A prompt from OpenAI's ChatGPT use-case library for a team facing a decision. You list the research, planning documents, market context, stakeholder comments, budget limits and criteria to use, and it writes a short memo with a recommendation, evidence, alternatives, costs, risks, open questions and links. It keeps outside research apart from internal evidence and does not make or share the decision.
I'm deciding whether [team or company] should [decision].
Use these sources:
- prior recaps and research: [files or links]
- planning docs, models, and dashboards: [files or links]
- audience, account, or market context: [files or links]
- stakeholder comments and meeting notes: [files, channels, or thread links]
- budget guardrails: [files or links]
- decision criteria: [criteria]
If I ask for web research, keep external findings separate from internal evidence. Write a concise decision memo with a recommendation, supporting evidence, alternatives and tradeoffs, costs, risks, a decision log, open questions, owner confirmations, and source links.
If the audience needs a pre-read, organize the memo and supporting analysis into a review-ready packet while keeping the memo as the source of truth. Flag assumptions and do not make the decision or share the output.
Prepare a meeting brief from your calendar, docs and threadsA prompt from OpenAI's ChatGPT use-case library that prepares you for a meeting using only the calendar event, documents, Slack threads and Gmail messages you name. It first lists what it can access, then gives the objective, attendee background, facts backed by sources, a likely agenda, open questions, follow-ups you may owe and a notes template. Claims it cannot support go in a separate gaps section, and it changes or sends nothing until you approve. It needs access to the calendar, documents and messages you name.Personal assistant
Diagnose why a sales deal has stalledOpenAI's starter prompt for working out why a deal stopped moving, using stage history, activity, call transcripts, emails, internal threads and procurement or security notes. It asks for the likely blocker with evidence, earlier attempts, dependencies, the escalation path, the next move with the customer, and missing context. It keeps sourced facts apart from inference and only drafts follow-ups, without sending them or editing the CRM.Sales
Map who agrees and disagrees across a stack of research papersA prompt from Anthropic's Claude use-case library for when you have read many papers and lost track of how they relate. It asks Claude to group the papers by argument, show where groups disagree and, for each group, summarise what it offers and what it overlooks. Attach the papers or your notes first, and treat the map as one reading to test against your own. In Claude the answer can appear as an interactive map; other assistants reply in text.Research
Analyze a research paper's findings and rigorThis prompt reads a research paper and returns a structured review with a summary, authors and organizations, key findings, and a study overview. It checks study quality items such as design, sample size, p-value, effect size, and reproducibility, rates novelty, rigor, and empiricism from 1 to 10 with a chart, and notes conflicts of interest. It ends with a letter grade and a short summary statement, and the grading leans critical.Research
Find research gaps from the papers you collectThis prompt acts as a research analyst that first asks for your topic, goal, collected papers or abstracts, academic level, and constraints. It then works through six steps: literature analysis, comparison of studies, possible gaps with evidence, a confidence rating for each gap, a feasibility check, and a ranked recommendation with a research plan. It forbids invented papers or citations and treats every gap as provisional until checked against the literature.Research
Group product feedback from Slack, GitHub, Linear and Drive into themesOpenAI's starter prompt for reviewing feedback on a product area over a set period, drawing on connected apps (Slack, Linear, GitHub, Google Drive) along with surveys, support tickets and research. It sorts the input into themes backed by evidence, says what needs attention and saves the findings to a Google document or spreadsheet. It may draft follow-ups but must not post or send them.Research
Prepare a meeting brief from your calendar, docs and threads
Diagnose why a sales deal has stalled
Map who agrees and disagrees across a stack of research papers
Analyze a research paper's findings and rigor
Find research gaps from the papers you collect
Group product feedback from Slack, GitHub, Linear and Drive into themes