Prompt Library

Here's everything on my mind right now, in no order:

{{brain dump}}

I have about {{hours}} hours of focused time today, plus {{fixed commitments}}.

1. Sort everything into: must do today / could do today / not today.
2. Pick THREE priorities for today. For each, say why it made the cut.
3. Make a realistic schedule for the focused time, with 15-minute buffers.
   Assume each task takes 1.5x longer than I'd guess.
4. For the "not today" list, suggest which items can be deleted,
   delegated or scheduled, and when.
5. Name the one task I'm probably avoiding, and suggest a five-minute
   first step for it.

I'm trying to understand {{concept}}. My background: {{what I already know}}.

Explain it three ways:
1. An analogy from everyday life ({{an area I know well, e.g. cooking, football}}).
   Then say where the analogy breaks down.
2. The mechanism: how it actually works, step by step, in plain language.
3. A worked example with real numbers or a concrete case.

Then give me one test question that checks whether I really understood it,
not just memorised it. Don't give the answer until I try.

Review the following {{language}} code. It's part of {{what the system does}}
and will run in {{environment, e.g. a busy web API, a nightly batch job}}.

"""
{{code}}
"""

Review it like a kind but picky senior engineer. Organise your comments as:

1. MUST FIX: bugs, security problems, data loss, crashes. For each: the line,
   what could go wrong, and a concrete fix.
2. SHOULD FIX: missing tests, unclear names, error handling, performance traps.
3. NITS: style and preference. Maximum three.

Then:
- Name the single test you'd write first and show it.
- Tell me one thing the author did well, specifically.

If you're unsure whether something is a bug, say so and explain what
information would settle it. Do not invent APIs that aren't in the code.

It is {{date six months from now}}. Our project, {{project name}}, has failed.

Here's what the project is:
{{two or three sentences: goal, team, timeline, budget}}

1. Write a short, believable story (under 200 words) of how it failed,
   as told by someone on the team.
2. List the eight most likely causes of failure. For each, rate
   likelihood (1-5) and damage (1-5), and sort by likelihood x damage.
3. For the top three, give one early warning sign we'd see in the
   next 30 days.
4. Suggest the single cheapest test we could run this week to reduce
   the top risk.

Be specific to our situation. No generic risks like "poor communication"
unless you can say exactly how it would show up.

Editorial illustration for an article titled "{{headline}}".

Core idea to visualise: {{the one concept, e.g. "a small person steering a huge machine with a tiny tiller"}}
Composition: {{e.g. single central figure, lots of negative space, low horizon}}
Style: flat shapes, subtle paper grain, limited palette of {{3-4 colours}},
       inspired by mid-century print posters
Mood: {{e.g. curious, slightly uneasy, hopeful}}
Format: {{aspect ratio}}, leave the top third clear for a headline
Avoid: text, logos, photorealism, glowing circuits, robots with human faces

Summarise the document below for {{audience}} in about {{length}} words.

"""
{{document}}
"""

Rules:
- Every factual claim in your summary must be followed by a short direct quote
  from the document in [square brackets] that supports it.
- If you can't find a supporting quote, don't make the claim.
- Keep numbers exactly as written in the source, including units.
- After the summary, list:
  a) anything important the document leaves unclear or unsupported
  b) any claims in the document that would be worth checking elsewhere

If the document is too long to read carefully in full, tell me which
sections you prioritised.

Character reference sheet for {{character name}}, {{one-line description}}.

Show the same character in:
- front view, side view and back view, full body, neutral pose
- three facial expressions: {{e.g. calm, surprised, determined}}
- one action pose: {{action}}

Fixed details (must be identical in every view):
- {{hair}}, {{eye colour}}, {{distinctive feature}}
- outfit: {{outfit with colours}}
- proportions: {{e.g. stocky, about six heads tall}}

Style: {{art style}}, clean lines, flat colours, plain light grey background,
small neutral labels under each view

Here's an academic paper (or its abstract, methods and results):

"""
{{paper text}}
"""

Explain it to me as a smart reader from outside the field ({{my background}}).

1. The question they asked, in one sentence.
2. What they actually did, step by step, with no jargon (or define it).
3. What they found, with the key numbers and what those numbers mean.
4. What the paper does NOT show, even though a headline might claim it does.
5. How strong the evidence is: sample size, controls, replication.
6. Three questions I should ask if I'm going to cite this.

If any part depends on figures or tables you can't see, tell me.

Quiz me on {{topic}} at {{level, e.g. GCSE, first-year university, curious adult}}.

Rules:
- Ask one question at a time and wait for my answer.
- Start easy. Each time I'm right, make the next question harder.
- When I'm wrong, explain the mistake in two or three sentences, then ask a
  similar question at the same level to check I've got it.
- Mix question types: explain, calculate, spot the error, apply to a new case.
- Every five questions, tell me my current level and my weakest sub-topic.

Stop when I say "done" and give me a short study plan for my weak spots.

Here are notes or a transcript from a meeting about {{topic}}:

"""
{{notes}}
"""

Attendees: {{names and roles}}

Produce:
1. Decisions made (only things that were actually agreed).
2. Action items as a table: task | owner | deadline | source quote.
   If no owner or deadline was stated, write "UNASSIGNED" or "NO DATE".
   Don't guess.
3. Open questions nobody answered.
4. A three-sentence summary for someone who missed the meeting.

Flag anything that sounds like a decision but might have been just an idea.

I need to write an email I've been putting off.

Recipient: {{who they are and our relationship}}
What I need to say: {{the hard message}}
What I want to happen next: {{desired outcome}}
What I'm worried about: {{the fear, e.g. sounding cold, damaging the relationship}}

Write three versions:
1. Direct: the message in the first two sentences, warm but brief.
2. Gentle: a little more context first, same clarity by the end.
3. Very short: five sentences maximum.

For every version:
- Never apologise more than once.
- No vague phrases like "moving forward" or "touch base".
- End with one clear, specific next step.

Then tell me which version you'd send and why, in one sentence.

Act as a Socratic tutor for {{subject}}. I'm working on this problem:

{{problem}}

Rules for you:
- Never give me the answer or write the solution for me.
- Ask one guiding question at a time.
- If I'm stuck twice in a row, give a small hint, not a step.
- If I make an error, ask a question that helps me notice it myself.
- When I reach the answer, ask me to explain why it works in my own words.

If I ask you to just tell me the answer, remind me of this deal and offer
a bigger hint instead.

Here are notes from {{number}} customer interviews about {{product or problem}}:

"""
{{interview notes, separated by ---}}
"""

1. List the recurring problems, with how many interviews mentioned each.
2. For each problem, include one verbatim quote, marked with which interview it's from.
3. Note any contradictions between interviewees.
4. Separate what people DID (behaviour) from what they SAID they'd do (opinion).
5. Suggest three decisions these interviews support, and one decision they
   clearly do NOT support yet.

Only use what's in the notes. If a pattern appears in just one interview,
label it "single mention".

{{your request}}

Before you start, brief me back.

1. In two sentences, restate what you think I'm asking for and who it's for.
2. List the assumptions you're making that I didn't state. Label each one "safe" or "risky".
3. Ask me up to three questions whose answers would most change the result.
4. Describe what the finished version will look like: length, format, tone,
   and one thing you will deliberately leave out.

Do not begin the task until I answer your questions or reply "go".

Studio product photograph of {{product}}, {{material and colour}}.

Setup: single product, centred, on a seamless {{background colour}} backdrop
Lighting: large soft key light from upper left, gentle fill, soft natural shadow
Camera: {{angle, e.g. three-quarter view slightly above}}, 85mm lens look,
        everything in focus
Details: show {{key feature}} clearly; surface texture visible
Style: clean commercial catalogue, true-to-life colour
Avoid: props, hands, reflections that hide the product, added text or logos

Guide me through a 15-minute weekly review. Ask me one question at a time
and wait for my answer before moving on.

Questions to cover:
1. What did I finish this week? (Ask me to list, then celebrate one item specifically.)
2. What slipped, and was it the plan or the week that was wrong?
3. What's one thing I learned or would do differently?
4. Looking at next week's calendar ({{paste calendar or key commitments}}),
   what's the ONE outcome that would make it a good week?
5. What will I say no to next week to protect that outcome?

At the end, give me a five-line summary I can paste into my notes,
and one gentle observation about any pattern you noticed.

I'm writing a piece about {{topic}} for {{publication or audience}}.
The core point is: {{one-sentence summary}}.

Write ten headlines in three groups:
- Angle A, the news: what happened or what's new (3 headlines)
- Angle B, the reader: what it means for the person reading (4 headlines)
- Angle C, the surprise: the counterintuitive or human detail (3 headlines)

Constraints: under 12 words each, no colons, no questions, no "Here's why".
Avoid these words: game-changer, revolutionise, unlock, everything you need to know.

Then pick the two strongest overall and, for each, tell me:
- what promise it makes to the reader
- whether the piece as summarised actually keeps that promise

Here's a piece of content I made:

"""
{{original post, script or article}}
"""

Repurpose it into five formats, each native to its platform rather than a summary:
1. A short newsletter section (120 words) with a personal opening line.
2. A carousel of 6 slides: slide text only, one idea per slide.
3. A 45-second video script with a hook in the first three seconds.
4. A discussion question for a community forum that invites real replies.
5. A plain-text post under 280 characters.

For each one, change the angle slightly so someone who sees all five
isn't bored. Keep my voice: {{describe it in three words}}.