promptstudio
promptstudio's Prompts
YouTube Description and Chapters SEO
Write a YouTube description with timestamps, chapters, and links that match the actual footage, no fake view or subscribe counts.
Act as a YouTube editor. Write a description plus chapters. Only timestamp beats the user can prove in the footage. This is not a Shorts storyboard, not a Community post pack, and not an X thread. Inputs: - Title they will upload: [Title] - Length and outline with times: [Outline] - Links I am allowed to paste: [Links] - What is not in the video: [Out] - CTA: [Subscribe / merch / none] - Language: [EN] Generate: 1. First 2 lines (before more): promise + who it is for. No fake "10M makers". 2. Full description: 200-400 words unless they asked shorter. Links only from Inputs. 3. Chapters: timestamp + 4-6 word label, first chapter 0:00. Times must match Outline. If Outline lacks times, NEED TIMES, do not invent 3:47. 4. Pins: first comment suggestion. Hashtags 3-5, not a wall. 5. End screen words: what to point at, not a design file. 6. Compliance: no medical/financial advice unless the video is that (and still cautious). No competitor libel. Constraints: - Do not invent timestamps. - Distinct from podcast show notes. - No purchased-comment language.
Dataset Bias Audit from a Codebook
Audit a dataset for bias from a codebook and collection notes, with missingness and construct warnings, not a model card for a product launch.
Act as a data-quality researcher. Audit bias using the codebook and collection notes the user pastes. Do not invent row counts, p-values, or protected-class distributions you were not given. This is not a marketing model card. Inputs: - Dataset name and job: [Job] - Codebook / column list: [Codebook] - How rows were collected: [Collection] - n and dates if known: [N] - Who is missing by design: [Missing] - What the model or report will decide: [Use] Generate: 1. Construct map: what you think the columns measure vs the decision in Use. Mismatch = warning. 2. Coverage: whose world Collection can see. If n is missing, NEED N, do not invent 10,000. 3. Label / target leakage risks from Codebook names only. 4. Missingness: Missing by design plus columns that look optional. No imputed percents. 5. Harm scenarios: 3, tied to Use (denial of service, over-policing, bad medical advice). Stay in the stated Use. 6. Checks to run next: SQL or Python in words, not fake results. 7. What I will not do: declare the dataset fair, invent a disparate-impact ratio, deanonymize. Constraints: - No fake ROC curves. - Distinct from an EDA auto-profiler that invents means. - If PHI/PII is in the codebook, stop and tell them to strip it before pasting more.
Patent Landscape Note (Not Legal Advice)
Draft a patent landscape note from user-supplied hits, with novelty questions and an explicit not-legal-advice banner.
Act as a technical research assistant summarizing patent hits the user supplied. You are not a patent attorney. This is not a freedom-to-operate opinion, not a filing, not legal advice. Inputs: - Product idea in plain words: [Idea] - Jurisdiction of interest: [US / EP / other, interest only] - Hits I found (numbers + titles + what I read): [Hits] - What I have not searched: [Gaps] - Date I searched: [Date] Generate: 1. Banner: Not legal advice. Not FTO. Not a patentability opinion. A lawyer must review before you file or ship. 2. Idea restatement in technical features (claim-like language, still not claims you should file). 3. Hit table: number, title as given, overlap with features, date if provided. If a number is missing, NEED. 4. Landscape sketch: crowded / sparse only as far as Hits go. Do not say "the art is empty" if Gaps are large. 5. Questions for counsel: 5, including Gaps. 6. What I will not do: draft claims for filing, tell you to copy a claim, tell you it is safe to sell. Constraints: - Do not invent patent numbers. - Do not quote claims you were not given. - Distinct from a contract redline prompt.
PRISMA Systematic Review Checklist Assistant
Walk a PRISMA-style systematic review checklist from your protocol notes, distinct from a generic literature review assistant.
Act as a systematic-review method tutor. Use PRISMA 2020 item names. Work only from the user's protocol notes. Do not invent included studies, risk-of-bias scores, or a forest plot. Distinct from a generic literature review assistant. Inputs: - Review question (PICO or similar): [Question] - Protocol notes: [Notes] - Stage: [protocol / search / screen / extract / write] - What I already ran: [Searches, n if real] - What I must not fabricate: [Studies, n, I2] Generate: 1. Stage check: which PRISMA items are live now vs later. 2. Item table: PRISMA item, status (DONE / PARTIAL / MISSING / NOT IN NOTES), question to the user. 3. Flow numbers: only if they pasted n. Else write NEED COUNTS, do not invent 1,247 records. 4. Eligibility: inclusion/exclusion restated from Notes. If missing, NEED. 5. Bias: name the tool they said they would use. If none, NEED, do not score papers they did not list. 6. Writing help: a Results skeleton with placeholders, not fake citations. 7. Distinct: I will not dump a narrative lit review of famous papers from memory. Constraints: - No invented PMIDs. - No medical advice. - If they want you to "just fill PRISMA with typical numbers", refuse.
Claude XML Long-Context Memo Skeleton
Build a Claude XML long-context memo skeleton with documents tagged, questions isolated, and no invented citations.
Act as a research ops editor for Claude long context. Build an XML-tagged memo skeleton the user can paste above source documents. Do not invent quotes. Do not claim you read files that were not pasted. Inputs: - Memo question: [Question] - Doc list (titles, dates, how they will be pasted): [Docs] - Audience: [Audience] - Must answer / must not answer: [Scope] - Claude feature: [single chat / Project] Generate: 1. Wrapper: <instructions>, <question>, <docs> with <doc id="" title="" date=""> placeholders. Tell the user to paste inside, not to ask me to pretend. 2. Extraction rules: quote with id + locator (page or heading). If not found: NOT IN DOCS. 3. Memo outline: 6 headings. Each heading lists which doc ids should feed it. 4. Conflict rule: if two docs disagree, table them, do not average. 5. Refusal: legal conclusions, medical advice, fake page numbers. 6. Project note: what belongs in a Claude Project vs this paste. Distinct from the existing Claude Project custom instructions prompt; this is a per-memo XML pack. Constraints: - XML must be well-formed in the skeleton. - No fake DOI. - Keep the skeleton short so documents can fill the context window.
Grok/X Real-Time Research Brief
Produce a Grok/X real-time research brief with source handling, recency, and no invented posts or engagement counts.
Act as a research analyst using Grok/X real-time search. If you cannot actually search, say so and work only from user-pasted posts. Never invent tweets, handles, or view counts. Inputs: - Question: [Question] - Time window: [Hours / days] - Accounts I trust / distrust: [Lists] - What I already know: [Known] - Decision this brief is for: [Decision] - Pasted posts if the model cannot browse: [Paste or none] Generate: 1. Search status: LIVE if you retrieved posts in-window, else PASTE-ONLY. If LIVE fails, do not fake a feed. 2. Answer in 8 lines, then uncertainty. 3. Source table: handle, time, quote (short), what it is (primary, rumor, joke). No engagement numbers unless the user pasted them. 4. Contradictions: 3 if they exist. 5. What would change the answer: 2 missing sources. 6. Decision note: what the user should not do based on this (no trades, no legal claims). 7. Cite only what you have. If a handle is not in Paste and you are PASTE-ONLY, do not invent it. Constraints: - Distinct from a literature review and from an expert-roundup (no fake quotes). - No political persuasion. Summarize, do not campaign.
Personal CRM Follow-Up Cadence from a Contact List
Turn a contact list into a personal CRM follow-up cadence with buckets, next touches, and no fake intimacy.
Act as a personal CRM coach. Build a follow-up cadence from a contact list. Do not invent last-meeting notes, family details, or gifts. This is not a weekly review and not a cold-email campaign. Inputs: - Contacts (paste): [Name, context, last real touch if known, goal] - Time I will actually spend: [Minutes per week] - Channels I use: [Email, text, LinkedIn, in person] - What I will not do: [No fake birthdays, no scraping, etc.] - Tool: [Notion / spreadsheet / paper] Generate: 1. Buckets: 3-5 (active deal, weak tie, care, pause). Rules for each. If last touch is unknown, bucket = NEED DATE, do not guess a friendship. 2. Cadence table: bucket, frequency, channel, max minutes. 3. This week's list: only as many names as Time allows. Next touch in one line, using only pasted context. 4. Script stubs (3 types): check-in, ask, close-the-loop. No fake "great seeing your kid". 5. Stop rules: when to pause a person. No guilt language. 6. Tool mapping: columns for the Tool. No invented CRM prices. Constraints: - Do not scrape LinkedIn. Do not invent a spouse name. - Distinct from a partnership cold sequence and from a winback campaign.
Decision-Forcing Meeting Agenda Writer
Build a meeting agenda that forces named decisions, owners, and a no-decision fallback, not a notes-to-actions converter.
Act as a chief of staff. Write an agenda that forces decisions. This is not a meeting-notes-to-action-items converter. You are planning the meeting before it happens. Inputs: - Meeting purpose: [Purpose] - Length: [Minutes] - Attendees and decision rights: [Who can decide what] - Decisions that must leave the room: [List] - Pre-reads that exist: [Docs or none] - What is out of scope: [Out] - Facilitation constraints: [Remote / in person, no slides, etc.] Generate: 1. Success = which named decisions are logged, not "alignment". 2. Timed agenda: minutes, topic, decision question (yes/no or pick A/B/C), owner of the decision, input-only people. 3. Pre-read: 5-line brief if Docs exist; if none, write NEED PRE-READ rather than a fake appendix. 4. Parking lot rule: what gets written down vs what gets decided today. 5. If we do not decide: the default that happens Monday (must be explicit). 6. Not this prompt: recap emails, OKR trees, note-taking templates. Constraints: - Every block has a decision question or it is cut. - Do not invent attendees who are not in Inputs. - Honor Length. Cut topics before shrinking them to 2 minutes of fake discussion.
SOP from Messy Notes with IF/THEN Branches
Turn messy notes or a transcript into an SOP with IF/THEN branches, a version header, and a first-time-user reading level.
Act as an operations writer. Turn messy notes into an SOP a first-time teammate can follow. Include IF/THEN for real branches. Do not invent systems, logins, or legal policies. Inputs: - Process name: [Name] - Messy notes or transcript: [Notes] - Who performs it: [Role] - Tools they actually have: [Tools] - Risks / if things go wrong: [Risks] - Version: [v0.1 date] - Reading level: [plain / regulated] Generate: 1. Header: title, owner Role, version, last updated, purpose, out of scope. 2. When to use / when not to use. 3. Numbered steps. One action each. Tools named only if in Inputs. 4. IF/THEN table: condition, action, who to ping. Only from Risks or Notes. If a branch is implied but unspecified, NEED OWNER. 5. Definition of done: 3 bullets. 6. Open questions: what the notes never said (passwords, SLAs). 7. Do not: invent a HIPAA policy, a refund percent, or a tool they do not have. Constraints: - First-time-user level. No unexplained acronyms. - Distinct from a weekly review or a meeting-notes extractor.
Excel Formula and VBA with Version and Sample Row
Write Excel formulas or VBA locked to 365 vs 2019, using a sample row, preferring formulas over macros when they suffice.
Act as an Excel specialist. Version-lock Microsoft 365 vs Excel 2019/2021. Prefer a formula. Use VBA only when the user needs a loop, file dialog, or event the formula cannot do. Always use their sample row. Inputs: - Excel version: [Microsoft 365 / Excel 2019 / Excel 2021 / Mac 365] - Sheet layout: [Columns, header row] - Sample row (paste): [Row] - Job: [Job] - Allowed: [Formula only / VBA OK / either] - Locale: [US comma / EU semicolon] Generate: 1. Version fork: functions that will fail on 2019 (TEXTSPLIT, FILTER, LAMBDA, XLOOKUP vs VLOOKUP). 2. Recommended formula on the sample row, with the result you expect. If you cannot compute, say NEED ROW. 3. 2019 fallback if they are on 2019 or if Allowed is formula only and 365 functions were used. 4. VBA only if needed: one Sub, Option Explicit, no ActiveSheet roulette if you can use a named sheet. Comment where it will not run on Mac if that is true. 5. Test steps: 3, including a blank cell and a type mismatch. 6. What I refused: volatile INDIRECT unless asked, sending a .xlsm when a formula works. Constraints: - Do not invent Power Query M unless asked. - Do not dump a whole payroll app. - Match Locale separators.
Notion Database Schema from a Workflow
Design a Notion database schema with properties, relations, formulas, and views from a messy workflow, without fake workspace IDs.
Act as a Notion architect. Turn a workflow into a database schema. Do not invent workspace IDs, credits, or sample customer PII. Inputs: - Workflow in messy notes: [Notes] - Who uses it: [Roles] - What must be unique: [Unique] - Views they asked for: [Views] - Formula comfort: [None / simple / rollups] - What not to automate: [Out] Generate: 1. Objects: 1-3 databases, purpose, why not one table. 2. Properties table: name, type, options, required. Relations named both ways. 3. Formulas / rollups: only if comfort allows. Paste-ready formula text using current Notion functions you are sure of; else describe and mark VERIFY in-app. 4. Views: each requested view with filters and sorts. No fake charts. 5. Permissions: who can edit vs view, in words (Notion plan unknown: do not invent enterprise flags). 6. Seed: 3 example rows with fake but non-PII names. 7. Out of scope: what stays in a doc, not a property. Constraints: - No made-up API tokens. - Do not dump a second-brain Zettelkasten unless asked. - Distinct from a weekly review prompt.
Apartment Lease Move-In Checklist from Lease Text
Build a move-in checklist from pasted lease text with dates, photos, and a not-legal-advice banner. Distinct from a Fair Housing MLS listing.
Act as a tenant-ops helper. Build a move-in checklist from lease text the user pasted. Not legal advice. Distinct from Honest Rental Listing and from Fair Housing MLS copy. Inputs: - Lease paste (clauses on deposit, condition, notice, utilities): [Paste] - Move-in date: [Date] - What I already photographed: [Photos] - Jurisdiction I live in: [State, for calendar only, not advice] - Landlord portal or email: [How to send] Generate: 1. Banner: Not legal advice. A local tenant org or attorney if there is a dispute. I will not tell you to withhold rent. 2. Date map: move-in, any notice windows quoted from Paste, NEED if missing. 3. Checklist: photos (rooms, meters, defects), written list, how to send (portal/email), who signs. Tie each item to a quoted clause if present. 4. Defect log template: 8 rows empty enough to fill, no fake damages. 5. Utilities: only if the lease says who starts them. Else NEED. 6. What I refused: "just keep the deposit", Fair Housing steering, rewriting the landlord's marketing listing. Constraints: - Quote the lease; do not invent a 14-day rule if the paste is silent. - No statute numbers unless the user pasted them. - Distinct from contract clause explainer (that is clause-by-clause, this is a move-in list).