APEX — The Master Website Build Prompt
Fill a short brief, paste the whole file, and the AI designs and builds one complete award-grade website — no lorem ipsum, no stalling.
Use when: Any new website build where you want an authored piece, not a template instantiation.
WebsiteArt DirectionWebGLAwwwards
The Anti-Slop Mandate
Refines a website into a distinctive, senior-crafted product — with an explicit blacklist of the clichés that make a site read as AI-generated.
Use when: An existing site or draft that looks templated and needs a principal-level craft pass.
WebsiteDesign ReviewCraftAnti-Slop
The Cosmic Master Prompt — Award-Grade Builder
Answer an intake, paste, and Claude Code plans, builds, then screenshots and judges its own work across three self-review passes.
Use when: Immersive, experience-led sites where you want the builder to critique and iterate on itself.
WebsiteWebGLSelf-ReviewClaude Code
APEX Part B — Build Directive Only
The execution half of APEX: craft system, archetype paradigms, performance and accessibility budgets, build order and acceptance checklist.
Use when: You already have a brief written — paste this to set the quality bar and build order.
WebsiteCraft SystemAcceptance Criteria
Few-Shot Chain-of-Thought Solver
Use when you can supply a few worked examples and want the model to imitate their step-by-step reasoning on a new problem of the same type.
reasoningcotfew-shotverification
Zero-Shot Step-by-Step Reasoner
Use for any single hard question when you have no examples to show - it upgrades the classic 'think step by step' trigger into a full scaffold with a…
reasoningcotzero-shot
Self-Consistency Majority Voting
Use when a single chain of reasoning is error-prone - the model produces several independent solution paths and the majority answer wins.
reasoningreliabilityvoting
Generated Knowledge Grounding
Use for commonsense or factual questions where hidden wrong assumptions cause errors - the model must list and grade relevant facts before it is allowed to…
reasoninggroundingfactsanti-hallucination
ReAct Tool-Use Agent Loop
Use to turn any LLM into a tool-using agent that alternates explicit reasoning with actions and never fabricates tool results - the backbone of most agent…
agentstoolsreactgrounding
Tree-of-Thoughts Expert Panel
Use for hard open-ended questions where the first idea is often wrong - three simulated experts explore different branches, attack each other, and prune dead…
reasoningtotdebateexploration
Plan-and-Solve Structured Reasoning
Use for multi-step quantitative or logical problems - forcing a complete plan before any calculation catches the setup errors that plain step-by-step…
reasoningplanningmathzero-shot
Least-to-Most Decomposition Ladder
Use when a problem is too hard to answer in one pass - it gets broken into a ladder of subquestions, easiest first, with each answer feeding the next.
reasoningdecompositioncomplex-problems
Prompt Template Generator
Use to have the model design a complete, reusable prompt template for any task you describe - it engineers the inputs, structure, guardrails, and output…
meta-prompttemplatesprompt-design
Four-Block System Prompt Skeleton
Use as the starting skeleton for any assistant's system message - identity, testable rules, demonstrations, and reference data in the order models parse best.
system-promptstructureassistants
Agent Persistence Directive
Drop this block into any agent's system prompt to stop it from quitting halfway - it enforces decomposition, reflection after every tool call, and a…
agentsautonomyreliability
One-Shot Web App Builder
Use to get a complete, polished, dependency-free web app in a single response - the model grades its own draft against a hidden rubric and iterates before…
codingfrontendone-shotrubric
LangGPT Structured Role Card
Use when you want a reusable, maintainable persona prompt - the sectioned role card gives the model a stable identity, bounded skills, testable rules, and a…
rolestructurepersonaframework
CO-STAR Communication Framework
Use for any writing task where voice and audience matter - posts, emails, landing pages, scripts - by forcing the context, objective, style, tone, audience,…
frameworkcopywritingaudiencetone
RACE Professional Deliverable Framework
Use for work deliverables with a clear done standard - reports, reviews, runbooks - because the Expectation section defines measurable success criteria before…
frameworkbusinessdeliverablesacceptance-criteria
Iterative Prompt Perfector
Use to co-write a prompt with the model itself - each round it delivers a full revised prompt, explains its changes, and asks the highest-leverage questions…
meta-promptiterationcollaboration
Configurable AI Tutor
Use to turn any LLM into a personal tutor whose depth, teaching style, and tone you configure - with slash commands for planning lessons, teaching, and testing.
educationpersonacommandsadaptive
Screenshot to Pixel-Perfect Code
Use when you have a screenshot of a UI and need it rebuilt in code with exact colors, spacing, and copy instead of a loose approximation.
design-to-codetailwindreactreplication
Figma Frame to Production Page
Use when converting a Figma design into a responsive, semantic, token-driven page instead of a brittle absolute-positioned copy.
figmadesign-to-codenextjsresponsive
High-Converting Landing Page Architect
Use when designing or rewriting a single-offer landing page and you want conversion strategy, real copy, and implementation in one pass.
landing-pageconversioncopywritingsaas