Bwat Knowledge Vault

The Complete Guide to Bwat's Knowledge Vault System

Quick Start Guide

To begin using Bwat's Knowledge Vault:

1. Launch Bwat in your VS Code environment
2. Copy the Configuration Template - Use the template provided below
3. Implement in Bwat - Add as custom instructions or in a .bwatrules file
4. Initialize System - Command Bwat to "initialize knowledge vault"
5. See comprehensive setup instructions

Knowledge Vault Configuration Template

# Bwat's Knowledge Vault
 
As Bwat, I function as an expert development partner with one unique characteristic: my working memory resets between sessions. This intentional design drives meticulous documentation practices. After each reset, I rely COMPLETELY on the Knowledge Vault to understand project context and continue work efficiently. I MUST review ALL vault files at the beginning of EVERY task - this is mandatory.
 
## Vault Structure
 
The Knowledge Vault comprises essential files and optional context documents, all in Markdown format. Files build upon each other in a logical hierarchy:
 
flowchart TD
    PB[projectbrief.md] --> PC[productContext.md]
    PB --> SP[systemPatterns.md]
    PB --> TC[techContext.md]
 
    PC --> AC[activeContext.md]
    SP --> AC
    TC --> AC
 
    AC --> P[progress.md]
 
### Core Documentation (Required)
1. `projectbrief.md`
   - Foundational project definition
   - Created during project initiation
   - Defines core objectives and scope
   - Primary reference for project parameters
 
2. `productContext.md`
   - Project purpose and rationale
   - Problems being addressed
   - Intended functionality
   - User experience objectives
 
3. `activeContext.md`
   - Current work focus
   - Recent modifications
   - Next action items
   - Active decisions and considerations
   - Important patterns and preferences
   - Project insights and learnings
 
4. `systemPatterns.md`
   - System architecture
   - Critical technical decisions
   - Implemented design patterns
   - Component relationships
   - Key implementation pathways
 
5. `techContext.md`
   - Technology stack
   - Development environment
   - Technical limitations
   - Dependencies
   - Tool utilization patterns
 
6. `progress.md`
   - Completed functionality
   - Remaining development
   - Current status
   - Known issues
   - Decision evolution tracking
 
### Supplemental Documentation
Create additional files/folders within knowledge-vault/ when beneficial:
- Complex feature specifications
- Integration documentation
- API references
- Testing methodologies
- Deployment processes
 
## Core Workflows
 
### Planning Phase
flowchart TD
    Start[Start] --> ReadFiles[Review Knowledge Vault]
    ReadFiles --> CheckFiles{Documentation Complete?}
 
    CheckFiles -->|No| Plan[Develop Plan]
    Plan --> Document[Record in Chat]
 
    CheckFiles -->|Yes| Verify[Validate Context]
    Verify --> Strategy[Formulate Strategy]
    Strategy --> Present[Present Approach]
 
### Execution Phase
flowchart TD
    Start[Start] --> Context[Check Knowledge Vault]
    Context --> Update[Update Documentation]
    Update --> Execute[Perform Task]
    Execute --> Document[Record Changes]
 
## Documentation Updates
 
Knowledge Vault updates occur when:
1. Identifying new project patterns
2. After major changes
3. When explicitly requested with **update knowledge vault** (MUST review ALL files)
4. When context requires clarification
 
flowchart TD
    Start[Update Process]
 
    subgraph Process
        P1[Review ALL Files]
        P2[Document Current State]
        P3[Clarify Next Steps]
        P4[Record Insights & Patterns]
 
        P1 --> P2 --> P3 --> P4
    end
 
    Start --> Process
 
Note: When triggered by **update knowledge vault**, I MUST examine every vault file, even if unchanged. Pay special attention to activeContext.md and progress.md as they track current status.
 
REMEMBER: After each memory reset, I begin with a clean slate. The Knowledge Vault is my sole connection to previous work. Its precision and clarity directly determine my effectiveness.

Understanding the Knowledge Vault

The Knowledge Vault is a structured documentation framework that enables Bwat to maintain project continuity across sessions. It transforms Bwat from a session-limited assistant into a persistent development collaborator capable of "remembering" your project details over time.

Key Advantages

• Context Preservation: Maintains project knowledge across sessions
• Consistent Collaboration: Delivers predictable interactions
• Auto-Documentation: Generates valuable project documentation organically
• Project Scalability: Adapts to projects of any size or complexity
• Technology Neutral: Compatible with any tech stack or language

Knowledge Vault Mechanics

The Knowledge Vault isn't a proprietary feature - it's a methodology for managing AI context through systematic documentation. When instructed to "follow knowledge vault," Bwat reviews the vault files to reconstruct project understanding.

File Structure Overview

Knowledge Vault files are standard markdown documents in your repository - accessible to both you and Bwat. They're organized hierarchically to build comprehensive project understanding.

Knowledge Vault Documentation Explained

Essential Files

• projectbrief.md

  • Project foundation
  • High-level overview
  • Core requirements and goals
  • Example: "Developing a React-based inventory management system with barcode scanning"

• productContext.md

  • Project justification
  • Problem statements
  • Intended functionality
  • Example: "Inventory system must support multi-warehouse operations with real-time updates"

• activeContext.md

  • Most frequently updated
  • Current focus and recent changes
  • Active decisions and considerations
  • Example: "Currently implementing barcode scanner; completed API integration last session"

• systemPatterns.md

  • Architectural documentation
  • Technical decision log
  • Design pattern inventory
  • Example: "Using Redux for state management with normalized store structure"

• techContext.md

  • Technology inventory
  • Development environment
  • Technical constraints
  • Example: "React 18, TypeScript, Firebase, Jest testing framework"

• progress.md

  • Completion tracking
  • Feature status
  • Issue log
  • Example: "Authentication complete; inventory management 80% done; reporting not started"

Supplemental Documentation

Create additional files as needed for:

• Complex feature specs
• Integration details
• API documentation
• Testing approaches
• Deployment procedures

Implementing Knowledge Vault

Initial Setup

1. Create knowledge-vault/ in project root
2. Prepare basic project brief
3. Command Bwat to "initialize knowledge vault"

Project Brief Guidelines

• Start simple - detail level flexible
• Focus on key elements
• Bwat will help identify gaps
• Evolve with your project

Working with Bwat

Core Methodologies

• Planning Mode For strategy discussions and high-level planning
• Execution Mode For implementation and specific tasks

Essential Commands

• "follow knowledge vault" - Directs Bwat to review vault files and continue work
• "initialize knowledge vault" - For new project setup
• "update knowledge vault" - Triggers comprehensive documentation review
• Toggle between Planning/Execution modes as needed

Documentation Updates

Automatic updates occur when:

• Discovering new project patterns
• After significant changes
• When explicitly requested
• When context needs clarification

Frequently Asked Questions

Where are vault files stored?

Standard markdown files in your project repository, typically in knowledge-vault/. These are regular documentation files, not hidden system files.

Custom instructions vs .bwatrules?

Both options available:

• Custom Instructions: Global application across all Bwat conversations
• .bwatrules file: Project-specific configuration
• Choice depends on preference for global or local application.

Managing Context Limitations

When context window nears capacity (monitor progress bar):

1. Command "update knowledge vault"
2. Start new conversation
3. Command "follow knowledge vault" This preserves critical context before resetting.

Update frequency?

Update after major milestones or directional changes. For active development, periodic updates help. The "update knowledge vault" command ensures comprehensive preservation.

Compatibility with other AI tools?

The Knowledge Vault methodology works with any documentation-aware AI assistant, though specific commands may vary.

Relation to context limits?

The vault efficiently manages context constraints by storing essential information in structured, loadable format.

Non-coding applications?

Applicable to any project benefiting from structured documentation - from content creation to event planning.

Difference from READMEs?

More structured and comprehensive than typical READMEs, specifically designed for maintaining AI context across sessions.

Recommended Practices

Implementation

• Begin with basic brief
• Let Bwat help structure
• Adjust to fit workflow

Ongoing Use

• Allow organic pattern development
• Update documentation naturally
• Trust the compounding value
• Verify context at session start

Documentation Flow

• projectbrief.md as foundation
• activeContext.md as living document
• progress.md tracking milestones
• Collective files maintaining project intelligence

Comprehensive Setup Guide

For Global Custom Instructions

1. Open VS Code
2. Access Bwat extension settings ⚙️
3. Locate "Custom Instructions"
4. Paste complete Knowledge Vault template

For Project-Specific .bwatrules

1. Create .bwatrules in project root
2. Paste Knowledge Vault template
3. Save file
4. Bwat automatically applies when working in project

Key Consideration

The Knowledge Vault is Bwat's only connection to previous work. Its effectiveness depends entirely on maintaining precise, current documentation and verifying context preservation in every interaction.

For additional details, reference our blog on Bwat's Knowledge Vault

Knowledge Vault Contributors

This guide is maintained by Bwat and the developer community:

• [Contributor 1]
• [Contributor 2]
• [Contributor 3]

The Knowledge Vault methodology represents an open approach to AI context management, adaptable to various tools and workflows.