simmer-ai-divergence

What This Skill Does

The simmer-ai-divergence skill serves as a high-performance analytical bridge between proprietary AI price modeling and live decentralized betting markets like Polymarket. By calculating the mathematical spread between the Simmer AI’s internal probability assessments and the crowd-sourced Polymarket odds, this agent identifies statistical anomalies. Users can quickly isolate where the "market" might be mispricing an event, effectively highlighting potential alpha opportunities based on the AI's conviction level. The tool normalizes complex data feeds into actionable insights, allowing traders to monitor market sentiment versus AI model predictions in real-time.

Installation

To integrate this skill into your OpenClaw agent, execute the following command in your terminal: clawhub install openclaw/skills/skills/adlai88/simmer-ai-divergence

Ensure that you have your SIMMER_API_KEY configured in your environment variables. You can obtain this key from the dashboard at simmer.markets/dashboard under the SDK tab. For standard setups, verify that your environment variables include SIMMER_API_KEY for authentication and optionally SIMMER_API_URL if you are utilizing custom endpoints.

Use Cases

• Alpha Generation: Quickly scan for high-divergence markets where the crowd might be lagging behind AI-driven intelligence.
• Portfolio Monitoring: Check the status of current positions to see if market prices are swinging toward or away from the AI's prediction.
• Market Sentiment Analysis: Determine where the AI is consistently more bullish or bearish than the general public, providing a pulse on unconventional market sentiment.
• Risk Management: Identify highly volatile markets where massive divergence may indicate either a massive trading opportunity or a significant disconnect requiring deeper research before entry.

Example Prompts

1. "Simmer, identify all markets where your probability is at least 15% higher than the current Polymarket price."
2. "Show me the top three bearish opportunities where the market is significantly overestimating the likelihood of a 'Yes' outcome."
3. "Is there any major disagreement between your AI models and Polymarket for the upcoming election markets?"

Tips & Limitations

• Threshold Tuning: Don't rely solely on the default 5% threshold. For high-conviction scalping, use --min 15 to filter out market noise.
• Resolution Timelines: Always cross-reference the Resolves date in the output. A high divergence on a long-term market carries different risk profiles than an event resolving in 24 hours.
• API Availability: The accuracy of this skill is strictly tied to the latency of the Simmer API. Ensure your connection is stable to receive the most current probability updates.