Prompt engineering
Writing, improving and testing prompts for any AI assistant.
Download all 64
Start here 2
The general, start-from-zero entry for a need in this category: it works from a blank start, asks for what it needs and hands off to the specific entries. Example: Plan a trip.
- Improve a prompt
Diagnoses why a prompt gives weak or inconsistent results and rewrites it with clear context, task, constraints and output format while keeping its intent. Use on any prompt for any AI assistant.
- Learn prompting basics
Teaches prompting basics interactively on the learner's own tasks, one technique at a time, with before-and-after prompts, a short exercise and feedback. For beginners to AI assistants.
Specific 60
One situation, audience, tool or format inside the category. Example: Plan a road trip with children.
- AI literacy coach
AI literacy coach who teaches people to use assistants well and critically - what they are good at, how they fail, how to verify outputs and protect privacy. Use for learning to work with AI.
- Compress a prompt
Shortens a long prompt while preserving its behaviour, maps every original instruction to where it now lives, reports the real size reduction and lists test inputs to check nothing changed.
- Create few-shot examples
Builds a small set of diverse, representative few-shot examples for a task, including tricky and negative cases, balanced so the model learns the rule rather than copying surface patterns.
- Diagnose prompt failures
Diagnoses why a prompt produces bad answers from failing examples, traces each failure to a root cause, proposes targeted fixes and a quick regression test set.
- Prompt engineer
Prompt engineer who writes clear, testable instructions, iterates against real examples and evals, and avoids model-specific tricks. Use for designing, debugging and maintaining prompts.
- Prompt iteration track
Improves a prompt in gated steps - define success, build test cases, run and grade, diagnose failures, revise, then compare versions on the same cases before adopting the change.
- Design a prompt chain
Splits a complex task into a chain of focused prompts with defined inputs and outputs, checks between steps, failure handling and a test plan. Use when automating multi-step work with AI.
- Write a system prompt
Writes a system prompt for a custom assistant from its purpose, audience, boundaries and tone, with handling for missing information and off-topic requests, plus a set of test questions.
- Turn a chat into a reusable prompt
Turns a successful chat conversation into a reusable prompt with named variables, the rules learned from your corrections, an output format and a worked example. Use for tasks you repeat with AI.
- Red-team a prompt
Tests a prompt or assistant setup against adversarial inputs - injection, edge cases, off-topic and harmful requests, data leaks - predicts failures and proposes fixes. For assistant builders.
- Write a reusable task prompt
Writes a reusable prompt from a plain description of a task, with context, typed variables and defaults, constraints, an output format, an example and a rule to ask for missing inputs.
- Practise spotting AI errors
Trains AI literacy with rounds of assistant-style answers containing planted errors, such as invented citations, wrong maths or outdated facts, and coaches the user to catch and verify them.
- Write an interview-first prompt
Writes a prompt that makes an assistant ask targeted questions one at a time before producing a document, plan or answer, with a question budget, a stop rule and a summary check.
- Write a deep-research brief
Writes a brief for an AI deep-research run - precise question, scope, source rules, output format and how to judge the result - so a research agent investigates the right thing.
- Build an AI output review checklist
Builds a checklist a team uses to review AI outputs before use, tiered by risk, covering facts, sources, numbers, tone, privacy, rights and sign-off. For teams adopting AI in daily work.
- Write a length-limited copy prompt
Writes a prompt that reliably produces copy within hard character or word limits - ads, SMS, meta descriptions, bios, form fields - using target ranges, counts, code checks and trimmed variants.
- Write a reliable maths prompt
Writes a prompt that gets correct numbers from an AI for quotes, budgets and conversions - code or calculator use, shown workings, units, rechecked totals and flagged estimates.
- Write a report template prompt
Writes a reusable prompt that turns raw notes into a fixed recurring report - weekly status, incident, site visit or shift handover - under exact headings, with missing inputs flagged, never filled.
- Write a summarization prompt
Writes a reusable summarization prompt for a recurring document type and reader, with keep and drop rules, length, structure, faithfulness rules for numbers and uncertainty, and a test plan.
- Write a translation prompt
Writes a reusable AI translation prompt with audience, register, glossary, do-not-translate list, formatting and placeholder preservation, and a self-check for omissions and added meaning.
- Anonymise text before pasting it into an AI tool
Replaces names, clients, codenames and confidential figures in text with consistent stand-ins before you paste it into an assistant, with a restore key and a check for clues left behind.
- Choose temperature and other generation settings
Picks temperature, top-p, maximum output length, stop sequences and seed for an API task, explains each choice and gives a quick repeat-run test for consistency versus variety.
- Write a roleplay character prompt
Writes a roleplay character prompt with personality, voice samples, knowledge limits, boundaries and consistency rules. For interactive fiction, language practice, training and games.
- Write a data analysis prompt
Writes a prompt for an assistant with code execution to analyse an uploaded CSV or spreadsheet correctly - inspect first, state assumptions, show code, reconcile totals, flag dirty data.
- Write an image analysis prompt
Writes a reusable prompt for reading screenshots, charts, receipts, forms or photos with an AI that can see images - region by region, illegible parts marked, seen kept apart from inferred.
- Write prompts for spreadsheet AI functions
Writes prompts for AI functions inside spreadsheet cells that classify, extract or rewrite each row consistently, with fixed outputs, blank handling, cell assembly and a spot-check plan.
- Build a test set for a prompt
Builds a hand-run test set for a prompt with happy, edge and negative inputs, expected behaviour and checkable pass criteria per case, and a scoring sheet to compare prompt versions side by side.
- Run prompt regression tests
Runs a prompt test set against the current and candidate prompt versions with the project's command, grades outputs with the stated checks and reports regressions, wins and flaky cases side by side.
- Write a classification prompt
Writes a classification prompt with sharp label definitions, include and exclude rules, boundary cases, balanced examples and an abstain option, plus a plan to measure accuracy on labelled data.
- Write a ranking prompt
Writes a prompt that scores and ranks a list of items such as leads, applications, ideas or tickets against weighted criteria, with evidence per score, tie rules and checks for order bias.
- Choose the right prompting technique for a task
Matches a real task to the prompting techniques that fit it - examples, decomposition, attached sources, step-by-step reasoning, structured output or a tool - explains why, and writes a first draft.
- Fix a prompt that keeps breaking a rule
Finds why a prompt keeps breaking a length, format or banned-word rule and fixes it with better placement, positive phrasing, matching examples and an output check where needed.
- Write a task brief for a browsing agent
Writes a task brief for a browsing or computer-use agent with a clear goal, allowed sites, choice criteria, stop-and-confirm points before purchases or submissions, and a result format.
- Write prompts for an office suite assistant
Writes prompts for the AI assistant inside a word processor, spreadsheet, email or chat app that point to the right files, ranges and threads and ask for output you can check.
- Audit a prompt for bias
Audits a prompt for biased framing, stereotyped examples, proxy attributes, exclusionary assumptions and unequal treatment across groups, then suggests neutral rewrites and paired tests.
- Build a team prompt library
Designs a shared prompt library for a team - structure, naming, an entry template with variables, ownership, review and versioning - and drafts the first entries for the team's top use cases.
- Convert an SOP into a prompt
Converts a standard operating procedure into assistant or agent instructions with ordered steps, decision rules, checks, escalation triggers, a gap list and test scenarios traced to the SOP.
- Port a prompt to another model
Ports a working prompt to another model family or vendor, adjusting structure, examples, format instructions and call settings, with a parity test plan. For builders switching models.
- Reduce made-up answers from a prompt
Rewrites a prompt so answers stay grounded, using source-only answering, quote-then-answer, an explicit not-found reply, known-versus-inferred labels and checkable citations.
- Write a batch processing prompt
Writes a prompt for processing many items consistently through an API or script, with a per-item output schema, stable labels, ID echo, error records and a QA sampling plan.
- Write an expert panel prompt
Writes a prompt that has an AI critique a plan from several distinct expert perspectives, with forced disagreement, evidence for each view, and a synthesis that keeps the unresolved conflicts.
- Write a long document prompt
Writes a prompt for analysing long documents with document placement, metadata tags, quote-first answering, citations, a not-found rule and a chunking plan for documents that do not fit.
- Write a long-form generation prompt
Writes a prompt set for getting long documents such as reports, guides or manuals from an AI past output limits - outline first, section by section, continuity notes and a final consistency pass.
- Write a research table prompt
Writes a prompt that fills a comparison matrix of vendors, tools or options criterion by criterion, with a source and date per cell and "unknown" instead of guessed values.
- Write a structured output prompt
Writes a prompt that returns schema-valid JSON reliably, with a JSON Schema, field rules, examples, edge-case handling and a validate-and-retry plan. For extraction and app integrations.
- Write a synthetic customer persona prompt
Writes a prompt that simulates a target customer to give early feedback on ideas, copy or flows, grounded in supplied research, with clear caveats on what simulated feedback cannot tell you.
- Build a prompt from input and output examples
Reverse-engineers a reusable prompt from input and output example pairs, naming the implicit rules, format and edge cases, then dry-runs the prompt against the examples, including held-out ones.
- Choose a model tier for a task
Picks a first-choice model tier and reasoning setting for a task from its difficulty, volume, latency, cost and risk, with signs to step up or down and a check sized to the stakes. No model names.
- Compare AI assistants on your own task
Sets up a blind side-by-side test of two or three assistants on your real prompts, then grades the pasted outputs against your criteria and recommends which to use and when.
- Run a prompting workshop for your team
Plans a 60 to 90 minute hands-on team workshop on using AI assistants, built on the team's own tasks, with timed exercises, before-and-after prompts, verification habits and data rules.
Showing 50 of 60. See them all in search
Advanced 2
Expert or deep work that assumes the user can judge expert output. Example: Plan croissant lamination.
- Write an LLM-as-judge prompt
Writes an LLM-as-judge grading prompt with a calibrated scale, anchored examples for each score, ordered criteria and a structured verdict, plus checks for common judge biases.
- Harden a prompt against injection
Hardens a prompt or assistant against prompt injection from untrusted content with input separation, an instruction hierarchy, least-privilege actions, output limits and an attack test set.
Not: coding-agent operations (meta).