> ## Documentation Index
> Fetch the complete documentation index at: https://docs.weventures.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Claude Code Prompting Guide

> Master the art of crafting powerful prompts for building production-ready applications with Claude Code

<Hero>
  Master Claude Code Prompting

  Transform your ideas into production-ready applications with precision prompting
</Hero>

<CardGroup cols={3}>
  <Card title="Explicit Instructions" icon="code" color="#8B5CF6">
    Never let Claude guess. Be explicit about every library, framework, and service.
  </Card>

  <Card title="Structured Planning" icon="map" color="#10B981">
    Start every project with comprehensive planning documents.
  </Card>

  <Card title="Agent Collaboration" icon="users" color="#F59E0B">
    Leverage specialized agents for complex multi-step tasks.
  </Card>
</CardGroup>

***

## 🎯 Core Philosophy

<Note>
  **The Golden Rule:** Every successful Claude Code project starts with thorough discovery, followed by `claude.md` and `plan.md`. These driver documents guide the entire development process.
</Note>

<Info>
  You are the architect and supervisor. Claude Code is your implementation partner that executes your vision with precision when given clear, detailed instructions.
</Info>

## 📚 Complete Guide Sections

<CardGroup cols={2}>
  <Card title="Discovery Phase" icon="magnifying-glass" href="/ai-development/prompting/discovery" color="#8B5CF6">
    Before you prompt - understand your project requirements thoroughly
  </Card>

  <Card title="Planning Documents" icon="file-code" href="/ai-development/prompting/planning" color="#10B981">
    Master claude.md and plan.md templates for structured development
  </Card>

  <Card title="Frontend Excellence" icon="palette" href="/ai-development/prompting/frontend" color="#F59E0B">
    Component architecture, design systems, and responsive development
  </Card>

  <Card title="Backend Strategy" icon="database" href="/ai-development/prompting/backend" color="#3B82F6">
    Architecture patterns for Supabase, FastAPI, and LangGraph
  </Card>

  <Card title="Testing & Quality" icon="flask" href="/ai-development/prompting/testing" color="#EF4444">
    Comprehensive testing strategies and quality assurance
  </Card>

  <Card title="Deployment & DevOps" icon="rocket" href="/ai-development/prompting/deployment" color="#9333EA">
    Production deployment, monitoring, and optimization
  </Card>

  <Card title="Agent Collaboration" icon="users-gear" href="/ai-development/prompting/agents" color="#14B8A6">
    Multi-agent workflows for complex tasks
  </Card>

  <Card title="Patterns & Reference" icon="book" href="/ai-development/prompting/reference" color="#6B7280">
    Common patterns, quick reference, and production checklist
  </Card>
</CardGroup>

## 🚀 Quick Start Process

<Steps>
  <Step title="Complete Discovery Phase">
    Thoroughly understand project requirements, audience, and constraints using our [discovery guide](/ai-development/prompting/discovery)
  </Step>

  <Step title="Create Planning Documents">
    Always begin by creating `claude.md` and `plan.md` with explicit requirements using our [templates](/ai-development/prompting/planning)
  </Step>

  <Step title="Specify Tech Stack">
    Be explicit about versions, libraries, and frameworks - never let Claude guess
  </Step>

  <Step title="Phase Implementation">
    Break down the project into independently testable phases
  </Step>

  <Step title="Iterate with Context">
    Provide rich context, errors, and logs when debugging
  </Step>
</Steps>

## 📋 Core Principles

<AccordionGroup>
  <Accordion title="Clarity Over Brevity" icon="eye">
    Provide comprehensive context rather than minimal instructions. Claude Code performs best with detailed, explicit guidance.
  </Accordion>

  <Accordion title="Structure Over Chaos" icon="sitemap">
    Use consistent formatting, templates, and patterns across all prompts for predictable results.
  </Accordion>

  <Accordion title="Libraries Over Custom Code" icon="box">
    Always prefer battle-tested open-source libraries over custom implementations. This ensures reliability and maintainability.
  </Accordion>

  <Accordion title="Context Over Assumptions" icon="info-circle">
    Include exact error messages, current state, and attempted solutions when debugging.
  </Accordion>

  <Accordion title="Planning Over Improvisation" icon="map">
    Invest time in planning to save time in implementation. A well-thought-out plan prevents costly refactoring.
  </Accordion>
</AccordionGroup>

## ⚡ Pro Tips

<Info>
  Save successful prompts as templates and build a library of patterns that work for your specific use cases.
</Info>

<Tip>
  Use `ultrathink` for complex architecture decisions, `think deeply` for data modeling, and `think` for user experience considerations.
</Tip>

<Warning>
  Always specify your tech stack explicitly - don't let AI guess what libraries you're using.
</Warning>

<Note>
  Small, atomic prompts work better than large, complex ones. Break down big tasks into manageable steps.
</Note>

## 🎯 Quick Reference Templates

<Tabs>
  <Tab title="🚀 New Project">
    ```text theme={null}
    ultrathink and create [PROJECT NAME]
    incorporating all discovery findings:
    [Paste discovery summary]

    Start with claude.md and plan.md
    Use Next.js 14, Supabase, shadcn/ui
    [Specific stack from discovery]
    ```
  </Tab>

  <Tab title="✨ Add Feature">
    ```text theme={null}
    Add [FEATURE] to existing project

    Context from discovery:
    - User need: [from user research]
    - Priority: [from stakeholder input]
    - Success metric: [from KPIs]

    Following patterns in:
    [reference files]

    Include:
    - Tests for [critical paths from user journey]
    - Error handling for [edge cases from discovery]
    - Analytics for [KPIs from discovery]
    ```
  </Tab>

  <Tab title="🐛 Debug Issue">
    ```text theme={null}
    Debug this issue:

    Error: [exact error message]
    File: [path/to/file:line_number]
    User Impact: [from discovery priorities]
    Frequency: [from monitoring]

    Context: [when it occurs in user journey]
    Code: [relevant snippet]

    Attempted solutions:
    1. [what you tried]
    2. [another attempt]

    Consider constraints from discovery:
    - [Performance requirement]
    - [Security requirement]
    ```
  </Tab>

  <Tab title="⚡ Optimize">
    ```text theme={null}
    Optimize [component/feature]

    Current metrics:
    - [Current performance data]

    Target from discovery:
    - [Performance budget]
    - [User expectations]

    Constraints:
    - [Technical limitations]
    - [Time/resource limits]

    Provide specific improvements
    ```
  </Tab>

  <Tab title="📚 Document">
    ```text theme={null}
    Document [feature/system]

    For audiences from discovery:
    - [End users - skill level]
    - [Developers - team size]
    - [Admins - technical level]

    Include:
    - Setup instructions
    - Usage examples
    - Troubleshooting
    - API reference
    ```
  </Tab>
</Tabs>

***

<Note>
  **Remember:** Success with Claude Code comes from thorough discovery, explicit instructions, and structured planning. You're the architect - Claude Code is your precise implementation partner.
</Note>

<Card title="Start Your Journey" icon="rocket" color="#FFD700">
  Ready to master Claude Code prompting? Begin with our comprehensive [Discovery Phase Guide](/ai-development/prompting/discovery) to ensure project success from the start.
</Card>