Backtests of crypto trading signals on 94 coins: what price did next, and how likely it was luck.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Lists the coin-and-signal studies with the strongest historical evidence, best first. Filter by coin, by signal, by lean (bullish or bearish) and by minimum proof score (default 55). Answers questions such as "which crypto signals actually worked" or "what has worked on <coin>". Statistics about past price behaviour on Binance daily candles. Not financial advice and not a prediction.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http provyan https://provyan.com/mcp
Lists the coin-and-signal studies with the strongest historical evidence, best first. Filter by coin, by signal, by lean (bullish or bearish) and by minimum proof score (default 55). Answers questions such as "which crypto signals actually worked" or "what has worked on <coin>". Statistics about past price behaviour on Binance daily candles. Not financial advice and not a prediction.
| Argument | Type | Description |
|---|---|---|
| coin | string | |
| lean | string | |
| limit | number | |
| signal | string | |
| min_score | number |
{
"type": "object",
"properties": {
"coin": {
"type": "string"
},
"lean": {
"enum": [
"bullish",
"bearish"
],
"type": "string"
},
"limit": {
"type": "number"
},
"signal": {
"type": "string"
},
"min_score": {
"type": "number"
}
}
}What historically happened to a crypto coin's price after a specific signal (e.g. RSI oversold, a 20% weekly drop, a breakout), measured on every past occurrence and compared with the coin's ordinary days. Returns case count, win rate, median move before and after trading costs, typical dip on the way, the luck probability after multiple-testing correction, and a 0-100 proof score. Statistics about past price behaviour on Binance daily candles. Not financial advice and not a prediction.
| Argument | Type | Description |
|---|---|---|
| coin* | string | Ticker or slug, e.g. BTC or bitcoin |
| signal* | string | Signal id, e.g. rsi-oversold, weekly-drop-20, golden-cross |
{
"type": "object",
"required": [
"coin",
"signal"
],
"properties": {
"coin": {
"type": "string",
"description": "Ticker or slug, e.g. BTC or bitcoin"
},
"signal": {
"type": "string",
"description": "Signal id, e.g. rsi-oversold, weekly-drop-20, golden-cross"
}
}
}Signals recently detected and publicly revealed in Provyan's hash-chained ledger, with the call made at the time and the graded result once seven days have passed, plus the live hit record. Statistics about past price behaviour on Binance daily candles. Not financial advice and not a prediction.
No arguments.
{
"type": "object",
"properties": {}
}The coins and signal ids Provyan studies, with each signal's exact rule.
No arguments.
{
"type": "object",
"properties": {}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 15, 2026.