ClaudeCode-Custom-Skills/skills/claw-setup/README.md

<div align="center">

# 🦞 Claw Setup

### Professional AI Agent Deployment Made Simple

**End-to-end setup of Claw platforms with 25+ AI providers, security hardening, and personal customization**

---

<p align="center">
  <a href="https://z.ai/subscribe?ic=R0K78RJKNW">
    <img src="https://img.shields.io/badge/Designed%20by-GLM%205%20Advanced%20Coding%20Model-blue?style=for-the-badge" alt="Designed by GLM 5">
  </a>
</p>

<p align="center">
  <i>✨ Autonomously developed by <a href="https://z.ai/subscribe?ic=R0K78RJKNW"><strong>GLM 5 Advanced Coding Model</strong></a></i>
</p>

<p align="center">
  <b>⚠️ Disclaimer: Test in a test environment prior to using on any live system</b>
</p>

---

</div>

## Overview

Claw Setup handles complete deployment of AI Agent platforms with **25+ AI provider integrations** (OpenCode compatible).

```
┌─────────────────────────────────────────────────────────────────┐
│                    CLAW SETUP WORKFLOW                          │
├─────────────────────────────────────────────────────────────────┤
│                                                                 │
│  Phase 1          Phase 2          Phase 3          Phase 4    │
│  ────────         ────────         ────────         ────────   │
│                                                                 │
│  ┌─────────┐     ┌─────────┐     ┌─────────┐     ┌─────────┐  │
│  │ SELECT  │────►│ INSTALL │────►│CUSTOMIZE│────►│ DEPLOY  │  │
│  │ Platform│     │& Secure │     │Providers│     │ & Run   │  │
│  └─────────┘     └─────────┘     └─────────┘     └─────────┘  │
│                                                                 │
└─────────────────────────────────────────────────────────────────┘
```

## Platforms Supported

| Platform | Language | Memory | Startup | Best For |
|----------|----------|--------|---------|----------|
| **OpenClaw** | TypeScript | >1GB | ~500s | Full-featured, 1700+ plugins |
| **NanoBot** | Python | ~100MB | ~30s | Research, customization |
| **PicoClaw** | Go | <10MB | ~1s | Embedded, $10 hardware |
| **ZeroClaw** | Rust | <5MB | <10ms | Maximum performance |
| **NanoClaw** | TypeScript | ~50MB | ~5s | WhatsApp integration |

## AI Providers (25+ Supported)

### Tier 1: Major AI Labs

| Provider | Models | Features |
|----------|--------|----------|
| **Anthropic** | Claude 3.5/4/Opus | Extended thinking, PDF support |
| **OpenAI** | GPT-4o, o1, o3, GPT-5 | Function calling, structured output |
| **Google AI** | Gemini 2.5, Gemini 3 Pro | Multimodal, long context |
| **xAI** | Grok | Real-time data integration |
| **Mistral** | Mistral Large, Codestral | Code-focused models |

### Tier 2: Cloud Platforms

| Provider | Models | Features |
|----------|--------|----------|
| **Azure OpenAI** | GPT-5, GPT-4o Enterprise | Azure integration |
| **Google Vertex** | Claude, Gemini on GCP | Anthropic on Google |
| **Amazon Bedrock** | Nova, Claude, Llama 3 | AWS regional prefixes |

### Tier 3: Aggregators & Gateways

| Provider | Models | Features |
|----------|--------|----------|
| **OpenRouter** | 100+ models | Multi-provider gateway |
| **Vercel AI** | Multi-provider | Edge hosting, rate limiting |
| **Together AI** | Open source | Fine-tuning, hosting |
| **DeepInfra** | Open source | Cost-effective |

### Tier 4: Fast Inference

| Provider | Speed | Models |
|----------|-------|--------|
| **Groq** | Ultra-fast | Llama 3, Mixtral |
| **Cerebras** | Fastest | Llama 3 variants |

### Tier 5: Specialized

| Provider | Use Case |
|----------|----------|
| **Perplexity** | Web search integration |
| **Cohere** | Enterprise RAG |
| **GitLab Duo** | CI/CD integration |
| **GitHub Copilot** | IDE integration |
| **Cloudflare AI** | Gateway, rate limiting |
| **SAP AI Core** | SAP enterprise |

### Local/Self-Hosted

| Provider | Use Case |
|----------|----------|
| **Ollama** | Local model hosting |
| **LM Studio** | GUI local models |
| **vLLM** | High-performance serving |

## Model Selection

**Option A: Fetch from Provider**
```bash
# Fetch available models
curl -s https://openrouter.ai/api/v1/models -H "Authorization: Bearer $KEY" | jq '.data[].id'
curl -s https://api.groq.com/openai/v1/models -H "Authorization: Bearer $KEY"
curl -s http://localhost:11434/api/tags  # Ollama
```

**Option B: Custom Model Input**
```json
{
  "provider": "openai",
  "modelId": "ft:gpt-4o:org:custom:suffix",
  "displayName": "My Fine-Tuned Model"
}
```

## Quick Start

```
"Setup OpenClaw with Anthropic and OpenAI providers"
"Install NanoBot with all available providers"
"Deploy ZeroClaw with Groq for fast inference"
"Configure Claw with local Ollama models"
```

## Configuration Example

```json
{
  "providers": {
    "anthropic": { "apiKey": "${ANTHROPIC_API_KEY}" },
    "openai": { "apiKey": "${OPENAI_API_KEY}" },
    "google": { "apiKey": "${GOOGLE_API_KEY}" },
    "openrouter": { "apiKey": "${OPENROUTER_API_KEY}" },
    "groq": { "apiKey": "${GROQ_API_KEY}" },
    "ollama": { "baseURL": "http://localhost:11434" }
  },
  "agents": {
    "defaults": { "model": "anthropic/claude-sonnet-4-5" },
    "fast": { "model": "groq/llama-3.3-70b-versatile" },
    "local": { "model": "ollama/llama3.2:70b" }
  }
}
```

## Security

- API keys via environment variables
- Restricted config permissions (chmod 600)
- Systemd hardening (NoNewPrivileges, PrivateTmp)
- Network binding to localhost

---

<p align="center">
  <a href="https://z.ai/subscribe?ic=R0K78RJKNW">Learn more about GLM 5 Advanced Coding Model</a>
</p>