User insights
Audit product requirements for ambiguity and missing decisions
Find language that permits incompatible implementations and expose the owner and evidence needed to decide
Open recipeAI tools
Claude prompt templates for clear, structured tasks
Find Claude prompt templates for reading, writing, evidence review, project work, learning, communication, and daily decisions
These recipes provide Claude with explicit context, boundaries, output requirements, and checks. Recipes with a recorded Claude run carry a test badge; the rest are portable prompt templates offered for adaptation and review, not claimed test coverage.
How to use this collection
- Replace every visible placeholder before using a template in Claude.
- Review facts, calculations, sources, and consequential decisions outside the model.
Choose a recipe
318 recipesCustomer support
Audit whether a support response actually answers the customer
Compare a draft response with every customer question, fact, constraint, and requested outcome
Open recipeCreative ideas
Brainstorm multiple directions from a theme and constraints
Generate genuinely different creative directions while respecting audience, facts, resources, exclusions, and decision needs
Open recipeResearch insights
Audit the evidence status of claims
Map each atomic claim to evidence, contradiction, source quality, and a calibrated status
Open recipeTeam growth
Audit recruitment or evaluation language for potential bias
Find vague, exclusionary, stereotyped, or inconsistently applied language and replace it with job-relevant evidence criteria
Open recipeCampaign planning
Audit marketing copy for exaggeration and evidence gaps
Trace factual and implied claims to evidence before copy is published
Open recipeSpreadsheet insights
Audit Excel headers, field definitions, and data types
Compare a workbook with its intended schema before import, analysis, or automation
Open recipeSpreadsheet insights
Audit a chart for misleading design
Review scale, encoding, labels, missing context, and the claim against source values
Open recipeUseful feedback
Audit content for repetition, vagueness, and machine-like writing
Diagnose observable writing problems without unreliable claims about whether AI produced the text
Open recipeClear writing
Audit email tone for harshness, ambiguity, and passive aggression
Identify wording risks in context and revise tone without weakening facts or boundaries
Open recipeLanguage practice
Audit bilingual content for semantic consistency
Find material differences in facts, scope, obligations, tone, and terminology across two language versions
Open recipeRead clearly
Audit a summary against its source
Find unsupported claims, changed meaning, and material omissions before a summary is shared
Open recipe