Smart Agent Skill

aion-sdk powered Polymarket thesis trader. Scan prediction markets for a configurable thesis, compare your probability to the market price, and trade only wh...

AI Skill Trader

This is a template. The default signal is a user-supplied thesis probability for markets matching a query string. Remix it with your own model, news feed, research pipeline, or alternative market-selection logic. The skill handles market discovery, context checks, execution tags, dry-run safety, and operator summaries.

What It Does

This skill searches Polymarket for markets matching a thesis query such as bitcoin, fed, or election, then compares your fair probability to the current market price. If the gap exceeds the configured minimum edge, it proposes or executes a trade.

Default behavior:

• Read candidate markets from AION market discovery.
• Compute edge as fair_value - yes_price for YES trades or (1 - fair_value) - no_price for NO trades.
• Check market context before acting.
• Skip trades when warnings or flip-flop risk are present.
• Default to dry-run unless --live is passed.
• Tag every trade with source and skill_slug.

Strategy Inputs

The entrypoint accepts CLI flags or environment variables:

• --query or MARKET_QUERY: Search term for market discovery.
• --thesis-probability or THESIS_PROBABILITY: Your fair probability from 0 to 1.
• --min-edge or MIN_EDGE: Minimum required edge before trading.
• --trade-amount or TRADE_AMOUNT_USD: Position size per trade.
• --limit or MARKET_LIMIT: Max markets to scan.
• --live or RUN_LIVE=true: Enable real trading.

Example Usage

Dry-run:

python trade_thesis.py --query bitcoin --thesis-probability 0.62

Live trading:

python trade_thesis.py --query bitcoin --thesis-probability 0.62 --min-edge 0.08 --trade-amount 5 --live

Environment-driven run:

export AION_API_KEY=your_api_key
export MARKET_QUERY=bitcoin
export THESIS_PROBABILITY=0.62
python trade_thesis.py

Required Credentials

• AION_API_KEY: Required for market reads and trade execution.

Remix Ideas

• Replace the thesis probability input with a custom forecast model.
• Use get_briefing() opportunity markets instead of query search.
• Add venue-specific sizing rules.
• Add position caps, PnL-aware de-risking, or stale-order cancellation.
• Schedule multiple automaton runs for different themes.

Publishing

From inside this skill folder:

npx clawhub@latest publish . --slug ai-skill-trader --version 1.0.0

Then verify installation:

npx clawhub@latest install ai-skill-trader