AI prompt directory

Build a structured 60-minute lesson with objectives, activities, timing, and an outline.

Develop an A/B testing framework with protocols, hypotheses, and success metrics.

Copyedit prose through staged revisions for clarity, grammar, tone, and style.

Generate advertising angles and copy variants for a product.

Choose the next agent tool call from a user goal and available schemas, returning arguments in JSON.

Extract IATA airport codes from itinerary text in the order they appear as JSON.

Write accessible alt text for a batch of product image descriptions.

This prompt writes an Amazon ad for a newly launched product. The copy has a headline, an opening line, key benefits, a reason to act now, social proof, and a call to action, plus one idea for the image.

Require verification-first behavior to reduce unsupported or hallucinated answers.

Design an anti-inflammatory meal plan based on cooking skill and preparation time.

Convert JSON API documentation into example curl commands.

Summarize an article with author, topics, coherence, persuasion, and a grounded counterpoint in JSON.

Implement asynchronous job polling with exponential backoff, timeouts, and explicit state handling.

This prompt reviews a YouTube channel from its link, details, screenshots, analytics, or videos. It checks search visibility, the videos themselves, thumbnails, titles and tags, subscriber growth, rival channels, and Shorts. You get 7-, 30-, and 90-day plans, ideas for new videos and their titles, and scores from 1 to 100.

This prompt tells a coding agent to map an entire repository, find bugs and security issues, and rank them by severity. It then fixes them with tests and writes up each fix. It works with any language or stack.

This prompt finds recurring charges in statement lines you paste and estimates what each costs per year. Each one gets a call to keep it, downgrade it, pause it, or cancel it, and the JSON ends with a brief to-do list.

This prompt sets up an agent that reads customer spreadsheets and types the data into a target app using Playwright scripts. It runs validation tests on the result and logs any errors. The agent needs Playwright access.

Suggest clearer variable and function names for a code snippet.

Add system instructions that direct an agent toward autonomous execution.

Generate a structured blog outline with H2/H3s, angle, and CTA from a topic.