- Home
- Skills
- Trading & Investing
- polyweather agent
polyweather agent
A Python agent for simulated weather market trading using real-time forecasts, LLM analysis, and Kelly sizing.
Free
Works with the AI Agents you already use
polyweather agent
Example session with this skill installed
Run a single trading cycle to analyze current weather markets and log the results to the paper ledger.
- Read your context and instructions
- Compiled the polyweather agent
One trading cycle completed. Fetched weather data for 7 cities, scanned Polymarket Gamma API, and updated the local JSON ledger with simulated positions.
Simulation Report: Cycle #0412
| Market | Outcome | Forecast Prob | Market Price | Edge | Action | Kelly Size |
|---|---|---|---|---|---|---|
| NYC Temp > 90°F (July 20) | YES | 72% | 0.61 | 11% | BUY_YES | $45.00 |
| London Rain > 5mm (July 21) | NO | 85% | 0.78 | 7% | BUY_NO | $28.00 |
| Dubai Max Temp < 110°F | YES | 15% | 0.22 | -7% | SKIP | $0.00 |
| Tokyo Humidity > 80% | YES | 55% | 0.52 | 3% | SKIP | $0.00 |
Cycle Statistics
- Simulated Bankroll: $1,073.00
- Active Paper Positions: 2
- Daily P&L: +$12.50 (Simulated)
- Dashboard Status: Active at http://localhost:5000
Next steps
- Open
results/paper_trades.jsonto view the full trade history. - Run
python main.py --ui-onlyto visualize performance trends in the browser. - Check
logs/for the detailed LLM reasoning behind the NYC and London recommendations.
Connects securely to your tools. The creator never sees your data.
About this skill
Testing weather market strategies often requires manual data collection and complex probability modeling. This skill automates the entire pipeline, from fetching real-time meteorological data to simulating fractional Kelly-sized trades on Polymarket.
What it does
- Forecast aggregation pulls data from Open-Meteo and Apify for major global cities including New York, London, and Tokyo.
- Market discovery scans the Polymarket Gamma API for active weather-related prediction markets.
- Probability modeling estimates outcome likelihoods by comparing raw forecast data against current market odds.
- LLM analysis uses OpenRouter models to evaluate market conditions and provide confidence-weighted trade recommendations.
- Position sizing applies a quarter-Kelly criterion with a 5% edge threshold to manage a simulated bankroll.
- Simulated hedging runs a secondary agent to identify and offset risk in open paper positions.
How it works
- Install dependencies from the
scripts/folder and configure API keys for OpenRouter and Apify in a local.envfile. - Run the
main.pyscript to start a trading cycle or launch the local web dashboard. - Review simulated trade logs and performance metrics in the
results/directory. - Adjust risk parameters like
MAX_BET_FRACTIONorSTARTING_BANKROLLin the configuration files to test different strategies.
Frameworks & tools
The agent is built with Python 3.11+. It integrates with the Polymarket Gamma API, OpenRouter (supporting models like Hermes), Open-Meteo, and Flask for the local dashboard.
Why this beats prompting it yourself
Manual prompts cannot stay synced with live weather APIs or track a persistent ledger of simulated P&L. This skill handles the math for Kelly sizing and data ingestion, ensuring your "paper" bankroll stays consistent across sessions.
Use cases
- Test weather-market trading strategies without risking real capital.
- Compare meteorological forecast accuracy against prediction market probabilities.
- Audit LLM reasoning for financial decision-making in niche markets.
- Build a historical dataset of weather market price movements and outcomes.
Known limitations
This is a paper-trading simulation only and cannot execute real orders or hold private keys. It is limited to specific tracked cities and requires an OpenRouter API key for the LLM analysis.
Paper trading simulation only. Requires Python 3.11+. Limited to specific cities like NY, London, and Tokyo. No real wallet or execution support.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed today
- Free with an Agensi account
Requires
Permissions
Allowed Hosts