New prompts and trending repos: week 46
This week: a system prompt for research agents, a safer browser automation pattern, and three repositories gaining stars fast.
Search prompts, categories, platforms and repositories
AIPrompt123 is a curated library of 538 tested prompts across 25 platforms. Every prompt includes fill-in template variables, negative constraints, and worked examples so you can skip the trial-and-error and get direct results.
Goal: Technical architecture & SEO audit for multi-tenant SaaS
Structured XML, few-shot edge cases, strict negative constraints
Auto-formatted for Claude 3.7 Sonnet, ChatGPT 4o or Cursor
Production code, schema markup & clean documentation
Every template is engineered to solve a concrete task across industry-standard AI engines. Select a discipline to inspect the anatomy of a production-grade prompt.
Structure content so search bots and generative LLMs cite your brand as the canonical authority in direct answers.
Optimized for
+320%
AI Citations
84% Win
Position Zero
Writinggoodpromptsisslow,inconsistentandhardtorepeat—AIPrompt123givesyoutestedonesinseconds.
Stop treating AI like a magic slot machine. We treat prompt design as rigorous interface engineering: tested variables, failure modes, and platform-specific syntax.
Engineers and creators waste 20+ minutes coaxing LLMs past lazy summaries and generic corporate fluff. Ad-hoc prompts rarely produce production-ready output.
A phrasing that works well in ChatGPT often fails completely in Claude 3.7, Cursor, or Midjourney. Each model demands distinct role boundaries and token discipline.
High-value prompts get buried in individual browser histories. Teams end up solving the same problems in isolation without standardized, versioned templates.
No bloat, no subscription paywalls, no login friction. Curated and maintained daily by practitioners who build with generative AI and search engines every day.
Every prompt marks dynamic inputs in brackets [like this] so you know exactly what parameters to fill before sending.
Calibrated for Claude XML tags, Cursor file references, ChatGPT Canvas mode, and Midjourney aspect parameters.
Each prompt includes failure-mode blockers that stop models from inventing citations or spewing generic filler.
No account creation, no paid barriers. The entire 538+ library is open and accessible from any device.
Stop guessing model syntax. From Claude 3.7 reasoning tags to Cursor Composer instructions and Sora shot trajectories, our library provides battle-tested recipes with zero login requirements.
0+
Prompts
0
Platforms
0
Categories
From basic chat completions to autonomous multimodal agents
Dedicated Website Growth Hub
Audit technical SEO, win Google AI Overviews & Featured Snippets (AEO), and get cited inside ChatGPT, Perplexity & Claude (GEO) with inbuilt + live-updated prompts.
Swipe through newly indexed prompts and starred GitHub AI repositories.
20 Specialized Categories
Every category contains 26 original prompts with worked examples, failure modes and platform-specific variants.
Draft, rewrite, edit and structure prose for articles, emails, reports and fiction. Prompts built around voice, audience and revision loops.
26 prompts
Keyword research, content briefs, meta tags, ad copy and campaign planning prompts that respect search intent and editorial standards.
44 prompts
Strategy memos, customer research synthesis, pricing analysis, meeting briefs and operating documents for founders and operators.
26 prompts
Debugging, code review, test generation, refactors, architecture and documentation prompts for developers using Cursor, Copilot, Claude and ChatGPT.
26 prompts
Turn raw tables, surveys and logs into clean summaries, charts, SQL and decision-ready findings with explicit assumptions.
26 prompts
Study plans, explanations, flashcards, essay feedback and exam practice that support learning instead of replacing it.
26 prompts
Resume tailoring, cover letters, LinkedIn profiles, interview practice and salary negotiation with honest, evidence-based framing.
26 prompts
Hooks, scripts, titles, thumbnails, shorts, carousels and content calendars for YouTube, TikTok, Instagram, LinkedIn and X.
26 prompts
Product descriptions, collection pages, ad creative, review responses and retention emails that convert without misleading claims.
26 prompts
Financial analysis, budgeting, unit economics and market research prompts with explicit educational-only guardrails.
26 prompts
Plain-language contract summaries, policy drafts, compliance checklists and research prompts that require qualified legal review.
26 prompts
Fresh Every Morning
One hand-checked prompt, rotated daily. The full page includes the variable form, platform tabs and a worked example.
This copy-ready healthcare prompt helps you create a health information source checklist with fewer blank-page decisions. Paste it into ChatGPT and Claude, fill the details, and you get a structured first draft you can edit. It is written for repeatable results and tells you what to check before publishing.
Create a checklist for evaluating online health information. Cover author credentials, evidence citations, date, commercial bias, sensational claims and peer review. Give one example of a trustworthy source type and one of a red flag. Keep it practical.
Fill-in variables are highlighted on the full page.
Craft Guides
Practical guides on prompt engineering, AI search optimisation, RAG and the tools worth your attention.
This week: a system prompt for research agents, a safer browser automation pattern, and three repositories gaining stars fast.
The frameworks, inference engines, agent tools and learning resources that are actually worth your time, with licence notes and what each is best for.
Marketing, customer service, pricing and admin prompts for teams without a dedicated AI department, plus the guardrails to keep you out of trouble.
How prompts are written, tested and maintained, and what you get for free.