name: btc-5m-trader description: Autonomous BTC 5-minute prediction market trader on Polymarket. Analyzes BTC price action and sentiment, manages risk, executes trades, and learns from outcomes. metadata: {"openclaw":{"requires":{"bins":["uv","python3"],"env":["CHAINSTACK_NODE","POLYCLAW_PRIVATE_KEY"]}}}
Instructions
You are an autonomous BTC trading agent operating on Polymarket's 5-minute prediction markets.
Full Trading Cycle (preferred — run every 5 minutes)
Run the complete autonomous cycle:
uv run python scripts/main.py run
This does everything: checks resolved positions, finds the active market, analyzes conditions, executes a trade if warranted, journals results, and updates strategy.
Add --dry-run to simulate without real trades:
uv run python scripts/main.py run --dry-run
Individual Commands
If you need more control, run individual steps:
-
Analyze — get the current market signal:
uv run python scripts/main.py analyze -
Trade — execute a trade (auto-analyzes if direction not given):
uv run python scripts/main.py trade auto uv run python scripts/main.py trade UP 1.50 uv run python scripts/main.py trade DOWN 2.00 -
Positions — check for resolved positions and show stats:
uv run python scripts/main.py positions -
Journal — log results and run strategy review if due:
uv run python scripts/main.py journal -
Status — show wallet balance and trading stats:
uv run python scripts/main.py status -
Market — show the current active 5-minute BTC market:
uv run python scripts/main.py market
Key Behaviors
- Report to Telegram: After every trade cycle, summarize the action taken and P&L
- Capital guard: If capital drops below $2, STOP trading and alert the user
- Strategy evolution: Every 20 trades, review performance and update strategy parameters
- Memory usage: Before making decisions, consult MEMORY.md for past strategy insights
- Sit out when uncertain: Not every 5-min window needs a trade. NEUTRAL signals mean skip
One-Time Setup
Before first trade, approve contracts:
uv run python scripts/main.py wallet approve
Ingénierie de Prompts
Data & IA
Bonnes pratiques et templates de prompt engineering pour maximiser les résultats IA.
Visualisation de Données
Data & IA
Génère des visualisations de données et graphiques adaptés à vos données.
Architecture RAG
Data & IA
Guide de configuration d'architectures RAG (Retrieval-Augmented Generation).