tauri-eeg

Imported from fordelkon/tauri-eeg (AGENTS.md). Install upstream with npx skills add fordelkon/tauri-eeg. Copyright stays with the author.

tauri-eeg

Multi-agent orchestration framework for agentic coding

Project Overview

A Claude Flow powered project

Tech Stack: TypeScript, Node.js Architecture: Domain-Driven Design with bounded contexts

Quick Start

Installation

npm install

Build

npm run build

Test

npm test

Development

npm run dev

Agent Coordination

Swarm Configuration

This project uses hierarchical swarm coordination for complex tasks:

When to Use Swarms

Invoke swarm for:

• Multi-file changes (3+ files)
• New feature implementation
• Cross-module refactoring
• API changes with tests
• Security-related changes
• Performance optimization

Skip swarm for:

• Single file edits
• Simple bug fixes (1-2 lines)
• Documentation updates
• Configuration changes

Available Skills

Use $skill-name syntax to invoke:

Agent Types

Execution Model

• claude-flow = LEDGER (coordinates: memory, routing, swarm state)
• Codex = EXECUTOR (writes code, runs tests, creates files)

Critical rule: DON'T STOP after calling claude-flow commands. Coordination commands return instantly — continue immediately with the next implementation step.

MCP Integration

Use MCP tools for coordination, then keep coding:

Code Standards

File Organization

• NEVER save to root folder
• /src - Source code files
• /tests - Test files
• /docs - Documentation
• /config - Configuration files

Quality Rules

• Files under 500 lines
• No hardcoded secrets
• Input validation at boundaries
• Typed interfaces for public APIs
• TDD London School (mock-first) preferred

Commit Messages

<type>(<scope>): <description>

[optional body]

Co-Authored-By: claude-flow <ruv@ruv.net>

Types: feat, fix, docs, style, refactor, perf, test, chore

Security

Critical Rules

• NEVER commit secrets, credentials, or .env files
• NEVER hardcode API keys
• Always validate user input
• Use parameterized queries for SQL
• Sanitize output to prevent XSS

Path Security

• Validate all file paths
• Prevent directory traversal (../)
• Use absolute paths internally

Memory System

Storing Patterns

npx @claude-flow/cli memory store \
  --key "pattern-name" \
  --value "pattern description" \
  --namespace patterns

Searching Memory

npx @claude-flow/cli memory search \
  --query "search terms" \
  --namespace patterns

Quick Commands

npx @claude-flow/cli memory search --query "relevant patterns"
npx @claude-flow/cli hooks route --task "current task description"
npx @claude-flow/cli swarm init --topology hierarchical
npx @claude-flow/cli hooks pre-task --description "task summary"

Links

• Documentation: https://github.com/ruvnet/ruflo
• Issues: https://github.com/ruvnet/ruflo/issues

Performance Targets

Testing

Running Tests

# Unit tests
npm test

# Integration tests
npm run test:integration

# Coverage
npm run test:coverage

# Security tests
npm run test:security

Test Philosophy

• TDD London School (mock-first)
• Unit tests for business logic
• Integration tests for boundaries
• E2E tests for critical paths
• Security tests for sensitive operations

Coverage Requirements

• Minimum 80% line coverage
• 100% coverage for security-critical code
• All public APIs must have tests

MCP Integration

Claude Flow exposes tools via Model Context Protocol:

# Start MCP server
npx ruflo mcp start

# List available tools
npx ruflo mcp tools

Available Tools

Hooks System

Claude Flow uses hooks for lifecycle automation:

Core Hooks

Session Hooks

Intelligence Hooks

Example Usage

# Before starting a task
npx @claude-flow/cli hooks pre-task \
  --description "implementing authentication"

# After completing a task
npx @claude-flow/cli hooks post-task \
  --task-id "task-123" \
  --success true

# Route a task to agents
npx @claude-flow/cli hooks route \
  --task "implement OAuth2 login flow"

Background Workers

12 background workers provide continuous optimization:

Managing Workers

# List workers
npx @claude-flow/cli hooks worker list

# Trigger specific worker
npx @claude-flow/cli hooks worker dispatch --trigger audit

# Check worker status
npx @claude-flow/cli hooks worker status

Intelligence System

The RuVector Intelligence System provides neural learning:

Components

• SONA: Self-Optimizing Neural Architecture (<0.05ms adaptation)
• MoE: Mixture of Experts for specialized routing
• HNSW: Hierarchical Navigable Small World for fast search
• EWC++: Elastic Weight Consolidation (prevents forgetting)
• Flash Attention: Optimized attention mechanism

4-Step Pipeline

1. RETRIEVE - Fetch relevant patterns via HNSW
2. JUDGE - Evaluate with verdicts (success/failure)
3. DISTILL - Extract key learnings via LoRA
4. CONSOLIDATE - Prevent catastrophic forgetting via EWC++

Debugging

Log Levels

# Set log level
export CLAUDE_FLOW_LOG_LEVEL=debug

# Enable verbose mode
npx @claude-flow/cli --verbose <command>

Health Checks

# Run diagnostics
npx @claude-flow/cli doctor --fix

# Check system status
npx @claude-flow/cli status