SwarmClaw Integration Plugin

Version: 1.0.0
License: MIT
Status: Production Ready

Multi-provider LLM integration plugin for Heretek OpenClaw with automatic failover, health monitoring, and provider statistics.

Features

• Multi-Provider Support: OpenAI, Anthropic, Google Gemini, Ollama (local)
• Automatic Failover: Seamless provider switching on failure (OpenAI → Anthropic → Google → Ollama)
• Health Monitoring: Continuous provider health checks with configurable thresholds
• Provider Statistics: Request counts, latency tracking, success rates
• Event-Driven: Real-time events for failover, recovery, and status changes
• TypeScript Support: Full type definitions included

Quick Start

1. Install Dependencies

cd plugins/swarmclaw-integration
npm install

2. Configure Environment

cp .env.example .env
# Edit .env with your API keys

3. Initialize Plugin

import { createPlugin } from '@heretek-ai/swarmclaw-integration-plugin';

const plugin = await createPlugin({
  failoverOrder: ['openai', 'anthropic', 'google', 'ollama'],
  healthCheckInterval: 30000
});

// Send a chat message
const response = await plugin.chat([
  { role: 'user', content: 'Hello!' }
]);

console.log(`Response from ${response.provider}: ${response.content}`);

Architecture

┌─────────────────────────────────────────────────────────────────┐
│                    SwarmClaw Plugin                              │
│                                                                  │
│  ┌─────────────────────────────────────────────────────────┐    │
│  │                   Failover Manager                       │    │
│  │  ┌─────────┐  ┌─────────┐  ┌─────────┐  ┌─────────┐    │    │
│  │  │ OpenAI  │→│Anthropic│→│ Google  │→│ Ollama  │    │    │
│  │  │ (P0)    │  │ (P1)    │  │ (P2)    │  │ (P3)    │    │    │
│  │  └─────────┘  └─────────┘  └─────────┘  └─────────┘    │    │
│  └─────────────────────────────────────────────────────────┘    │
│                              │                                   │
│              ┌───────────────┼───────────────┐                  │
│              ▼               ▼               ▼                  │
│     ┌────────────────┐ ┌────────────┐ ┌──────────────┐         │
│     │ Provider Config│ │Health Check│ │  Statistics  │         │
│     │                │ │  Manager   │ │   Tracker    │         │
│     └────────────────┘ └────────────┘ └──────────────┘         │
│                                                                  │
│  Events: providerSelected, providerFailed, failoverTriggered    │
│          allProvidersFailed, providerRecovered                  │
└─────────────────────────────────────────────────────────────────┘

Configuration

Environment Variables

| Variable | Description | Default | |----------|-------------|---------| | SWARMCLAW_FAILOVER_ORDER | Comma-separated provider order | openai,anthropic,google,ollama | | OPENAI_API_KEY | OpenAI API key | - | | OPENAI_BASE_URL | OpenAI base URL | https://api.openai.com/v1 | | OPENAI_MODELS | Comma-separated models | gpt-4o,gpt-4-turbo,gpt-3.5-turbo | | ANTHROPIC_API_KEY | Anthropic API key | - | | ANTHROPIC_BASE_URL | Anthropic base URL | https://api.anthropic.com | | ANTHROPIC_MODELS | Comma-separated models | claude-sonnet-4-20250514,... | | GOOGLE_API_KEY | Google API key | - | | GOOGLE_BASE_URL | Google base URL | https://generativelanguage.googleapis.com/v1beta | | GOOGLE_MODELS | Comma-separated models | gemini-2.0-flash,gemini-1.5-pro | | OLLAMA_BASE_URL | Ollama base URL | http://localhost:11434 | | OLLAMA_MODELS | Comma-separated models | llama3.1,qwen2.5,mistral | | HEALTH_CHECK_INTERVAL | Health check interval (ms) | 30000 | | FAILURE_THRESHOLD | Failures before unhealthy | 3 | | SUCCESS_THRESHOLD | Successes before healthy | 2 |

Constructor Options

const plugin = new SwarmClawPlugin({
  // Provider failover order
  failoverOrder: ['openai', 'anthropic', 'google', 'ollama'],
  
  // Retry configuration
  maxRetries: 2,           // Max retries per provider
  retryDelay: 1000,        // Initial retry delay (ms)
  backoffMultiplier: 2,    // Exponential backoff multiplier
  
  // Health check configuration
  healthCheckInterval: 30000,  // Health check interval (ms)
  failureThreshold: 3,         // Failures before unhealthy
  successThreshold: 2,         // Successes before healthy
});

API

Chat

const response = await plugin.chat([
  { role: 'system', content: 'You are helpful.' },
  { role: 'user', content: 'Hello!' }
], {
  model: 'gpt-4o',        // Optional: specific model
  temperature: 0.7,       // Optional: temperature
  maxTokens: 1024,        // Optional: max tokens
  timeout: 30000          // Optional: timeout (ms)
});

// Response format
{
  content: "Hello! How can I help you?",
  role: "assistant",
  usage: { promptTokens: 10, completionTokens: 20, totalTokens: 30 },
  model: "gpt-4o",
  provider: "openai"
}

Embeddings

const result = await plugin.embed('Text to embed', {
  model: 'text-embedding-3-small'
});

// Response format
{
  embedding: [0.1, 0.2, ...],  // Embedding vector
  usage: { promptTokens: 5, totalTokens: 5 },
  model: "text-embedding-3-small",
  provider: "openai"
}

Health Monitoring

// Get all provider statuses
const status = plugin.getStatus();
console.log(status);

// Get specific provider health
const health = plugin.getProviderHealth('openai');
console.log(health);
// { provider: 'openai', status: 'healthy', lastCheck: {...}, ... }

// Get statistics
const stats = plugin.getStats('openai');
console.log(stats);
// { totalRequests: 100, successfulRequests: 98, failedRequests: 2, ... }

Events

// Provider selected for request
plugin.on('providerSelected', (event) => {
  console.log(`Provider: ${event.provider}, Attempt: ${event.attempt}`);
});

// Provider failed
plugin.on('providerFailed', (event) => {
  console.warn(`Provider ${event.provider} failed: ${event.error}`);
});

// Failover triggered
plugin.on('failoverTriggered', (event) => {
  console.warn(`Failover: ${event.fromProvider} → ${event.nextProvider}`);
});

// All providers failed
plugin.on('allProvidersFailed', (event) => {
  console.error(`All providers failed. Tried: ${event.attemptedProviders}`);
});

// Provider recovered
plugin.on('providerRecovered', (event) => {
  console.log(`Provider ${event.provider} recovered: ${event.status}`);
});

Provider Support Matrix

| Provider | Chat | Embeddings | Health Check | Auth Type | |----------|------|------------|--------------|-----------| | OpenAI | ✅ | ✅ | ✅ | Bearer Token | | Anthropic | ✅ | ❌ | ✅ | API Key Header | | Google | ✅ | ✅ | ✅ | Query Param | | Ollama | ✅ | ✅ | ✅ | None |

Failover Logic

Request → Check Provider Health
    │
    ├─→ Healthy? → Send Request
    │       │
    │       ├─→ Success → Return Result
    │       │
    │       └─→ Failure → Retry (maxRetries)
    │               │
    │               └─→ Still Failing → Mark Unhealthy → Next Provider
    │
    └─→ Unhealthy? → Skip → Next Provider

Failover Order

Default order: OpenAI → Anthropic → Google → Ollama

1. OpenAI (Priority 0): Primary provider, GPT-4o, GPT-4 Turbo
2. Anthropic (Priority 1): Secondary, Claude Sonnet, Claude Opus
3. Google (Priority 2): Tertiary, Gemini 2.0, Gemini 1.5
4. Ollama (Priority 3): Local fallback, Llama 3.1, Qwen 2.5

Testing

# Run tests
npm test

# Run health check
npm run healthcheck

# Lint
npm run lint

Integration Examples

With OpenClaw Gateway

// In your agent workspace
import { createPlugin } from '@heretek-ai/swarmclaw-integration-plugin';

const swarmclaw = await createPlugin();

// Use in agent message handler
async function handleUserMessage(message) {
  try {
    const response = await swarmclaw.chat([
      { role: 'user', content: message }
    ]);
    return response.content;
  } catch (error) {
    console.error('All providers failed:', error);
    throw error;
  }
}

With LiteLLM

# litellm_config.yaml
model_list:
  - model_name: "responsible-llm"
    litellm_params:
      model: "openai/gpt-4o"
    fallbacks:
      - anthropic/claude-sonnet-4-20250514
      - gemini/gemini-2.0-flash
      - ollama/llama3.1

Troubleshooting

Common Issues

All providers failing:

• Verify API keys are correct
• Check network connectivity
• Review provider status pages
• Check rate limits

High latency:

• Monitor provider health status
• Consider adjusting failover order
• Review timeout settings

Provider marked unhealthy:

• Check consecutive failure count
• Review health check logs
• Manually mark healthy if needed: plugin.markProviderHealthy('openai')

License

MIT

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Run tests
5. Submit a pull request

Support

• Documentation: SKILL.md
• Issues: https://github.com/heretek-ai/heretek-openclaw/issues
• Heretek OpenClaw: https://github.com/heretek-ai/heretek-openclaw