SPT - Quick Reference

⚡ Quick Start

bash

# Load a single skill
claude-code --skill skill-name "Your task"

# Load multiple skills (recommended for complex tasks)
claude-code --skill skill1 --skill skill2 "Complex task"

# Example
claude-code --skill pair-programming --skill verification-quality "Build login form"

🎯 SPT Skills (25 Total)

🧠 SptDB Skills - Vector Search & Memory (5)

sptdb-vector-search: Semantic search, RAG systems, knowledge bases (<100µs)
sptdb-memory-patterns: Persistent memory, session management, context preservation
sptdb-optimization: 4-32x memory reduction, 150x faster search, HNSW indexing
sptdb-learning: 9 RL algorithms (Decision Transformer, Q-Learning, Actor-Critic, SARSA)
sptdb-advanced: QUIC sync (<1ms), multi-database coordination, hybrid search, distributed

🔄 Flow Nexus Skills - Cloud Platform (3)

flow-nexus-platform: Authentication, sandboxes, deployment, payments, challenges
flow-nexus-swarm: Cloud swarms, event-driven workflows, message queues
flow-nexus-neural: Train & deploy neural networks in E2B sandboxes

🐙 Swarm Skills - Multi-Agent Orchestration (3)

swarm-orchestration: Multi-agent coordination, parallel execution, dynamic topology
swarm-advanced: Research, development, testing, complex workflows
hive-mind-advanced: Queen coordinator, worker specialists, adaptive swarms

🔧 Workflow Skills - Automation (4)

stream-chain: Stream-JSON chaining, pipelines, data transformation
hooks-automation: Pre/post hooks, Git integration, session management, MCP coordination
smart-methodology: Specification → Modeling → Architecture → Refinement → Trust-Validation
skill-builder: Generate React components (TypeScript, hooks, tests, Storybook)

📊 GitHub Skills - Integration & DevOps (5)

github-code-review: Multi-agent PR review, security analysis, best practices
github-project-management: Issue tracking, boards, sprints, Agile workflow
github-release-management: Automated versioning, testing, deployment, rollback
github-workflow-automation: CI/CD optimization, parallelization, artifact management
github-multi-repo: Cross-repo sync, package alignment, architecture management

🤝 Development Skills - Quality & Performance (3)

pair-programming: Driver/navigator modes, TDD, refactoring, debugging, verification
verification-quality: Code quality checks, automated testing, validation workflows
performance-analysis: Profiling, bottleneck detection, optimization, benchmarking

🧩 ReasoningBank Skills - Adaptive Learning (2)

reasoningbank-intelligence: Pattern recognition, strategy optimization, meta-cognition
reasoningbank-sptdb: Trajectory tracking, verdict judgment, 150x speedup, experience replay

📚 Skill Selection Quick Guide

Building something? Use these:

Task	Skills
Semantic search/RAG	sptdb-vector-search + sptdb-optimization
React component	skill-builder + pair-programming + verification-quality
Code review	github-code-review + pair-programming
Release/deploy	github-release-management + github-workflow-automation
Multi-repo work	github-multi-repo + swarm-orchestration
Self-learning agent	sptdb-learning + reasoningbank-intelligence
Complex system	smart-methodology + swarm-orchestration
Cloud scale	flow-nexus-platform + flow-nexus-swarm

🔄 Legacy Commands (For Reference)

Agent Spawning

bash

/spawn <agent-type> <task-description>
/spawn coder "Implement JWT authentication"
/spawn tester "Write unit tests for UserService"
/spawn reviewer "Review PR #123"

Swarm Operations

bash

/swarm-init <strategy> <objective>
/swarm-init hierarchical "Build authentication system"
/swarm-monitor
/swarm-status

SMART Workflow

bash

/smart-architect "Design microservices architecture"
/smart-coder "Implement user service"
/smart-tester "Test API endpoints"
/smart-reviewer "Review implementation"
/smart-documenter "Generate API docs"

💾 Memory Operations

javascript

// Store data
mcp__spt__memory_usage({
  action: "store",
  key: "path/to/data",
  namespace: "coordination",
  value: JSON.stringify(data)
})

// Retrieve data
mcp__spt__memory_usage({
  action: "retrieve",
  key: "path/to/data",
  namespace: "coordination"
})

// Search data
mcp__spt__memory_usage({
  action: "search",
  query: "search terms",
  namespace: "coordination"
})

📦 Memory Namespaces

coordination/
├── swarm/          # Swarm coordination
│   ├── queen/      # Queen coordinator
│   ├── workers/    # Worker agents
│   └── shared/     # Shared context
├── task/           # Task-specific
└── agent/          # Agent-specific

🛠️ MCP Tools Reference

Swarm Tools

javascript

mcp__flow-nexus__swarm_init({ strategy, objective, max_agents })
mcp__flow-nexus__agent_spawn({ type, task, context })
mcp__flow-nexus__task_orchestrate({ tasks, coordination })

Optimization Tools

javascript

mcp__sublinear-time-solver__solve({ matrix, vector, method })
mcp__sublinear-time-solver__pageRank({ adjacency, dampingFactor })
mcp__sublinear-time-solver__analyzeMatrix({ matrix, checkDominance })

Performance Tools

javascript

mcp__spt__benchmark_run({ type, iterations })
mcp__spt__bottleneck_analyze({ component, metrics })

👥 Agent Metadata & Types

Agent Types

Type	Purpose	Examples
coder	Code implementation	Write features, refactor
planner	Task planning	Break down work, roadmaps
researcher	Research	Research tech, analyze docs
reviewer	Code review	PR review, quality checks
tester	Testing	Unit tests, integration
architect	System design	Architecture, tech selection

Core Agents (agents/core/)

coder - Code implementation
planner - Task breakdown and planning
researcher - Information gathering
reviewer - Code and quality review
tester - Testing and validation

Specialized Agents

GitHub (agents/github/): code-review-swarm, pr-manager, issue-tracker, release-manager Hive Mind (agents/hive-mind/): queen-coordinator, worker-specialist, scout-explorer, swarm-memory-manager Consensus (agents/consensus/): raft-manager, quorum-manager, byzantine-coordinator, crdt-synchronizer SMART (agents/smart/): specification, modeling, architecture, refinement

Agent Metadata Template

yaml

---
name: my-agent
type: developer|coordinator|analyst|specialist
color: "#FF6B35"
description: Brief description
capabilities:
  - capability_1
  - capability_2
priority: high|medium|low
hooks:
  pre: |
    echo "Starting..."
  post: |
    echo "Complete!"
---

🔗 Common Commands

Analysis

bash

/bottleneck-detect <component>
/performance-report
/token-usage

GitHub

bash

/github/pr-manager review <PR#>
/github/issue-tracker triage <repo>
/github/release-manager create <version>

Hive Mind

bash

/hive-mind-init <strategy>
/hive-mind-spawn <type> <count>
/hive-mind-consensus <decision>
/hive-mind-status

Optimization

bash

/topology-optimize
/parallel-execute <tasks>
/cache-manage

🎯 Swarm Strategies

Strategy	Best For	Organization
Hierarchical	Complex projects	Tree with coordinator
Mesh	Distributed tasks	Peer-to-peer network
Adaptive	Dynamic workloads	Auto-scaling topology

📐 File Organization

src/
  modules/
    feature/
      feature.service.ts      # Business logic
      feature.controller.ts   # HTTP handling
      feature.repository.ts   # Data access
      feature.types.ts        # Type definitions
      feature.test.ts         # Tests

✅ Code Quality Checklist

[ ] Clear naming conventions
[ ] Single Responsibility Principle
[ ] Proper error handling
[ ] TypeScript types defined
[ ] Tests written (>80% coverage)
[ ] Documentation added
[ ] Linter passing
[ ] Security validated

🚨 Error Handling Pattern

typescript

try {
  const result = await operation();
  return result;
} catch (error) {
  logger.error('Operation failed', { error, context });
  
  if (error instanceof ValidationError) {
    throw new BadRequestError('Invalid input', error);
  }
  
  throw new InternalServerError('Unexpected error', error);
}

🪝 Hook Examples

bash

# Pre-hook: Validate environment
hooks:
  pre: |
    if [ ! -f "package.json" ]; then
      echo "⚠️  No package.json found"
      exit 1
    fi

# Post-hook: Store results
hooks:
  post: |
    memory_store "task_complete" "$(date -Iseconds)"
    npm run lint --if-present

⚙️ Configuration Files

settings.json - Global config

json

{
  "agents": { "max_concurrent": 10 },
  "mcp": { "server_url": "..." },
  "checkpoints": { "enabled": true }
}

settings.local.json - Local overrides

json

{
  "debug": true,
  "agents": { "max_concurrent": 5 }
}

🚀 Helper Scripts

bash

./helpers/quick-start.sh                 # Quick start
./helpers/github-setup.sh                # Setup GitHub
./helpers/setup-mcp.sh                   # Setup MCP
./helpers/checkpoint-manager.sh create <n>    # Create checkpoint
./helpers/checkpoint-manager.sh restore <id>  # Restore checkpoint
./helpers/checkpoint-manager.sh list          # List checkpoints

🔐 Consensus Protocols

Protocol	Consistency	Fault Tolerance	Performance
Raft	Strong	f+1 failures	Medium
Byzantine	Strong	2f+1 failures	Low
Gossip	Eventual	Very High	High
Quorum	Configurable	Configurable	Medium-High
CRDT	Eventual	Very High	Very High

💡 Performance Tips

Load appropriate skills - Choose 2-4 complementary skills
Enable checkpoints - For long-running tasks
Use swarms - For parallel work
Cache results - In memory namespaces
Batch operations - When possible
Monitor performance - Use performance-analysis skill

🔒 Security Best Practices

Never hardcode secrets
Use environment variables
Validate all inputs
Sanitize outputs
Implement proper auth/authz
Use memory namespaces for isolation

🐛 Troubleshooting

Skill won't load:

Verify skill name spelling
Check prerequisites (Node 18+, gh CLI, MCP)
Review MCP connection

Agent won't spawn:

Check metadata syntax
Verify capabilities
Review memory limits

Memory errors:

Check MCP connection
Verify namespace permissions
Review memory quotas

Slow performance:

Load performance-analysis skill
Enable checkpoints
Reduce concurrent agents

📚 Quick Links

Skills: --skill flag in Claude Code
Agents: agents/**/*.md
Commands: commands/
Helpers: helpers/
Settings: settings.json
MCP Server: flow-nexus (required for cloud features)

SPT Quick Reference v3.0Optimized for Claude Code Skills

Last updated: 2025 For detailed docs: See individual SKILL.md files in skills/ directory

SPT - Quick Reference ​

⚡ Quick Start ​

🎯 SPT Skills (25 Total) ​

🧠 SptDB Skills - Vector Search & Memory (5) ​

🔄 Flow Nexus Skills - Cloud Platform (3) ​

🐙 Swarm Skills - Multi-Agent Orchestration (3) ​

🔧 Workflow Skills - Automation (4) ​

📊 GitHub Skills - Integration & DevOps (5) ​

🤝 Development Skills - Quality & Performance (3) ​

🧩 ReasoningBank Skills - Adaptive Learning (2) ​

📚 Skill Selection Quick Guide ​

🔄 Legacy Commands (For Reference) ​

Agent Spawning ​

Swarm Operations ​

SMART Workflow ​

💾 Memory Operations ​

📦 Memory Namespaces ​

🛠️ MCP Tools Reference ​

Swarm Tools ​

Optimization Tools ​

Performance Tools ​

👥 Agent Metadata & Types ​

Agent Types ​

Core Agents (agents/core/) ​

Specialized Agents ​

Agent Metadata Template ​

🔗 Common Commands ​

Analysis ​

GitHub ​

Hive Mind ​

Optimization ​

🎯 Swarm Strategies ​

📐 File Organization ​

✅ Code Quality Checklist ​

🚨 Error Handling Pattern ​

🪝 Hook Examples ​

⚙️ Configuration Files ​

🚀 Helper Scripts ​

🔐 Consensus Protocols ​

💡 Performance Tips ​

🔒 Security Best Practices ​

🐛 Troubleshooting ​

📚 Quick Links ​

SPT - Quick Reference

⚡ Quick Start

🎯 SPT Skills (25 Total)

🧠 SptDB Skills - Vector Search & Memory (5)

🔄 Flow Nexus Skills - Cloud Platform (3)

🐙 Swarm Skills - Multi-Agent Orchestration (3)

🔧 Workflow Skills - Automation (4)

📊 GitHub Skills - Integration & DevOps (5)

🤝 Development Skills - Quality & Performance (3)

🧩 ReasoningBank Skills - Adaptive Learning (2)

📚 Skill Selection Quick Guide

🔄 Legacy Commands (For Reference)

Agent Spawning

Swarm Operations

SMART Workflow

💾 Memory Operations

📦 Memory Namespaces

🛠️ MCP Tools Reference

Swarm Tools

Optimization Tools

Performance Tools

👥 Agent Metadata & Types

Agent Types

Core Agents (agents/core/)

Specialized Agents

Agent Metadata Template

🔗 Common Commands

Analysis

GitHub

Hive Mind

Optimization

🎯 Swarm Strategies

📐 File Organization

✅ Code Quality Checklist

🚨 Error Handling Pattern

🪝 Hook Examples

⚙️ Configuration Files

🚀 Helper Scripts

🔐 Consensus Protocols

💡 Performance Tips

🔒 Security Best Practices

🐛 Troubleshooting

📚 Quick Links