Prompt Shortcuts

100 AI behavior commands organized into 10 categories. Copy and use at the start of any prompt.

KitsPortfolio

Total

100

Categories

10

Showing

100

00
start
01

Core Writing & Style

10
/blunt

Remove hedging, caveats, and filler. Output direct, declarative statements only.

/compress

Cut output to the shortest form that preserves all meaning. No padding.

/expand

Elaborate each point with evidence, examples, and full context.

/reformat

Restructure existing content with headers, bullets, and whitespace for clarity.

/formal

Switch register to professional, precise, and citation-ready language.

/casual

Shift tone to conversational, first-person, and approachable prose.

/eli5

Explain using simple analogies and vocabulary a fifth-grader would understand.

/expert

Respond assuming the reader has deep domain expertise. Omit basics.

/bullets

Convert prose into a scannable bulleted list without loss of information.

/oneliner

Distill the full response into a single, standalone sentence.

02

Artifacts & Creation

10
/draft

Produce a complete first-pass artifact ready for editing, not a skeleton.

/template

Output a reusable, fill-in-the-blank document structure for the given type.

/sop

Generate a numbered standard operating procedure with checkboxes and roles.

/email

Write a professional email with subject line, greeting, body, and CTA.

/readme

Generate a full project README with setup, usage, and contribution sections.

/proposal

Create a structured proposal document with problem, solution, timeline, and cost.

/brief

Produce a creative or project brief with goals, audience, constraints, and deliverables.

/changelog

Format a versioned changelog entry with added, changed, fixed, and removed sections.

/report

Structure findings into an executive-ready report with sections and a summary.

/checklist

Generate a task-completion checklist ordered by priority and dependency.

03

Thinking & Reasoning

10
/steelman

Construct the strongest possible version of the opposing argument before responding.

/firstprinciples

Break the problem down to foundational assumptions and rebuild from ground zero.

/contrarian

Challenge the premise. Argue the case against the assumed correct answer.

/inversion

Approach via inversion: define what failure looks like, then reverse to success.

/bayesian

Reason explicitly with probabilities, updating priors as evidence is introduced.

/falsify

Identify conditions or evidence that would prove the stated claim false.

/tradeoffs

Map every decision as explicit trade-offs — what is gained and what is sacrificed.

/secondorder

Extend analysis to second and third-order consequences of the stated action.

/devil

Identify every failure mode, edge case, and weak assumption in the plan.

/socratic

Respond only with questions that expose hidden assumptions and gaps in reasoning.

04

Learning & Mastery

10
/map

Produce a structured knowledge map of the topic showing relationships between concepts.

/roadmap

Generate a sequenced learning path from beginner to advanced for the given subject.

/quiz

Create 5-10 graded questions to test comprehension of the specified topic.

/flashcards

Generate Q&A flashcard pairs optimized for spaced-repetition memorization.

/analogy

Explain the concept using a precise analogy drawn from everyday experience.

/contrast

Compare two or more concepts by surfacing key differences in a table.

/feynman

Teach the concept from scratch as if the listener has zero prior knowledge.

/mental-model

Apply a named mental model (specify which) to analyze the given situation.

/prereqs

List all prerequisite knowledge needed to understand the stated topic.

/summary

Condense the material into a structured summary retaining all key ideas.

05

Analysis & Strategy

10
/swot

Run a full SWOT analysis — Strengths, Weaknesses, Opportunities, Threats.

/pestle

Analyze via Political, Economic, Social, Tech, Legal, and Environmental factors.

/compete

Perform a competitive analysis: identify rivals, positioning, and differentiation gaps.

/bottleneck

Identify the single constraint most limiting progress or performance in the system.

/root-cause

Apply 5 Whys iteratively to trace a problem back to its root cause.

/risk

Enumerate risks ranked by likelihood x impact with a mitigation for each.

/scenario

Model three scenarios — best case, worst case, and most likely — with drivers.

/metrics

Define the 3-5 most valid metrics to measure success for the stated goal.

/gaps

Identify what is missing: data, skills, resources, or assumptions not yet tested.

/decision

Frame the decision as a weighted scoring matrix with explicit criteria.

06

Creative & Content

10
/hook

Write 3 opening hooks for the content: curiosity, data, and narrative variants.

/thread

Transform a long-form idea into a numbered social media thread with a CTA.

/headline

Generate 5 headline variations: direct, bold, question, how-to, and listicle formats.

/storify

Rewrite the content as a narrative arc with a protagonist, conflict, and resolution.

/angles

List 5 unique editorial angles to cover the same topic from different perspectives.

/cta

Write 3 call-to-action variants: urgency, value-led, and curiosity-driven.

/repurpose

Map one piece of content to 5 formats: post, email, video script, tweet, short.

/persona

Rewrite the content in the voice of a specified character, brand, or archetype.

/series

Break the topic into a 5-part content series with titles and one-line synopses.

/counternarrative

Write the article that challenges the most popular take on this topic.

07

Coding & Technical

10
/refactor

Rewrite the code for clarity, efficiency, and idiomatic style without changing behavior.

/review

Perform a full code review: logic, edge cases, security, and performance.

/tests

Generate a comprehensive test suite covering unit, edge, and failure cases.

/doc

Write inline documentation, JSDoc comments, and a usage example for the code.

/debug

Trace the bug step by step, state the root cause, and provide the fix with explanation.

/optimize

Profile the code for time and space complexity and output a faster version.

/explain-code

Walk through the code line by line explaining what each block does and why.

/api

Design a RESTful or GraphQL API schema for the described resource with all endpoints.

/migrate

Write a migration plan from the current stack to the target with rollback steps.

/security

Audit for vulnerabilities: injection, auth flaws, exposure, and unsafe defaults.

08

Research & Deep Dives

10
/lit

Summarize the current state of research on the topic with key findings and gaps.

/timeline

Build a chronological timeline of events, decisions, or developments on the topic.

/interview

Generate expert interview questions that uncover non-obvious insight on the subject.

/sources

List the most authoritative primary sources to consult for the stated topic.

/myth

Identify and debunk 3-5 common misconceptions about the topic with evidence.

/data

Surface key statistics, benchmarks, or empirical data points relevant to the claim.

/compare-research

Compare two competing schools of thought or methodologies with their evidence bases.

/expert-view

Synthesize what leading practitioners in the field currently think about the topic.

/landscape

Map the full ecosystem: players, tools, communities, and emerging trends.

/critique

Critically evaluate the methodology, assumptions, and validity of the stated argument.

09

Power Commands

10
ULTRA

Activate maximum output depth. No length limits, no omissions, no summaries.

CHAIN

Chain multiple tasks sequentially and output all results without stopping for confirmation.

ITERATE

Produce 5 distinct versions of the output, each with a different strategic approach.

CRITIQUE+FIX

Self-critique the output immediately after generating it, then output an improved version.

PARALLELIZE

Break the task into independent subtasks and solve all of them simultaneously.

SIMULATE

Run a decision or plan through a simulation with assumptions stated and outcomes modeled.

VERIFY

Cross-check every claim, number, and assertion in the output for accuracy.

META

Analyze the structure of this prompt and suggest how to make it more effective.

AUTOPILOT

Complete the full project end-to-end making all reasonable decisions without pausing.

BENCHMARK

Compare this output against the best-known reference in the field and score the gap.

10

Hidden Modes

10
/oracle

Respond as a domain authority with high conviction. State what is true, not what might be.

/devil-gpt

Adopt a relentlessly critical stance. Find flaws in everything. Argue against approval.

/mirror

Reflect the user's core idea back with clarity, removing all noise and confusion.

/time-capsule

Write the output as if looking back from 10 years in the future on today's decision.

/unsaid

Identify what is not being said in the conversation that is critical to the outcome.

/pattern

Identify the underlying pattern or recurring structure in what the user is describing.

/reframe

Restate the problem from a completely different context that reveals a new solution path.

/minimum

Strip the solution to the irreducible minimum that still solves the core problem.

/wildcard

Generate the most unconventional, unexpected, and creative answer possible.

/ghost

Write as a silent collaborator: no meta-commentary, no preamble. Just the work.

10
100 total