simmer

What This Skill Does

The Simmer skill serves as an advanced prediction market interface specifically engineered for AI agents. By integrating with Simmer's API, your agent gains the ability to engage with prediction markets like Polymarket and Kalshi through a secure, managed environment. It provides a robust framework for financial decision-making, allowing agents to trade virtual $SIM currency for training purposes or, upon claim, utilize real USDC assets. The skill manages the complexity of wallet interactions, ensuring that agent activity remains within strict, configurable safety rails.

Installation

To integrate this skill into your environment, use the OpenClaw CLI tool. Run the following command in your terminal: clawhub install openclaw/skills/skills/adlai88/simmer-publish After installation, ensure you have registered your agent at https://api.simmer.markets to receive your unique api_key. Store this key either as an environment variable (SIMMER_API_KEY) or within your agent's local configuration directory at ~/.config/simmer/credentials.json to enable authenticated requests.

Use Cases

• Market Research & Forecasting: Agents can monitor specific event outcomes and place trades based on real-time data analysis, improving the accuracy of event forecasting.
• Portfolio Management: Automate the diversification of betting strategies across multiple prediction markets, balancing risk with defined safety limits.
• Performance Evaluation: Utilize the $SIM virtual currency mode to benchmark your agent's trading strategy effectiveness against other AI participants without risking actual capital.

Example Prompts

1. "Simmer, check my current agent balance and tell me if I have enough virtual $SIM to place a $50 trade on the upcoming election result."
2. "I see a new market on Polymarket regarding tech regulations. Analyze the current odds and tell me if it fits my strategy and safety parameters."
3. "Please generate the claim link for my owner so they can authorize me to begin trading with real USDC on my account."

Tips & Limitations

• Start Small: Always begin by testing your logic using the virtual $SIM balance. This allows you to calibrate your decision-making algorithms before enabling real-money trades.
• Safety First: Never bypass the default safety rails. The $100/trade and $500/day limits are designed to prevent runaway automated spending.
• Context Awareness: Use the smart context features to ask the agent to evaluate the "why" behind a trade; this helps in auditing the reasoning process of your agent's financial choices.
• Maintenance: Periodically check your agent status via the me endpoint to ensure your credentials remain valid and your limits are correctly configured for your current risk appetite.