Ask your AI agent about the Bitcoin power law
btclaw.io runs a free MCP server. Add it to Claude or any agent that speaks the Model Context Protocol, then ask in plain words: where is Bitcoin against the trend today, what is the trend in 2030, when does the trend reach $1M. The answers come from the same daily refit as this site.
https://btclaw.io/mcpRemote server (streamable HTTP). Free, read-only, no account and no key.
Add it in one step
Claude (web and desktop)
Open Settings → Connectors, choose Add custom connector, name it btclaw and paste the server URL.
Claude Code
Run once in a terminal:
claude mcp add --transport http btclaw https://btclaw.io/mcp
Other MCP clients
Any client that supports remote servers. The usual config:
{
"mcpServers": {
"btclaw": { "url": "https://btclaw.io/mcp" }
}
}
What the agent can ask
get_power_law_status
Live price, trend value, % from the trend, zone, both bands, the current fit and the next halving.
“Where is Bitcoin against the power law today?”
get_trend_on_date
The trend and the two bands on any date from 2010 to 2100.
“What is the power-law trend on 1 July 2030?”
when_will_bitcoin_reach
When the trend reaches a price, the earliest date on the upper band, and when even the lower band stays above it.
“When does Bitcoin reach $1 million on the power law?”
explain_power_law_model
The formula, how it is fitted, the bands, its limits, and the peer-reviewed paper behind it.
“How does the Bitcoin power law work, and how reliable is it?”
Good to know
- The trend is refit every day at 03:00 UTC on the full Coin Metrics history, the same fit as the dashboard, the daily card and the TradingView indicator.
- The bands are 0.4× (lower, −60%) and 3× (upper) the trend.
- Every answer carries its source and a short note: an educational model, not a forecast or investment advice.
- Requests are rate-limited per IP to keep the service fast for everyone.
The model and its limits are explained in What is the Bitcoin power law?, and the paper is in the library. A summary for language models is at /llms.txt.