Example prompts

Use Centaur list_feed to show the latest source-message groups. For each group, name the trader, quote the source text briefly, and list its events with asset, direction, and event type.

Use Centaur list_events to find trade events from the last 24 hours, and ignore events whose `autoGenerated` is true. Summarize the most notable trades and quote the source message of each with list_messages.

Use Centaur list_aggregate_summaries to summarize the main market themes of the last 24 hours. Use explicit UTC bounds and report each summary's `coverage` counts.

Use Centaur list_asset_stats for BTC, ETH, and SOL to compare current long and short positioning, with counts and percentages.

Use Centaur list_open_positions for BTC to list current open positions with trader, direction, entry price, and current return.

Use Centaur get_trending_signals with a 7-day window to show which assets gained the most net long or short traders. Explain each row in one line.

Use Centaur get_smart_money to show where leading traders and the rest of the crowd are positioned on opposite sides. Rank the assets by the gap.

Use Centaur get_leaders_book to list the assets where at least two traders on its roster hold the same side. Resolve their names with list_trader_directory.

Use Centaur rank_traders to rank the most active traders of the last 7 days by event count. Include their position counts and direction bias.

Use Centaur rank_traders to rank traders by 7D win rate for positions opened in the last 30 days. Flag thin samples using the returned position counts.

Use Centaur summarize_message_activity to show daily message and event volume for the last 7 days, grouped by trader.

Use Centaur list_trader_directory to resolve Crypto Capo, then summarize their profile with list_trader_stats, list_open_positions, and list_messages: asset focus, direction bias, time-based performance, open positions, and recent source messages.

Use Centaur list_trader_stats to compare a trader's 1D, 7D, and 30D time-based performance. Report scored and sampled positions for each window, and flag thin samples.

Use Centaur list_asset_stats to get ETH's current positioning, then compare it with list_aggregate_summaries from the same UTC day. Keep positioning and narrative claims separate.