Make a dataset into an explorable museum
Turns real figures into interactive exhibits that preserve provenance and uncertainty.
[DATASET][AUDIENCE][QUESTION]
A growing, practical AI prompt library
Better instructions.
More useful AI.
Find complete prompts for real tasks, understand what each one does, replace the variables and use it immediately.
693prompts
48categories
0login required
New: prompts for messy reality
Useful AI should survive a fact-check.
Turn claims, files, screenshots and risky offers into answers you can inspect. These prompts show what AI can and cannot verify.
Reality & evidenceReceipts, changing decisions and source checks.Check the claims →
Document detectiveTimelines, contradictions and missing records.Follow the evidence →
Safety & trustJob offers, suspicious links and privacy checks.Pause and verify →
From messy inputScreenshots, voice notes, photos and vague briefs.Make it usable →
Copy, paste, smile
A useful answer—or a good laugh—in one prompt.
These short prompts work in a chat you already use. For personal questions, the AI should say what it actually knows, not pretend it can see your whole life.
What AI knows about youAudit memories, assumptions, context and blind spots.Explore prompts →
Fun and gamesCharacters, tiny legends, friendly roasts and absurd side quests.Have some fun →
Everyday helpersQuick decisions, better messages and small practical plans.Find a helper →
Playful photo editsCaricatures, pop-up portraits and delightfully odd versions of your photo.Explore image prompts →
Find your starting point
What do you want AI to make?
Choose a format, then explore its topics. Every prompt has its own page.
Browse all 48 topics
Browse the library
Start with a task, not a buzzword.
Turn a memory into a playable web experience
Specifies a compact accessible site where exploring objects reveals a story.
[MEMORY OR FICTIONAL PREMISE][PLACE]
The impossible museum in one continuous shot
Plans a one-shot video through rooms whose objects gradually change the story.
[DURATION][THEME][OBJECT]
One walk, two possible worlds
Creates a split-screen video concept where one person makes a different small choice in each half.
[SETTING][CHOICE A][CHOICE B]
Turn a messy voice note into a tiny documentary
Structures a personal recording into a short documentary while preserving what was actually said.
[SUBJECT][AUDIENCE]
An audio drama told by an everyday object
Creates a short, original audio drama with sound doing most of the storytelling.
[OBJECT][LOCATION][THEME]
A four-door story that changes its meaning
Builds a short branching story where the choices alter the interpretation, not just the ending.
[ORDINARY PROBLEM][SETTING][ROLE]
Turn a vague business idea into a testable offer
Builds a small, honest market test with clear customer, value, cost and stop criteria.
[IDEA][CUSTOMER][LIMITS][WHAT WE KNOW]
Audit a bold claim before sharing it
Traces a claim to its original evidence and separates what is true, missing and overstated.
[CLAIM][WHERE AND WHEN IT APPEARED]
Create a practice exam with feedback, not answer leaks
Builds a balanced question set, marking rubric and targeted remediation path.
[SUBJECT][LEVEL][LEARNING OUTCOMES][MINUTES]
Turn a topic into an adaptive Socratic tutor
Teaches through questions, examples and misconception checks instead of dumping an answer.
[TOPIC][CURRENT LEVEL][GOAL]
Define dashboard metrics people can trust
Turns vague KPIs into formulas, source-of-truth rules, refresh cadence and audit examples.
[BUSINESS PROCESS][AUDIENCE][DATA SOURCES][DECISIONS]
Find the story and errors in a messy dataset
Audits a CSV for quality issues before building defensible findings and charts.
[DECISION OR QUESTION]
Organise messy files with a dry-run manifest
Plans reversible file organisation without deleting or silently overwriting anything.
[EXACT ROOT]
Monitor a topic and report only meaningful changes
Defines a repeatable research monitor with source quality, deduplication and quiet unchanged runs.
[TOPIC][ALLOWED PRIMARY SOURCES][SCHEDULE][ACTIONABLE THRESHOLD]
Test a site as several real user types
Walks key journeys with screenshots, failures and reproducible evidence rather than a vague quality score.
[SITE][PERSONAS][STAGING OR PRODUCTION][TEST ACCOUNTS AND DATA]
Create a bounded inbox-triage agent
Sorts incoming messages, drafts replies and escalates important cases without sending on its own.
[TEAM][ALLOWED INBOX AND TIME RANGE]
Build and test an accessible UI component
Turns a UI request into keyboard, screen-reader, responsive and visual acceptance tests.
[COMPONENT][FRAMEWORK][USER TASK][REFERENCE]
Build an API integration with failure handling
Plans an integration around real schemas, authentication boundaries, retries and tests.
[SOURCE SYSTEM][DESTINATION SYSTEM][BUSINESS OUTCOME][API MATERIAL]
Refactor code without changing its behaviour
Defines a safe refactor with invariants, characterisation tests and rollback checks.
[MODULE OR FILE][MAINTAINABILITY GOAL][LANGUAGE AND FRAMEWORK][FILES IN SCOPE]
Diagnose a bug from a reproducible case
Forces a minimal reproduction, hypothesis ranking and verification before proposing a patch.
[SYSTEM AND LANGUAGE][SYMPTOM][EXPECTED][MATERIAL]
Localise a spoken message for multiple languages
Creates performance-ready versions with meaning, timing and cultural checks rather than literal translation.
[SOURCE LANGUAGE][TARGET LANGUAGES][REGIONS AND AUDIENCES][SECONDS]
Build an original song production brief
Maps a song idea into sections, instrumentation, vocal direction and originality guardrails.
[DURATION][THEME][AUDIENCE][ENERGY ARC]
Design a believable soundscape from a scene
Produces layered ambience with foreground cues, spatial placement and a clean mix plan.
[DURATION][SCENE][LOCATION][TIME]
Need an AI worker, not one answer?
Use the agent blueprint.
Define tools, authority, proof, persistence and safe stop conditions.