Stack signals with cross-source confluence

Single signals are easy to fade. Confluence = same asset showing up across 2+ independent data streams in 24h. confluence_today does the join for you.

Signals it stacks:

insider Form 4 clusters
material 8-K filings
US spot ETF flows
DeFi yield spikes
airdrop activity
Hyperliquid vault concentration
Polymarket whale entries
Solana smart-wallet buys

Tool

confluence_today. params:

param	default	notes
`limit`	5 (free) / 25 (pro)	max items

Returns each asset with signal_count (how many sources hit) and sources (the list).

Agent prompt

Call confluence_today with limit=10. For each item with signal_count >= 3,
output: asset, signal_count, sources joined. Then for the top 3 by
composite_score, pull the actual underlying records from each source
(insider_buy_clusters / sec8k_material_today / etf_flow_today / etc) so
I can read the raw evidence.

Expected output

COIN       | signal_count=4 | insider_cluster, sec8k_2.02, etf_inflow_btc, hl_vault
MSTR       | signal_count=3 | sec8k_2.02, etf_inflow_btc, polymarket_whale
SOL        | signal_count=3 | hl_vault, onchain_smart_wallets, etf_inflow

Top by composite_score: COIN, MSTR, SOL

What to do next

signal_count >= 3 is the threshold worth looking at. 2 is noise.
The interesting moves are when independent sources (insider buys + onchain wallet activity + ETF flow) all point at the same asset on the same day.
This is the differentiator versus single-source data feeds. Don’t sleep on it.

Limits

Free tier returns top 5. Pro 25 default, 50 max. Recompute window: every 4h free, every 15min pro.