SkillAgentSearch skills...

Geo Satellite Trader

trade polymarket geopolitics using NASA satellite data + Claude AI. 10 free APIs, $0/month.

Install / Use

npx skills add helicerat/geo-satellite-trader

Installs into whichever agent you are using.

About this skill

Quality Score

0/100

Supported Platforms

Claude Code
Claude Desktop

README

🛰 Geopolitical Intelligence Terminal

Trade prediction markets using satellite data, aircraft tracking, and AI analysis.

$500 → $7,778 in 16 trades (backtested on real NASA FIRMS data)

How It Works

NASA FIRMS (fires) ──┐
OpenSky (aircraft) ──┤
USGS (earthquakes) ──┤
VIX + Gold + Crypto ─┼──→ Claude AI ──→ Trade Signal ──→ Polymarket
Fear & Greed ────────┤                                      ↓
Exchange Rates ──────┤                               Telegram Alert
Weather ─────────────┘

Quick Start

1. Install dependencies

pip install -r requirements.txt

2. Get API keys (all free)

  • NASA FIRMS → https://firms.modaps.eosdis.nasa.gov → Register → Get MAP_KEY
  • Anthropic Claude → https://console.anthropic.com → Get API key
  • Polymarket → https://polymarket.com → Settings → API Keys
  • Telegram (optional) → @BotFather → Create bot

3. Configure

cp .env.example .env
# Edit .env with your keys

Or set environment variables:

# Windows
set FIRMS_API_KEY=your_key
set ANTHROPIC_API_KEY=sk-ant-...

# Mac/Linux
export FIRMS_API_KEY=your_key
export ANTHROPIC_API_KEY=sk-ant-...

4. Run

Dashboard only (no API keys needed except FIRMS):

python geo_terminal.py

Full pipeline (needs Claude + Polymarket keys):

python geo_run.py

Backtest (no keys needed):

python geo_backtest_v3.py

Files

| File | What it does | |------|-------------| | geo_terminal.py | Dashboard — pulls data from 10 free APIs | | geo_claude_prompt.py | Sends data to Claude → trade signal | | geo_trader.py | Executes trade on Polymarket (dry run default) | | geo_alerts.py | Sends signal to Telegram | | geo_run.py | Runs full pipeline in sequence | | geo_backtest_v3.py | Backtest on historical NASA data | | .env.example | Template for API keys |

Trading Logic

Signal Detection

  • Scans 6 geopolitical regions every 15 minutes
  • Calculates 14-day rolling baseline for fire activity
  • Flags anomalies: current fires > 3x baseline

Position Sizing (Dynamic Kelly)

| Anomaly Ratio | FRP | Kelly Fraction | |---------------|-----|----------------| | ≥10x | any | 20-40% | | 5-10x | ≥50MW | 22% | | 5-10x | <50MW | 15% | | 3-5x | ≥20MW | 12% | | 3-5x | <20MW | 10% | | <3x | any | 2-6% |

Exit Rules

  • Take profit: +50%
  • Stop loss: -30%
  • Time stop: 14 days max hold

False Alarm Filtering

Single-source anomaly (only FIRMS, no VIX/Gold/Aircraft confirmation) → Kelly capped at 3%

Backtest Results (Real NASA Data)

$500 → $7,778  (+1,456%)
Win rate: 69% (11 wins, 5 losses)
Profit factor: 17.6x
Max drawdown: 7.3%
Avg hold: 5.3 days

Tested on 22 blind FIRMS anomaly signals across Israel/Gaza, Kursk, and Syria (Sep 2023 — Dec 2024).

⚠️ Disclaimer

  • Backtest entry/exit prices are estimated (Polymarket has no public price history API)
  • FIRMS data has 4-6h delay from satellite observation to API
  • Small sample size (7 events) — not statistically significant
  • Past performance ≠ future results
  • This is educational content, not financial advice
  • Always start with small amounts and dry-run mode

Related Skills

View on GitHub
GitHub Stars36
CategoryDevelopment
Updated9d ago
Forks10

Languages

Python

Security Score

75/100

Audited on Jul 29, 2026

No findings