Better prompts
Diagnose why a prompt performs poorly
Compare intent, inputs, instructions, and real outputs to identify evidenced failure modes instead of guessing
Open recipe
Browse prompts for writing, research, learning, and image creation. Add your context, inspect the full prompt, and use it with your preferred AI assistant.
Better prompts
Compare intent, inputs, instructions, and real outputs to identify evidenced failure modes instead of guessing
Open recipeStudy smarter
Use answer evidence to separate concept, procedure, interpretation, and careless errors before planning remediation
Open recipeWork together
Create psychologically safer, evidence-seeking questions that move from outcomes and context to controllable experiments
Open recipeUser insights
Turn research questions into realistic, non-leading tasks with success evidence, observations, prompts, and prototype limits
Open recipeLesson planning
Sequence preparation, guided application, feedback, and later retrieval around one learning target
Open recipeBetter prompts
Turn a prompt contract into repeatable cases with expected behavior, evidence rules, and pass criteria
Open recipeBetter emails
Build an evidence-centered question sequence that opens participation and deepens reasoning
Open recipeSpreadsheet insights
Turn a forecasting request into a testable analysis plan with baselines, time splits, leakage controls, uncertainty, and decision limits
Open recipeDaily operations
Connect sales, stock movements, purchasing, receiving, exceptions, and reconciliation into one controlled workflow
Open recipeSpreadsheet insights
Track usable stock, demand, lead time, open supply, reorder logic, and exception ownership without inventing purchasing decisions
Open recipeSpreadsheet insights
Map a decision question to rows, columns, values, filters, and validation totals
Open recipeSpreadsheet insights
Structure assignment, exposure, outcomes, exclusions, uncertainty, and decision rules without overstating a test
Open recipeMany paths, one reliable recipe
Browse by type, task, audience, tool, or collection. Inspect the examples and guidance for each recipe.
Start with what you want the AI to produce
Find a recipe for the outcome you need now
Start from the result you need to produce
Find prompts for recurring professional work
Build skills, teach, and prepare for career moves
Plan everyday decisions and personal projects
Curated starting points for common roles and goals
Explore visual styles and practical recipe workflows
Adapt portable recipes and see recorded test status
FAQ
Learn how to choose a recipe, add useful context, and review the result.
An AI prompt recipe is a reusable set of instructions for a specific task, such as drafting a message, analyzing a document, or planning a meeting. AILesson recipes combine a prompt template with context fields, a worked example, and checks you can use to review the AI response.
Open a recipe that matches your task, fill in the context fields, and inspect the assembled prompt. Copy it into your preferred AI assistant, then use the recipe's checks to review the response. Unfilled fields remain as placeholders, so replace or remove them before sending the prompt.
The recipes use plain-language instructions designed to be portable across AI assistants. Check each recipe's test record for the tools actually tested. Results can vary by model and version, and tasks involving files or other inputs require an assistant that supports those inputs.
Search by the task or outcome you need, or browse by prompt type, task, audience, AI tool, or workflow. Compare the expected result and worked example with your goal before choosing a recipe. Add your own source material, constraints, and output requirements rather than copying the example unchanged.
No. Prompt assembly happens in your browser; the recipe builder does not send your field inputs to an AILesson server or an AI model. If you paste the finished prompt into another service, that service handles the information under its own policies. Only include information you are allowed to share with it.
No. A worked example and a test record show how a recipe was used, not a guarantee of future results. Check facts, source references, calculations, and whether the response follows your constraints. Use the recipe's review checklist and verify time-sensitive claims before relying on the output.