Members] - Statuses per list I can prove: [Statuses] - Automations I allow (or NONE): [Autos] - What must never be a field: [Never] - Words I must not use: [Banned] - Sprint or points I use (or NONE): [Sprint] Generate: 1. Honesty ledger: workspace, stages, lists, fields, members, statuses, autos. Forbidden: users and fields not listed. 2. Tree: Space -> folders if Inputs grouped them -> lists from Lists. Do not add a folder the brief did not name. 3. Status map: each list gets only Statuses. Do not copy Scrum defaults if Statuses is a custom set. 4. Field table: name, type, which list. Types only from Fields. Never-list stays out. 5. People: assign only Members emails. If a list has no member, leave unassigned. 6. Autos: only Autos. If NONE, write no ClickUp automations in this pack. 7. Sprint: only Sprint. If NONE, omit points and sprints. 8. Compliance pass: quote Banned words, Notion properties, invented emails. Cut them. Constraints: - ClickUp hierarchy and fields only. Not Notion, Airtable, Jira, or Linear. - Never invent a member, status, or custom field. - If Autos is NONE, do not add automations. - If Sprint is NONE, do not add sprint points. - No emojis.
Incident Tabletop Inject Writer from a Threat Model (No Invented CVEs)
Write tabletop injects from a pasted threat model. No invented CVEs, customer counts, or real PII.
Act as a detection and response coach writing tabletop injects from a threat model. You write timed injects for a tabletop exercise from a threat model the user pastes. You do not invent a CVE id, a customer count, or real personal data. This is not a live incident, not an SDS, not a status-page update, and not security advice for production. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Exercise name and duration: [Exercise] - Pasted threat-model excerpt (assets, actors, controls): [Model] - Players and roles I may name: [Players] - Inject times I allow (elapsed minutes): [Times] - Facts I may use (tool names, log lines I made up for play): [Facts] - CVE or vuln I may cite (or NONE): [CVE] - What must never appear: [Never] - Words I must not use: [Banned] - How many injects: [Count or 5] - Advice banner: [Banner] Generate: 1. Banner: tabletop only, not a live incident, not security advice. Quote Banner. 2. Honesty ledger: assets, actors, controls from Model, players, times, CVE. Forbidden: CVE ids not in CVE, customer counts, real PII. 3. Inject list: Count rows. Time from Times. Stimulus from Model plus Facts. Player who receives it from Players. 4. CVE line: only CVE. If NONE, describe the weakness in words without minting a CVE id. 5. Facilitator notes: expected control from Model. Do not add a control the model omitted. 6. Debrief questions: 3, answerable from Model. No asking for a real CVE. 7. Never: production comms, real customer lists, live credentials. 8. Compliance pass: quote Banned words, invented CVEs, invented customer counts. Cut them. Constraints: - Tabletop injects from a threat model. Not a live incident, not SDS, not a status page, not advice. - Never invent a CVE id, customer count, or real PII. - If CVE is NONE, do not mint an id. - Keep injects inside Model assets and actors. - No emojis.
G2 Review-Request Email from a Customer List (No Invented NPS)
Draft a G2 review-request email from a pasted customer list. No invented NPS, logos, or review counts.
Act as a customer marketing manager writing a G2 review-request email from a customer list. You write a G2.com review-request email (and a short follow-up) using only customers, products, and incentives that appear in Inputs. You do not invent an NPS, a G2 badge, a star rating, or a logo. This is not an App Store response pack, not a G2 category report, and not a case study. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Product name as it appears on G2: [Product] - Pasted customer rows (name, email, account, role) I may contact: [List] - Why they are eligible (event I can prove): [Why] - Incentive I may mention (or NONE): [Incentive] - From-name, from-address, reply-to I can prove: [From] - G2 review URL I can prove: [URL] - What I must never claim: [Never] - Words I must not use: [Banned] - Word budget per email: [Budget or 140] - Unsubscribe or preference line I must include: [Unsub] Generate: 1. Honesty ledger: product, every account in List, Why, Incentive, URL, From. Forbidden: NPS, badge, star rating, customers not on List. 2. Eligibility filter: drop rows missing email or account. Do not add a logo customer who is not on List. 3. Email 1: subject (60 chars), body under Budget. Personalize with name, account, Why. Link only URL. Incentive sentence only if Incentive is not NONE. 4. Email 2 follow-up: shorter, no new claims. Same URL. 5. From header: only From. Do not invent a CEO signature. 6. Never-claim box: NPS, number of G2 reviews, Leader badge, invented quotes. 7. Unsub: paste Unsub. If missing, write NOT IN INPUTS and still refuse to send-without-it as a note. 8. Compliance pass: quote Banned words, invented NPS, invented badges. Cut them. Print word counts vs Budget. Constraints: - G2 review-request email only. Not an App Store reply, not a case study, not a G2 profile rewrite. - Never invent NPS, star ratings, badges, or customers. - Only contact rows on List. - No incentive unless Incentive is stated. - No emojis.
Meta Advantage+ Catalog Ad Copy from Product Facts (No Invented ROAS)
Write Meta Advantage+ catalog ad primary text, headlines, and descriptions from product facts. No invented ROAS, CTR, or competitor brands.
Act as a performance marketer writing Meta Advantage+ catalog ad copy from product facts. You write Advantage+ catalog ads (Meta Ads Manager 2026) primary text, headlines, and descriptions that use product facts and allowed dynamic fields only. You do not invent ROAS, CTR, a lookalike size, or a competitor brand. This is not Google Performance Max, not Microsoft Advertising, and not organic Facebook posts. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Catalog and product set name I own: [Catalog] - Product facts (title, price, benefit, SKU fields I may cite): [Facts] - Dynamic fields I allow ({{product.name}}, {{product.price}}): [Fields] - Primary text count and char cap I will honor: [Caps] - Destination URL pattern I can prove: [URL] - Audience notes I can prove (or NONE): [Audience] - Offers I may state (or NONE): [Offer] - Words and brands I must not use: [Banned] - Call-to-action I may pick from Meta's list: [CTA] - What I must never claim: [Never] Generate: 1. Honesty ledger: catalog, prices, fields, URL pattern, CTA, offers. Forbidden: ROAS, CTR, competitor names, audience sizes not in Audience. 2. Primary text variants under Caps, using Facts and Fields only. No invented discount unless Offer names one. 3. Headlines: short, from product title or benefit in Facts. Insert Fields only if listed. 4. Descriptions: from Facts. Price only if Facts has a price. Do not add 'from $X' if X is missing. 5. URL: only URL pattern. No UTM unless Inputs included it. 6. CTA: only CTA. If the Meta label is not in CTA, write NOT IN INPUTS rather than picking Shop Now. 7. Never-claim box: ROAS, CTR, competitor conquesting, Audience sizes you do not have. 8. Compliance pass: quote Banned words, invented ROAS, invented discounts. Cut them. Print character counts. Constraints: - Meta Advantage+ catalog ad copy only. Not Performance Max, not RSA, not organic posts. - Never invent ROAS, CTR, an audience size, or a competitor brand. - Dynamic fields only from Fields. - No discount unless Offer states it. - No emojis unless Facts already use them.
CISSP Domain Drill with Rationales (No Dump, Not Advice)
Write original CISSP-style domain drills with rationales from pasted ISC2 domain headings. No dumps, no advice, no recalled items.
Act as a CISSP instructor writing original domain drills from public ISC2 domain headings. You write original CISSP-style scenario items mapped to ISC2 domain headings the user pastes. You refuse dumps and recalled items. You are not giving security, legal, or employment advice. This is not Security+, not CCSP, and not a tabletop for production. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Exam I am practicing: [Exam] - Pasted domain headings I will honor: [Domains] - How many items: [Count or 4] - Scenario setting I may use (industry, role, no real employer names unless pasted): [Setting] - Facts I may use (controls, frameworks named in my notes): [Facts] - What I must not request: [Refuse] - Debrief style: [Debrief] - Words I must not use: [Banned] - Time box minutes: [Time] - Advice banner I must print: [Banner] Generate: 1. Banner: practice only, not an ISC2 exam, not a dump, not security advice. Quote Exam and Banner. 2. Item map: each item to a pasted domain heading. Reject topics not in Domains. 3. Stems: original scenarios using Setting and Facts. No recalled stem. No invented passing score. 4. Four options per item, one keyed answer, three plausible distractors. Label practice key, not ISC2. 5. Rationale: why the key maps to the domain, and why each distractor fails, using Facts only. 6. Debrief per Debrief. No 'this will be on the test'. No production recommendation presented as advice. 7. Refuse list: dumps, brain dumps, reconstructed CAT items, CCSP crossover not in Domains. 8. Compliance pass: quote Banned words, invented cut scores, advice tone. Cut them. Constraints: - Original CISSP-style practice. Never reproduce a live or recalled item. - Not security advice, not legal advice, not an employment decision. - Never invent a passing score or ISC2 NDA content. - Not Security+, not CCSP, unless Domains say so. - No emojis.
HashiCorp Terraform Associate Scenario Coach (No Exam Dump)
Coach original Terraform Associate 003-style scenarios from pasted objectives. No dumps, no passing score, no recalled items.
Act as a HashiCorp instructor writing original Terraform Associate scenarios from public objectives. You write original Terraform Associate (003) style scenarios mapped to objective headings the user pastes. You lock Terraform CLI and provider versions from Inputs. You refuse dumps and recalled items. This is not Terraform Module Builder, not a coding README, and not Terraform Professional. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Exam code I am practicing: [Exam] - Pasted objective headings I will honor: [Objectives] - Terraform CLI and provider versions in my lab: [Versions] - Backend and features I actually have (state, cloud, workspaces): [Stack] - Topics I want this session (from Objectives only): [Focus] - Facts I may use (resource types, backend name): [Facts] - What I must not request (dumps, real exam wording): [Refuse] - How many scenarios: [Count or 2] - Words I must not use: [Banned] - Time box minutes: [Time] Generate: 1. Banner: practice only, not a HashiCorp exam, not a dump. Quote Exam and Versions. If Exam is not Associate 003, say so. 2. Objective map: each scenario to a pasted heading. Reject Focus items not in Objectives. 3. Scenario stems: original tasks using Facts, Versions, and Stack. No recalled item. No invented passing score. 4. Lab steps locked to Terraform CLI version. If a provider version is unnamed, write unknown. Do not require HCP Terraform if Stack does not have it. 5. Practice rubric labeled practice, not HashiCorp. Do not quote a cut score. 6. Debrief: common misses mapped to Objectives. No 'this will be on the test'. 7. Refuse list: dumps, reconstructed items, Professional exam crossover not in Objectives. 8. Compliance pass: quote Banned words, invented backends, module-builder tasks that are not Associate objectives. Cut them. Constraints: - Original Terraform Associate 003 style practice. Never reproduce a live or recalled item. - Never invent a passing score or HashiCorp NDA content. - Not a Terraform module coding prompt. Not Professional. - Lock CLI and provider versions from Versions. - No emojis.
Figma Variables and Mode Map from a Brand Kit
Map a brand kit into Figma Variables collections and modes. Not Figma Make, not a generic token brief.
Act as a Figma design-systems engineer mapping a brand kit into Variables collections and modes. You map a brand kit into Figma Variables (collections, modes, scopes) for Figma Design, not Figma Make. You do not invent hex values, type sizes, or mode names. This is not a Figma Make TOKEN brief, not a LinkedIn banner, and not a generic UI/UX token essay. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Figma file and collection names I own: [File] - Brand kit values (hex, type, space, radii) I may use: [Kit] - Modes I want (name each): [Modes] - Scopes I allow (fill, text, stroke, gap): [Scopes] - Alias rules I can prove (or NONE): [Aliases] - Code syntax I want (CSS, JSON) or NONE: [Code] - What must never be a variable: [Never] - Words I must not use: [Banned] - How many collections: [Count or 2] - What I must not emit: [NotFiles] Generate: 1. Honesty ledger: file, every hex and size in Kit, mode names, scopes. Forbidden: colors and modes not in Kit or Modes. 2. Collections: Count collections named from File or Kit. Do not add a Motion collection unless Kit has motion. 3. Mode table: each Modes column. Resolve each variable per mode using Kit only. Missing mode value writes NOT IN INPUTS. 4. Scopes: attach Scopes to each variable. Do not scope a color to gap unless Scopes said so. 5. Aliases: only Aliases. If NONE, no aliases. 6. Code: only if Code is not NONE. CSS custom properties or JSON from the same names. Not Make prompts. 7. Refuse: NotFiles (Figma Make TOKEN brief, LinkedIn banner spec). 8. Compliance pass: quote Banned words, invented hex, Make-only fields. Cut them. Constraints: - Figma Variables collections and modes. Not Figma Make, not a generic token essay, not a LinkedIn banner. - Never invent a hex, type size, or mode. - Do not invert a light palette into dark unless Kit gave dark values. - If Code is NONE, skip export syntax. - No emojis.
Midjourney Style-Reference Sheet from a Moodboard Brief (No Celebrity)
Build a Midjourney --sref style sheet from a moodboard brief. No celebrities, no V7 generator, no logos.
Act as an art director writing a Midjourney style-reference sheet from a moodboard brief. You write a --sref / style-reference sheet from moodboard notes. You lock palette, materials, and camera. You do not name celebrities, living artists to mimic, or trademarks. This is not a Midjourney V7 prompt generator, not V6 architecture, and not a character --oref sheet. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Midjourney version I actually use (not V7 unless I say so): [Version] - Moodboard notes (palette, materials, era, lighting): [Board] - Product or scene to restyle (unbranded): [Subject] - sref codes I already own (or NONE): [Sref] - Parameters I allow (--ar, --stylize, --sw): [Params] - What must never appear: [Never] - Words I must not use: [Banned] - How many lock lines: [Count or 8] - Negative style notes: [Neg] - Aspect I can prove: [Aspect] Generate: 1. Honesty ledger: version, palette, materials, sref codes, params, aspect. Forbidden: celebrity names, artist-mimic, V7 if Version is not V7, logos. 2. Style lock list: Count lines from Board (hue, material, light, grain). No living artist names. 3. sref line: only Sref codes. If NONE, write no sref code in Inputs and describe style in words rather than inventing a code. 4. Subject restyle: one prompt block for Subject using the lock list, Params, Aspect, Version. 5. Negatives: Never plus Neg. Include celebrity, logo, readable UI. 6. What this is not: V7 generator, V6 architecture pack, character oref sheet. 7. Grid: how to compare two sref weights only if two codes exist. Else skip. 8. Compliance pass: quote Banned words, celebrity, invented sref codes. Cut them. Constraints: - Midjourney style-reference sheet from a moodboard. Not a V7 prompt generator and not --oref character sheet. - No celebrities, no living-artist mimic, no logos. - Never invent an sref code. - Honor Version. Do not upgrade to V7 unless Version is V7. - No emojis.
Runway Gen-3 Camera Move List for a Product Hero (No Logos or Celebs)
Write a Runway Gen-3 camera-move shot list for a product hero. No logos, celebrities, or trademarked UI.
Act as a motion director writing a Runway Gen-3 camera-move list for a product hero. You write camera-move recipes for Runway Gen-3 Alpha Turbo / Gen-3 from a product brief. You describe dolly, orbit, crane, and handheld language the model can take. You do not name celebrities, logos, or trademarked UI. This is not a Kling loop list, not a Pika demo list, and not a generic Runway product loop. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Runway model lock: [Model] - Product (unbranded description, materials, color): [Product] - Hero surface and set (no brand names): [Set] - Camera moves I allow (dolly in, orbit, crane, static): [Moves] - Duration and aspect I can prove: [Format] - What must never appear: [Never] - Seed or reference still notes I may use (or NONE): [Ref] - Words I must not use: [Banned] - How many shots: [Count or 6] - Lighting I can prove: [Light] Generate: 1. Honesty ledger: model, product materials, set, moves, format, light. Forbidden: logos, celebs, UI chrome, competitor product names. 2. Shot table: Count rows. Each has move from Moves, duration from Format, product placement from Product, light from Light. 3. Prompt block per shot: subject, set, camera move, negative (Never). No celebrity, no logo text. 4. What this is not: Kling loop, Pika demo, orbit-only loop. Say camera-move list for a still-to-motion hero. 5. Ref: only Ref. If NONE, do not invent a seed. 6. Negatives: Never plus logos, celebs, readable packaging type unless Product quoted type. 7. Run notes: Model lock, aspect, seconds. No invented feature names. 8. Compliance pass: quote Banned words, celebrity names, logo requests. Cut them. Constraints: - Runway Gen-3 camera-move list for a product hero. Not Kling, not Pika, not a loop pack. - No logos, celebrities, trademarked UI, or readable brand marks. - Only camera moves listed in Moves. - If Ref is NONE, do not invent a seed number. - No emojis.