Glossary

Integer that identifies which data endpoint to query. Each Heisenberg endpoint (Markets, Trades, Wallet 360, etc.) has a unique agent_id. Passed in the request body.

Proprietary 0–100 composite score measuring a wallet's prediction market trading skill. Factors in profitability, accuracy, consistency, risk management, and diversification.

Comprehensive wallet intelligence profile with 60+ computed metrics covering performance, risk, behavior, and activity patterns.

Market quality and health assessment — liquidity depth, volume dynamics, whale concentration, and price stability.

Request body field controlling response format. Use {"format_type": "raw"} for structured JSON responses.

Boolean in the pagination response. When true, more results exist beyond the current page. Increment offset by limit to fetch the next page.

Hex string uniquely identifying a market (question) on Polymarket. A market can have multiple outcomes, each with its own token_id. Example: 0xbcacd5a055f5...

Large integer identifying a specific outcome token within a Polymarket market. Each side of a binary market (Yes/No) has its own token_id. Used for trading and price queries.

The Polymarket proxy wallet address associated with a user. This is the address that executes trades on-chain, not the user's EOA.

Profit and Loss — the net financial result of a wallet's trading activity. Can be realized (closed positions) or unrealized (open positions).

Open, High, Low, Close, Volume — standard candlestick data format used in price charts. Each candle represents a fixed time period.

Time bucket size for candlestick data. Options: 1m, 5m, 15m, 1h, 4h, 1d. Smaller resolution = more granular data.

URL-friendly identifier for a market. Example: "will-bitcoin-hit-100k-by-2026". Used for lookups and search queries.

Slug identifying the parent event that groups related markets together. One event (e.g., "2024 US Election") can contain many markets.

A possible result of a prediction market. Binary markets have "Yes"/"No" outcomes. Multi-outcome markets can have many (e.g., candidate names).

Model Context Protocol — an open standard for connecting AI models to external data sources. Heisenberg supports MCP for Claude integration.