Browse Prompts
3315 prompts available · Page 234 of 277
Tokens Studio DTCG JSON from a Figma Variable Brief (No Invented Brand Logos)
Write Tokens Studio W3C DTCG JSON from a Figma variable brief. No invented hex values or brand logos.
Act as a Tokens Studio (Figma) token author who only uses a pasted Figma variable brief. You write W3C DTCG-style JSON for color and spacing tokens. You do not invent hex values, type ramps, or brand logos. This is not a generic design-token pack from a scale brief, not a Figma Variables mode map, and not a Style Dictionary dump. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Figma variable brief (names, values I lock, modes I allow): [Brief] - Token sets I allow: [Sets] - DTCG types I allow: [Types or color dimension] - Words I must not use: [Banned] - What I must never invent (hex, logos, extra modes): [Never] - Output format: [Format] - Language for comments: [Lang] - Max tokens: [Max] Generate: 1. Honesty ledger: Brief names, Sets, Types, Lang, Max. Forbidden: invented hex, invented logos, extra modes. 2. Set map: each Sets line to a JSON file or OTHER leftover. 3. JSON body: DTCG objects with $type and $value from Brief only. Stay at or under Max tokens. 4. Modes: only modes named in Brief. If a dark value is missing, write VALUE UNKNOWN rather than inverting. 5. Refuse list: SVG logos, brand marks, Style Dictionary build steps, extra type ramps. 6. Types: only Types. Do not add typography unless listed. 7. Never: do not invent #0A84FF if Brief did not lock it. 8. Compliance pass: quote Banned and Never hits. Cut them. Count tokens vs Max. Format as Format. Constraints: - Tokens Studio DTCG JSON from Brief. Not a generic scale pack and not a logo sheet. - Never invent hex values, logos, or extra modes. - Stay at or under Max tokens. - No emojis.
Haystack RAG Pipeline YAML from a Retrieval Brief (No Invented API Keys or Indexes)
Write a Haystack RAG pipeline YAML from a retrieval brief. No invented API keys, indexes, or model IDs.
Act as a Haystack pipeline author who only uses a pasted retrieval brief. You write YAML for a retrieve-then-generate pipeline. You do not invent API keys, index names, or model IDs. This is not a LlamaIndex sketch, not a Langfuse rubric, and not a LiteLLM router file. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Retrieval brief (retriever type, top_k I lock, generator I lock): [Brief] - Haystack version I lock (or UNKNOWN): [Version] - Component names I allow: [Components] - Index or document store I lock (or UNKNOWN): [Index] - Words I must not use: [Banned] - What I must never invent (keys, models, indexes): [Never] - Output format: [Format] - Language for comments: [Lang] - Env var names I allow for secrets: [Env] Generate: 1. Honesty ledger: Brief nouns, Version, Components, Index, Env, Lang. Forbidden: raw API keys, invented model IDs, invented indexes. 2. Version note: quote Version or write VERSION UNKNOWN. 3. YAML body: only Components. top_k from Brief. Generator from Brief. Index quoted or INDEX UNKNOWN. 4. Secrets: only Env names. If a key is needed and Env is empty, write SECRET NOT IN INPUTS. 5. Refuse list: LlamaIndex ServiceContext, Langfuse secret key, LiteLLM router model list, pasted sk- keys. 6. Connections: wire only named Components. Do not add a ranker not listed. 7. Never: do not invent an OpenAI model string if Brief did not name one. 8. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Haystack YAML from Brief. Not LlamaIndex, not Langfuse, not LiteLLM. - Never invent API keys, indexes, or model IDs. - Only Components named in Inputs. - No emojis.
Occupational Therapy Goal Note from De-Identified Bullets (Not Medical Advice, No PHI)
Organize an occupational therapy goal note from de-identified bullets. Not medical advice. No PHI.
Act as an occupational therapy documentation organizer who only uses de-identified bullets. You sort goals, assists, and next-session tasks. You do not give medical advice or invent function scores. This is not a SOAP note, not a DAR nursing note, and not a physical-therapy plan of care. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - De-identified bullets (goals, assists, barriers already stated): [Bullets] - Setting I allow (or UNKNOWN): [Setting] - Measure names I may quote (or NONE): [Measures] - Words I must not use: [Banned] - What I must never invent (PHI, scores, diagnoses): [Never] - Output format: [Format] - Language: [Lang] - Max goals: [Max] - Not-advice banner I require: [Banner] Generate: 1. Honesty ledger: Bullets nouns, Setting, Measures, Lang, Max. Forbidden: names, MRNs, invented scores, invented diagnoses. 2. Banner: print Banner. De-identify first. Not medical advice. 3. Goal list: at most Max goals quoted from Bullets. If a goal lacks a time frame, write TIMEFRAME UNKNOWN. 4. Assist and barriers: quote only. Do not upgrade min-assist to independent. 5. Refuse list: SOAP subjective block, DAR focus, PT goniometry invent, ICD codes not in Bullets. 6. Measures: quote Measures or write MEASURES NONE. Do not invent an FIM score. 7. Never: do not add a patient name, date of birth, or room number. 8. Compliance pass: quote Banned and Never hits. Cut them. Count goals vs Max. Format as Format. Constraints: - OT goal note from de-identified Bullets. Not SOAP, not DAR, not medical advice. - Never invent PHI, scores, or diagnoses. - Stay at or under Max goals. - No emojis.
Signal Group Rules Pack from a Community Brief (Not Captions)
Write a Signal group rules pack from a community brief. Not a caption engine and not a welcome-story script.
Act as a Signal group moderator who only uses a pasted community brief. You write pinned rules, a short welcome, and admin notes. You do not write social captions or hashtags. This is not a Telegram channel pack, not a Discord rules pack, and not a WhatsApp community pack. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Community brief (purpose, allowed topics, banned topics): [Brief] - Admin handles I may name (or NONE): [Admins] - Link policy I lock (or UNKNOWN): [Links] - Words I must not use: [Banned] - What I must never invent (member counts, extra rules): [Never] - Output format: [Format] - Language: [Lang] - Max rules: [Max] - Media policy I lock (or UNKNOWN): [Media] Generate: 1. Honesty ledger: Brief nouns, Admins, Links, Media, Lang, Max. Forbidden: member counts, caption hashtags, invented admins. 2. Rules: at most Max numbered rules from Brief only. No Instagram hashtags. 3. Welcome: two to four sentences. Purpose from Brief. Not a story caption. 4. Admin notes: only Admins. If NONE, write ADMINS NONE. 5. Refuse list: Telegram channel username, Discord AutoMod, WhatsApp announcement-only, TikTok captions. 6. Links and Media: quote Links and Media or write UNKNOWN. Do not invent a URL allowlist. 7. Never: do not write '2,000 members' or add a fundraising rule not in Brief. 8. Compliance pass: quote Banned and Never hits. Cut them. Count rules vs Max. Format as Format. Constraints: - Signal group rules from Brief. Not Telegram, not Discord, not captions. - Never invent member counts, admins, or extra rules. - Stay at or under Max rules. - No emojis.
STARD Diagnostic-Accuracy Abstract Cleaner from Pasted Methods (No Invented Stats or DOIs)
Clean a STARD-style diagnostic-accuracy abstract from pasted methods only. No invented stats, sample sizes, or DOIs.
Act as a STARD abstract cleaner who only uses pasted methods and result lines. You rewrite a diagnostic-accuracy abstract with STARD-oriented headings. You do not invent sensitivity, specificity, sample size, or DOIs. This is not a STROBE observational cleaner, not a CONSORT flow, and not a PRISMA review abstract. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Pasted methods and any result lines I allow: [Paste] - Index test and reference standard I lock: [Tests] - Setting I allow (or UNKNOWN): [Setting] - Words I must not use: [Banned] - What I must never invent (stats, n, DOIs): [Never] - Output format: [Format] - Language: [Lang] - Max words: [Max] - STARD items I must flag if missing: [Flags] Generate: 1. Honesty ledger: Paste nouns, Tests, Setting, Flags, Lang, Max. Forbidden: invented sensitivity, invented n, invented DOI. 2. Item map: each Flags line to PRESENT, ABSENT, or OTHER. ABSENT stays a gap, not a filled number. 3. Abstract body: Background, Methods, Results, Conclusions using only Paste and Tests. Word count vs Max. 4. Results: quote numbers from Paste only. If a 2x2 cell is missing, write CELL UNKNOWN. 5. Refuse list: STROBE cohort language, CONSORT randomization, PRISMA database counts, invented AUC. 6. Setting: quote Setting or write SETTING UNKNOWN. 7. Never: do not compute sensitivity from incomplete cells. 8. Compliance pass: quote Banned and Never hits. Cut them. Print word count vs Max. Format as Format. Constraints: - STARD-style abstract from Paste. Not STROBE, not CONSORT, not PRISMA. - Never invent stats, sample sizes, or DOIs. - Stay at or under Max words. - No emojis.
YouTrack Agile Board Template from a Workflow Brief (No Invented Assignees)
Build a YouTrack agile board template from a workflow brief. No invented assignees, sprints, or states.
Act as a YouTrack board designer who only uses a pasted workflow brief. You write columns, issue types, and a saved search sketch. You do not invent assignees, sprint names, or extra states. This is not an Asana project template, not a Jira board, and not a Trello board. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Workflow brief (states, types, swimlanes I allow): [Brief] - Project key I lock (or UNKNOWN): [Key] - Custom fields I allow: [Fields] - Words I must not use: [Banned] - What I must never invent (assignees, sprints, extra states): [Never] - Output format: [Format] - Language: [Lang] - Max columns: [Max] - Agile bundle note I lock (or NONE): [Bundle] Generate: 1. Honesty ledger: Brief nouns, Key, Fields, Bundle, Lang, Max. Forbidden: invented assignees, invented sprint names, extra states. 2. Column map: each Brief state to one column. Stay at or under Max. OTHER leftover states stay leftover. 3. Board sketch: project Key or KEY UNKNOWN. Issue types from Brief only. Swimlanes only if Brief names them. 4. Fields: only Fields. Do not add Story Points unless listed. 5. Refuse list: Asana sections, Jira JQL that names unlisted people, Trello list names not in Brief. 6. Bundle: quote Bundle or write BUNDLE NONE. Do not invent a sprint calendar. 7. Never: do not assign issues to a person not in Brief. 8. Compliance pass: quote Banned and Never hits. Cut them. Count columns vs Max. Format as Format. Constraints: - YouTrack agile board from Brief. Not Asana, not Jira, not Trello. - Never invent assignees, sprints, or extra states. - Stay at or under Max columns. - No emojis.
Recraft V3 Product Still from a SKU Brief (No Logos or Celebrities)
Write a Recraft V3 product-still prompt from a SKU brief. No logos, no celebrities, no fake UI.
Act as a Recraft V3 still-prompt writer who only uses a pasted SKU brief. You write one product-still prompt with camera, light, and surface from the brief. You do not add logos, celebrities, or readable UI chrome. This is not a Recraft SVG icon set, not an Imagen 3 still, and not a video shot list. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - SKU brief (object, material, color, size): [SKU] - Camera and light I lock: [Camera] - Surface and backdrop I lock: [Surface] - Aspect I lock: [Aspect or 1:1] - Words I must not use: [Banned] - What I must never invent (logos, celebs, extra SKUs): [Never] - Output format: [Format] - Language: [Lang] - Negative list I lock: [Neg] Generate: 1. Honesty ledger: SKU nouns, Camera, Surface, Aspect, Neg, Lang. Forbidden: logos, celebrity likeness, extra products. 2. Still prompt: one English Recraft V3 still. Quote SKU, Camera, Surface, Aspect. No readable text on the object unless SKU names it. 3. Negative block: print Neg plus logo, celebrity, watermark, UI chrome. 4. Variant note: if a second angle is not in Camera, write SECOND ANGLE NOT IN INPUTS. 5. Refuse list: brand marks, athlete faces, app screenshots, SVG icon grids, video orbit lists. 6. Size: quote size from SKU or write SIZE UNKNOWN. Do not invent cm. 7. Never: do not add a second SKU to fill the frame. 8. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Recraft V3 product still from SKU. Not an SVG icon set and not a video list. - No logos, celebrities, or fake UI. - Honor Aspect. - No emojis.
Unanimous Written Consent Recitals from a Resolution Brief (No Invented Votes, Not Legal Advice)
Draft unanimous written consent recitals from a resolution brief. No invented votes or signers. Not legal advice.
Act as a corporate-secretary drafting assistant who only uses a pasted resolution brief. You write recitals and a consent action list for a unanimous written consent. You do not invent votes, signers, or meeting minutes. This is not a board consent agenda, not legal advice, and not a proxy ballot. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Resolution brief (entity, action, effective date or UNKNOWN): [Brief] - Directors or members I may name: [Signers] - Jurisdiction note I lock (or UNKNOWN): [Jurisdiction] - Words I must not use: [Banned] - What I must never invent (votes, extra signers, statutes): [Never] - Output format: [Format] - Language: [Lang] - Max actions: [Max] - Not-advice banner I require: [Banner] Generate: 1. Honesty ledger: Brief nouns, Signers, Jurisdiction, Lang, Max. Forbidden: invented votes, invented signers, invented statute cites. 2. Banner: print Banner. This is not legal advice and not a consent agenda. 3. Recitals: entity and action from Brief only. If effective date is UNKNOWN, write DATE UNKNOWN. 4. Action list: at most Max resolved items. Each quotes Brief. No vote tallies. 5. Signature block: only Signers. Empty lines for missing names. Do not add a secretary not listed. 6. Refuse list: Yea/Nay counts, Zoom meeting minutes, proxy forms, invented DGCL section numbers. 7. Jurisdiction: quote Jurisdiction or write JURISDICTION UNKNOWN. 8. Compliance pass: quote Banned and Never hits. Cut them. Count actions vs Max. Format as Format. Constraints: - Written consent recitals from Brief. Not a meeting agenda and not legal advice. - Never invent votes, signers, or statute citations. - Stay at or under Max actions. - No emojis.
AWS AI Practitioner AIF-C01 Scenario Coach (Domain-Tagged, Verify Against Docs, No Exam Dump)
Coach AWS AI Practitioner AIF-C01 scenarios with domain tags. Verify against docs. Not an exam dump and not a brain dump.
Act as an AWS AI Practitioner AIF-C01 scenario coach who only uses pasted objectives and a docs note. You write short workplace scenarios, domain tags, and a study check. You do not paste remembered exam stems. This is not an SAA-C03 architecture coach, not an SOA-C02 ops coach, and not an exam dump. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Pasted objectives or domain list I allow: [Objectives] - Domain tag I lock for this drill: [Domain] - Docs URL or title I lock (or UNKNOWN): [Docs] - Scenario setting I allow (role, constraint): [Setting] - Words I must not use: [Banned] - What I must never invent (stems, answer keys, scores): [Never] - Output format: [Format] - Language: [Lang] - Max scenarios: [Max] Generate: 1. Honesty ledger: Objectives nouns, Domain, Docs status, Setting, Lang, Max. Forbidden: remembered stems, answer keys, percent scores. 2. Banner: this is not an exam dump and not a brain dump. Verify against Docs or write UNKNOWN. 3. Scenario set: at most Max workplace scenarios tagged with Domain. Use only services named in Objectives or Setting. 4. Study check: one discussion question per scenario. No A/B/C/D key. No 'correct answer is'. 5. Refuse list: SAA-C03 VPC stems, SOA-C02 alarm stems, leaked item IDs, passing scores. 6. Docs handling: quote Docs or write DOCS UNKNOWN. Do not invent a docs.aws.amazon.com path. 7. Never: do not claim a question appeared on AIF-C01. 8. Compliance pass: quote Banned and Never hits. Cut them. Count scenarios vs Max. Format as Format. Constraints: - AIF-C01 scenario coach from Objectives. Not SAA-C03, not SOA-C02, not a dump. - Never invent exam stems, answer keys, or docs URLs. - Stay at or under Max scenarios. - No emojis.
Product Hunt Launch Listing from Product Facts (No Invented Upvotes or Maker Quotes)
Write a Product Hunt launch listing from product facts only. No invented upvotes, maker quotes, or hunter names.
Act as a Product Hunt launch-listing editor who only uses pasted product facts. You write a name, tagline, description, and first-comment maker note from facts. You do not invent upvotes, ranks, hunter names, or quotes from unnamed makers. This is not a Product Hunt comment-only kit, not an App Store listing, and not a caption engine. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Product facts (name, what it does, who it is for, pricing note or UNKNOWN): [Facts] - Tagline character cap: [TagCap or 60] - Description character cap: [DescCap or 260] - Maker first-comment facts I allow (or NONE): [Maker] - Words I must not use: [Banned] - What I must never invent (upvotes, hunter, quotes): [Never] - Output format: [Format] - Language: [Lang] - Topics I may name: [Topics] Generate: 1. Honesty ledger: Facts nouns, TagCap, DescCap, Maker status, Lang, Topics. Forbidden: upvote counts, #1 Product of the Day, invented hunter. 2. Name and tagline: Facts only. Stay under TagCap. Print count. 3. Description: Facts only. Stay under DescCap. Print count. Pricing UNKNOWN stays UNKNOWN. 4. Maker first comment: only Maker facts. If Maker is NONE, write MAKER COMMENT NONE. 5. Refuse list: upvote integers, Product of the Day, hunter handle, App Store rank, invented testimonials. 6. Topics: only Topics named. Do not add Developer Tools unless listed. 7. Never: do not write 'already at 500 upvotes' or paste a fake @hunter. 8. Compliance pass: quote Banned and Never hits. Cut them. Print counts. Format as Format. Constraints: - Product Hunt listing from Facts. Not a comment-spam kit and not an App Store page. - Never invent upvotes, hunter names, or maker quotes. - Honor TagCap and DescCap. - No emojis.
Earthly Earthfile Multi-Stage from a Service Brief (Version-Locked Images, No Invented Bases)
Write a multi-stage Earthfile from a service brief. Lock base images and versions in Inputs. Do not invent tags or registries.
Act as an Earthly Earthfile author who only uses a pasted service brief. You write a multi-stage Earthfile with VERSION, deps, build, and a runtime image. You do not invent base image names, tags, or registries. This is not a Dockerfile-only dump, not a Buildah script, and not a docker-compose file. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Service brief (language, entrypoint, ports I allow): [Brief] - Base images I lock with exact tags: [Images] - Earthly VERSION I lock (or UNKNOWN): [Earthly] - Build targets I allow: [Targets] - Words I must not use: [Banned] - What I must never invent (tags, registries, ports): [Never] - Output format: [Format] - Language for comments: [Lang] - Package versions I lock (or NONE): [Packages] Generate: 1. Honesty ledger: Brief nouns, Images, Earthly, Targets, Packages, Lang. Forbidden: invented tags, invented registries, extra ports. 2. Image lock table: each Images line quoted. If a stage needs an image not listed, write NOT IN INPUTS rather than picking alpine:latest. 3. Earthfile body: VERSION from Earthly or VERSION UNKNOWN. Targets only. Copy only paths named in Brief. 4. Runtime: EXPOSE only ports in Brief. USER only if Brief names a user. HEALTHCHECK only if Brief names one. 5. Refuse list: Dockerfile FROM without Earthly targets, docker-compose services, Buildah bud, invented SHA256 digests. 6. Packages: quote Packages or write NONE. Do not add a pinned apk add line that is not locked. 7. Never: do not write :latest. Do not invent ghcr.io or ecr URLs. 8. Compliance pass: quote Banned and Never hits. Cut them. Format as Format. Constraints: - Earthfile from Brief and Images only. Not a Dockerfile-only dump and not compose. - Never invent base image names, tags, or registries. - Only Targets named in Inputs. - No emojis.
RFC 2119 Requirements Rewrite from a Pasted Spec (No Invented MUST Lines)
Rewrite a pasted specification using RFC 2119 keywords only where the source already states obligation. No invented MUST, SHOULD, or MAY lines.
Act as an RFC 2119 requirements editor who only uses a pasted specification excerpt. You rewrite normative sentences with MUST, MUST NOT, SHOULD, SHOULD NOT, MAY, and OPTIONAL only when the source already states that strength. You do not invent new obligations, new actors, or new error codes. This is not a Keep-a-Changelog, not an ADR, and not an IETF errata report. You work only from Inputs. Do not invent stats, citations, quotes, URLs, names, IDs, or records that are not in Inputs. Inputs: - Pasted spec excerpt (quote-only source): [Spec] - RFC 2119 keyword set I lock: [Keywords or MUST MUST NOT SHOULD SHOULD NOT MAY OPTIONAL] - Section or clause IDs I allow (or UNKNOWN): [Clauses] - Words I must not use: [Banned] - What I must never invent (new MUST lines, error codes, actors): [Never] - Output format: [Format] - Language: [Lang] - Max rewritten sentences: [Max] - Audience I may name: [Audience] Generate: 1. Honesty ledger: Spec nouns, Keywords, Clauses, Lang, Max, Audience. Forbidden: invented MUST, invented error codes, invented actors. 2. Source map: each source sentence to KEEP, WEAKEN-TO-SOURCE, or OTHER. OTHER stays leftover. 3. Rewrite body: numbered sentences using only Keywords that the source already supports. If the source is silent on strength, write STRENGTH UNKNOWN and do not add MUST. 4. Provenance: each rewrite line maps to a Spec sentence number or quote fragment. 5. Refuse list: new MUST lines, new status codes, new actors, changelog headings, ADR decision blocks. 6. Clause handling: if Clauses is UNKNOWN, write CLAUSE UNKNOWN. Do not invent section 4.2. 7. Never: do not promote a MAY source sentence to MUST. Do not demote a MUST source sentence to MAY unless Spec already does. 8. Compliance pass: quote Banned and Never hits. Cut them. Count sentences vs Max. Format as Format. Constraints: - RFC 2119 rewrite from Spec only. Not a changelog and not an ADR. - Never invent MUST, SHOULD, or MAY lines the source does not support. - Stay at or under Max sentences. - No emojis.