CoinBucha
Provides leading, non-price Bitcoin intelligence signals such as sovereign adoption, hiring velocity, network hashrate, ETF flows, and corporate treasuries.
CoinBucha MCP — Bitcoin signal intelligence for AI agents
Signal-first Bitcoin intelligence over MCP. Leading, non-price indicators — sovereign adoption, hiring velocity, network hashrate, ETF flows, corporate treasuries — each as a 0–100 strength, a direction, a one-line
why, and primary sources.Information, not financial advice. No tool returns a buy/sell/hold recommendation.
This is the public MCP interface for CoinBucha. The hosted server reads the same live signal data the dashboard renders, so the machine face and the human face never diverge.
Connect (hosted — recommended, no install)
The server is remote and speaks Streamable HTTP. Point any MCP client at:
https://coinbucha.com/api/mcpClaude Desktop / Claude Code — add a custom connector with that URL, or in
claude_desktop_config.json:
{
"mcpServers": {
"coinbucha": {
"type": "streamable-http",
"url": "https://coinbucha.com/api/mcp"
}
}
}Cursor / other clients — add an MCP server with transport streamable-http and the URL above.
Free tier is open (IP rate-limited). Higher limits are available with an API key via the
Authorization: Bearer <key> header — see coinbucha.com.
Related MCP server: PreReason-mcp
Connect (local stdio — offline/dev)
The bundled server.py exposes the same tools and fetches the live public snapshots from
coinbucha.com, so it needs no local data pipeline.
pip install -r requirements.txt
python3 server.py # stdio transport
# inspect: uv run mcp dev server.py{
"mcpServers": {
"coinbucha": { "command": "python3", "args": ["server.py"] }
}
}Tools
Tool | Returns |
| Ranked composite signal feed (filter by |
| Bitcoin/crypto-infra hiring velocity: open roles + 30/90-day deltas from live ATS boards |
| Sovereign Bitcoin holdings, ranked (filter by |
| Corporate Bitcoin treasury holdings: total BTC held by public companies + top holders |
| Spot Bitcoin ETF net flows (latest day + trailing 5/30-day) |
| Bitcoin network hashrate level + 30/90-day trend |
| Today's machine-readable signal digest (the CoinBucha Daily Brew) |
Resource coinbucha://methodology — how every signal is computed (see docs/methodology.md).
Prompt morning_bitcoin_brief — one-shot template to render the Daily Brew from live signals.
Every response carries an as_of timestamp, a why, and a disclaimer.
Example
Call scan_signals with { "min_strength": 60, "direction": "tailwind" } → the strongest
structural tailwinds right now, ranked, each with its rationale and sources. Or run the
morning_bitcoin_brief prompt for a written digest.
Methodology & guardrail
Signals are leading, non-price indicators rendered as signal_strength (0–100), a
direction, and a why. The full, versioned methodology is public — see
docs/methodology.md — because an auditable method is the point.
CoinBucha's public and MCP surfaces are self-audited against MiCA / FCA / GDPR. No surface presents a signal as advice.
Links
Dashboard: https://coinbucha.com
MCP endpoint: https://coinbucha.com/api/mcp
Discovery manifest: https://coinbucha.com/.well-known/mcp.json
License
MIT — see LICENSE. This repository is the public MCP interface and
methodology; the signal-collection pipeline is maintained separately.
Available Tools
7 toolsget_daily_brewBInspect
Today's machine-readable signal digest (the CoinBucha Daily Brew).
Inputs: no named parameters.
get_etf_flowsAInspect
Spot Bitcoin ETF net flows (latest day + trailing 5/30-day) — institutional demand.
Inputs: no named parameters.
get_hiring_signalAInspect
Bitcoin-native hiring velocity (open roles + 30/90-day deltas).
Args:
company: optional company name filter (case-insensitive substring).
Inputs: company.
get_network_signalAInspect
Bitcoin network hashrate level and 30/90-day trend (miner conviction / security).
Inputs: no named parameters.
get_sovereign_reservesBInspect
Sovereign Bitcoin holdings, ranked.
Args:
country: optional country name/code filter (case-insensitive substring).
tier: optional tier filter (1 = largest holders).
Inputs: tier, country.
get_treasury_holdingsBInspect
Corporate Bitcoin treasury holdings: total BTC held by public companies + top holders.
Inputs: no named parameters.
scan_signalsAInspect
Ranked composite feed of Bitcoin signals.
Args:
min_strength: only return signals with signal_strength >= this (0-100).
direction: filter by 'tailwind', 'headwind', or 'neutral'.
signal_types: filter by type, e.g. ['sovereign_adoption','hiring_velocity'].
Inputs: direction, min_strength, signal_types.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 7 tools
Each tool targets a distinct data category (hiring, sovereign reserves, treasury, ETF flows, network, digest, composite feed). While get_daily_brew and scan_signals both surface signals, their descriptions clearly differentiate a digest from a filterable ranked feed.
Six of seven tools follow the 'get_<noun>' pattern (e.g., get_etf_flows, get_network_signal). The outlier 'scan_signals' uses a different verb style, causing a minor deviation in an otherwise consistent scheme.
Seven tools is a well-scoped set for a Bitcoin signals server, covering distinct signal categories and aggregate views without bloat.
The surface covers key Bitcoin signal categories (institutional, corporate, sovereign, network, hiring) plus digest and composite feeds. A slight gap might be a direct price or on-chain metric tool, but the server appears purpose-focused and effective.
Maintenance
Related MCP Connectors
Narrative & signal intelligence for AI agents: crypto/AI/macro convergence & divergence.
Signed BTC and macro market intelligence for autonomous research, risk, and trading agents.
Verifiable crypto/KR/US market intelligence for AI agents: signals, hash-chained ledger. Not advice.
Whale & Institutional Flow MCP — 8 tools: TVL flows, alpha signals, stablecoin supply.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceAI-powered crypto signal intelligence for 20 assets (BTC, ETH, SOL, etc). 6 scoring dimensions: whale activity, technical analysis, derivatives flow, narrative strength, sentiment, market structure. Market regime detection (TRENDING/RANGING), portfolio optimization, and accuracy tracking. 9 read-only MCP tools. Free via MCP, $0.001 USDC via x402 on Base for REST API.3MIT
- AlicenseAqualityAmaintenancePreReason is a financial data MCP server and REST API for AI agents. Ask what a listed company reported, where Treasury yields stand or what the Federal Reserve decided, and one call returns a briefing reasoned in advance, across asset classes and regions, with every figure dated and every source named. Agents start at https://www.prereason.com/agents.md8148 npm8MIT
- AlicenseAqualityCmaintenanceThe conviction engine for autonomous crypto trading agents. 376 metrics across 8 factor classes, multi-factor backtesting, signal persistence, and regime analysis — 21 tools for AI agents via MCP.2147 npmMIT
- AlicenseAqualityDmaintenanceBitcoin market intelligence MCP server. Exposes Signal Lord's composite gauge scoring, on-chain and macro regime signals, and market data to AI agents.2610 npm1MIT