Curated AI prompts you can run right now.
212 system-prompt bundles. Paste an input, hit Run — get instant output from our AI. Free, no signup.
Type what you want to do — e.g. “write a cold email”, “summarise a contract”. Hit ✨ Ask AI if keyword search misses.
⭐ Featured
Reduces any long article to a 3-bullet TL;DR plus a single-sentence one-liner.
For when you have to brief someone before a meeting. Captures the *implications*, not just the facts.
Cuts an email in half without losing the ask.
Paste any draft email. Get a tighter version that still says everything that mattered.
Takes a vague prompt and rewrites it to be specific, structured, and reliable.
Adds output format, examples, constraints, and edge-case handling.
Reviews a git diff like a senior engineer — flags real bugs, perf issues, security risks, and style.
Doesn't bikeshed about formatting. Focuses on what would actually break in production.
Breaks a top-line metric (e.g. ARR) into a tree of contributing inputs the team can act on.
Stops "improve revenue" being the only thing in the company OKRs.
Corrects + explains errors in your second-language writing — like a tutor would.
Grammar + word choice + idiomatic phrasing. Coaching, not just correction.
15 headlines across 5 proven copywriting formulas for the same article.
Useful before publishing — A/B test the top 3 in email subjects + LinkedIn.
A 45-min lesson plan for a topic + student age — objectives, materials, activities, assessment.
Ready to print and hand to a substitute teacher.
Builds a weighted decision matrix — options × criteria × weights → scored recommendation.
For decisions where gut isn't enough.
Generates unit tests — happy path, edge cases, error paths — in the test framework you use.
Covers branches the AI can deduce from the code, not just trivial cases.
Drafts a starter privacy policy from a product description — GDPR + CCPA-aware.
Not a substitute for a lawyer. A solid starting point that covers the standard sections.
Builds a JSON template + 10 sample I/O pairs for a fine-tuning dataset.
For instruction-tuning tasks — start with clean, varied examples.
Internal memo: should we settle this dispute or fight? With BATNA + expected value math.
For founder / counsel decision — quantifies expected outcomes.
Translates text into the target language AND adapts cultural references / units / examples.
Not just word-for-word — adjusts examples, currencies, and idioms.
Reviews a git diff like a senior engineer — flags real bugs, perf issues, security risks, and style.
Doesn't bikeshed about formatting. Focuses on what would actually break in production.
Writes a polite-but-firm "no" reply — preserves relationship without saying yes.
For requests on your time you should decline.
New-hire onboarding plan for the first 90 days — learn, contribute, own.
Specific deliverables per phase. Stops "shadow people and figure it out".
Drafts a clean offer letter — role, comp, equity, start date, conditions, references to policies.
Indian / US conventions. NOT a substitute for employment counsel.
Turn a story or insight into a 1,200-character LinkedIn post that reads like a person wrote it.
Punchy opening line. Short paragraphs. One concrete lesson. No corporate buzzwords.
30 daily journaling prompts — varied themes, depth without being heavy.
For self-reflection, mental clarity, gratitude. Each prompt 1 sentence + a follow-up.
Writes the management commentary for a P&L — what drove the variance vs plan.
For CFO / board reporting. No "revenue was driven by good performance".
12 subject-line variants across 4 styles (curiosity, value, urgency, personal).
For email A/B tests. Real, not "🔥 You won't believe this".
Crafts a click-worthy SEO title (≤60 chars) + meta description (≤155 chars) targeting a keyword.
Title prioritises Google ranking factors. Meta prioritises CTR.
A 500-700 word bedtime story tailored to a child's age + favorite themes.
Gentle, age-appropriate vocabulary, satisfying ending, encourages sleep.
Reduces any long article to a 3-bullet TL;DR plus a single-sentence one-liner.
For when you have to brief someone before a meeting. Captures the *implications*, not just the facts.
Writes annotation guidelines that two annotators can apply consistently.
For labeled-data projects — sentiment, classification, NER.
Writes a 500-word short story in the genre + style requested.
Tight structure, beginning-middle-end. No exposition dump.
Reviews a screen description and lists usability issues ranked by severity.
Catches the obvious things a Friday-shipped feature missed. Heuristic-based, not subjective taste.
Writes a clean conventional-commits message (feat/fix/chore/etc.) from a diff or description.
Stops "wip fix stuff" commit history.
Rewrites text in a different tone (formal ↔ casual / friendly ↔ professional / etc.).
Same meaning, different register. Useful for code reviews, emails, support replies.
Categorises a list of expenses into Needs / Wants / Investments + sub-categories.
For monthly review, tax prep, or expense report cleanup.
Cuts an email in half without losing the ask.
Paste any draft email. Get a tighter version that still says everything that mattered.
Writes a firm but polite follow-up email for an overdue invoice.
Three escalation levels: gentle nudge, second reminder, formal final notice.
Defines explicit "kill this feature" criteria so you stop maintaining duds.
Most teams ship 100x more than they kill. This forces honest cleanup.
Decodes a regex into plain English, character-class by character-class.
No more squinting at /^\d{3}-\d{4}$/.
Decodes a cron expression into "Every X at Y" plain English with the next 5 run times.
Saves you from cronjob misfires at 03:00 on the wrong day.
Fills in all 9 boxes of the Osterwalder Business Model Canvas from a business description.
For early-stage thinking, repositioning workshops, or onboarding investors.
A day-by-day vacation itinerary balanced for activity + rest + local food.
Avoids over-scheduling. Leaves room for spontaneity.
Takes a vague prompt and rewrites it to be specific, structured, and reliable.
Adds output format, examples, constraints, and edge-case handling.
Builds a structured eval rubric to grade LLM outputs for a specific task.
Stops vibes-based "did the AI do good?" review.
Drafts an on-call runbook for a specific failure mode — detect, diagnose, mitigate, recover.
Saves the 3am scramble of "what do I do?".
Generates a multi-stage, small-image Dockerfile for a given stack.
Multi-stage build, non-root user, .dockerignore, healthcheck, cache layers.
10 CTA copy variants for a single button, sorted by intent (high vs low).
For A/B tests on landing pages, modals, emails.
A short cold email that reads like a 1:1, not a campaign. Personalises around the prospect's actual situation.
Subject line + 3-paragraph body. P.S. line. Single ask. Built for reply rates above 5%.
Reads a contract and surfaces the obligations, dates, risks, and exit clauses in plain English.
NOT legal advice. Saves time before you take it to your lawyer.
A system prompt template that pushes the model to show its reasoning before answering.
Higher accuracy on multi-step problems, debuggable when wrong.
A personalised plan to get to inbox zero — triage rules, batching, templates.
Stops constant inbox-checking. Uses Allen + Sunday Edition principles.
Turns a 2-3 line feature idea into a complete Product Requirements Document.
Stops the "let's just start coding" trap. Forces explicit thinking about success metrics.
3-5 min wedding speech — story-driven, warm, lightly funny, no offense risk.
Structured for delivery — opens with a hook, ends with a toast.
Pulls action items out of an email thread or meeting transcript — who, what, by when.
Goes after "todos" that hide behind "let me know" / "circle back".
Suggests 10 thoughtful gift ideas across budgets for a specific person.
Goes beyond generic "buy them a gift card". Considers interests + relationship.
Explains a grammar rule with examples + counter-examples + 3 practice sentences.
For language learners — replaces dense textbook explanations.
Drafts a model card with intended use, limitations, training data summary, evaluation results.
For shipping a model responsibly + onboarding new users.
Designs a system prompt that turns user questions into retrieval queries — multi-query, HyDE, etc.
For RAG systems where retrieval quality dominates answer quality.
Pulls a clean list of action items + owner + deadline from any wall of text.
Paste meeting notes, a long email thread, or a Slack dump. Get a structured to-do list.
Generates an Ash Maurya Lean Canvas — designed for early-stage startups, not big corps.
Same canvas, leaner: Problem, Solution, Unique Value Prop, Unfair Advantage, Channels, Customer Segments, Key Metrics, Cost Structure, Revenue Streams.
Generates 20 brandable name ideas for a product, grouped by style (descriptive / abstract / portmanteau).
Each name includes the .com / .ai availability heuristic (not authoritative).
Brief for a logo designer — usage, constraints, do's/don'ts, deliverables.
Reduces the "let me do 30 variations" surprise.
7-day meal plan with breakfast/lunch/dinner + macros for a given calorie target and diet style.
Currency/region-aware ingredient choices. Includes a grocery list.
Generates a 30-day content calendar across blog + email + social with themes.
Balances educational, promotional, and conversational posts.
AP-style press release for a product launch, fundraise, or partnership.
Headline, dateline, lede, body, boilerplate, contact — the format reporters expect.
Brief for a cohesive icon set — style, weight, grid, list of icons, deliverables.
Stops icon Frankensets (mix of Lucide + Heroicons + emoji).
Converts a podcast transcript into show notes — timestamps, summary, key quotes.
For listeners who skim, for search visibility, and for SEO.
Drafts a 750-word op-ed with a sharp argument + 3 supporting points + counter-argument addressed.
For publications like NYT, WSJ, Hindu BusinessLine — strong opinion, civil tone.
Page 1 of 4 · 212 prompts