OpenClaw Super Tasks ⚡

Curated OpenClaw skills for powerful AI use cases with ClawRouter cost optimization.

Save 90%+ on AI costs while accessing 30+ frontier models through intelligent routing.

Use these skills via Telegram, Discord, WhatsApp, or any OpenClaw channel!

---

🚀 Quick Start (5 Minutes)

Step 1: Install Skills

git clone https://github.com/blockrunai/OpenClaw-Super-Tasks
cd OpenClaw-Super-Tasks

# Copy to OpenClaw workspace
cp -r skills/* ~/.openclaw/workspace/skills/

Step 2: Set Up Telegram (Optional)

Talk to @BotFather on Telegram:

/newbot

Add token to ~/.openclaw/openclaw.json:

{
  "channels": {
    "telegram": {
      "enabled": true,
      "botToken": "YOUR_TOKEN_HERE",
      "dmPolicy": "pairing"
    }
  }
}

Restart and pair:

openclaw daemon restart
# Message your bot, it sends pairing code
openclaw pairing approve telegram <CODE>

Step 3: Use It!

Message your bot:

• "Debug: def foo(): return x/0"
• "Research AI agent trends"
• "Analyze github.com/openclaw/openclaw"

Each response shows real cost and savings via ClawRouter!

---

📚 Available Super Tasks

Skill	What It Does	Example	Expected Cost*
debug-code	Find bugs, explain fixes	"Debug this: [code]"	$0.002-0.02
web-research	Research topics, synthesize findings	"Research X"	$0.01-0.08
github-analyze	Analyze repos, architecture	"Analyze github.com/user/repo"	$0.15-0.40
explain-code	Explain how code works	"Explain: [code]"	$0.001-0.01
code-review	Review code quality	"Review: [code]"	$0.05-0.15

*With ClawRouter routing to optimal models. Compare to Claude Opus: $0.08-5.00 per task

---

💰 Cost Tracking

All skills automatically show ClawRouter cost info:

🐛 Bug Analysis

[Your analysis here]

---
💰 Cost: $0.0023 (saved 97% vs Claude Opus $0.08)
📊 Model: xai/grok-code-fast-1 (MEDIUM tier)
⚡ Tokens: 250 in, 180 out
⏱️ Time: 1.8s

---

🏗️ How It Works

You → Telegram/Discord → OpenClaw → Skill → ClawRouter → Optimal Model
                                              ↓
                                       Logs cost to file
                                              ↓
                                  Response with cost appended

ClawRouter routes to the cheapest capable model across 30+ options:

• SIMPLE tasks → Kimi K2.5 ($0.001/M)
• MEDIUM tasks → Grok Code Fast ($0.20/M)
• COMPLEX tasks → Gemini 2.5 Pro ($10/M)
• REASONING tasks → DeepSeek R1 ($0.50/M)

---

📦 Requirements

• ✅ OpenClaw 2026.2+ (install guide)
• ✅ ClawRouter plugin (install)
• ✅ Messaging account (Telegram, Discord, etc.)

---

🛠️ Adding More Skills

Create ~/.openclaw/workspace/skills/my-skill/SKILL.md:

---
name: my-skill
description: What it does
metadata:
  openclaw:
    requires:
      config: ["providers.blockrun"]
---

# My Custom Skill

## Description
Explain what this skill does.

## When to Use
- User asks X
- User needs Y

## Instructions
Step-by-step what the AI should do...

## Format Response
Show example output format...

## Append Cost
Read from ~/.openclaw/blockrun/logs/usage-YYYY-MM-DD.jsonl

---

📖 Resources

• OpenClaw Documentation
• ClawRouter Repository
• Creating OpenClaw Skills
• Telegram Setup Guide
• OpenClaw Configuration

---

🤝 Contributing

PRs welcome! To add a new skill:

1. Create skills/skill-name/SKILL.md
2. Follow the format in existing skills
3. Test with OpenClaw
4. Submit PR with example output

---

📝 License

MIT License - use freely in personal and commercial projects.

---

🏆 Built By

BlockRun - Pay-per-request AI inference with USDC micropayments

Related Projects:

• ClawRouter - Smart AI model routing
• BlockRun Platform - x402 micropayment infrastructure