Bitcoin-MCP

Bitcoin-MCP

give you access to bitcoinsAPI

Category
访问服务器

README

bitcoin-mcp

<!-- mcp-name: io.github.Bortlesboat/bitcoin-mcp -->

PyPI MCP Registry Python 3.10+ License: MIT Tests

MCP server for Bitcoin Core/Knots nodes. 35 tools, 6 prompts, 7 resources for AI agents to query the blockchain, analyze mempool fee markets, decode transactions, broadcast transactions, decode Lightning invoices, inspect mining economics, and explore the protocol.

Runs against your local node — no API keys, no rate limits, no third-party dependencies.

The first Bitcoin MCP server on the official Anthropic MCP Registry.

No Node? No Problem

Point at a hosted Satoshi API instance instead of running your own node:

SATOSHI_API_URL=https://bitcoinsapi.com bitcoin-mcp

This adds a query_remote_api tool that proxies requests through the REST API, with optional L402 Lightning micropayments.

60-Second Quickstart

# 1. Install
pip install bitcoin-mcp

# 2. Add to Claude Desktop (claude_desktop_config.json)
{
  "mcpServers": {
    "bitcoin": { "command": "bitcoin-mcp" }
  }
}

# 3. Ask Claude: "What's the current Bitcoin fee environment?"

That's it. The server auto-detects your node via cookie authentication.

Install

pip install bitcoin-mcp

Configuration

Override with environment variables if needed:

BITCOIN_RPC_HOST=127.0.0.1
BITCOIN_RPC_PORT=8332
BITCOIN_RPC_USER=myuser
BITCOIN_RPC_PASSWORD=mypassword
BITCOIN_DATADIR=E:/
BITCOIN_NETWORK=mainnet          # mainnet (default) | testnet | signet | regtest
SATOSHI_API_URL=https://bitcoinsapi.com  # optional: use remote API instead of local node

35 Tools

Node & Network (3)

Tool Description
get_node_status Chain, height, sync progress, disk, connections, version
get_peer_info Connected peer addresses, latency, services
get_network_info Protocol version, relay fee, warnings

Blockchain & Blocks (6)

Tool Description
analyze_block Full analysis: pool ID, SegWit/Taproot adoption, fee distribution
get_blockchain_info Chain, difficulty, softforks, chain work
get_block_stats Raw statistics: median fee, total output, subsidy
get_chain_tx_stats Transaction rate over N blocks
get_chain_tips Active chain, forks, and stale branches
search_blocks Block stats for a range of heights (max 10)

Mempool (4)

Tool Description
analyze_mempool Fee buckets, congestion level, next-block minimum fee
get_mempool_entry Details of a specific unconfirmed tx
get_mempool_info Quick stats: count, bytes, min relay fee
get_mempool_ancestors Ancestor transactions for CPFP analysis

Transactions (4)

Tool Description
analyze_transaction Full decode + inscription detection + fee analysis
decode_raw_transaction Decode raw hex without input lookup
send_raw_transaction Broadcast a signed transaction (with fee safety limit)
check_utxo Check if a specific output is unspent

Fee Estimation (4)

Tool Description
get_fee_estimates Rates for 1/3/6/25/144 block targets
get_fee_recommendation Plain-English send/wait advice
estimate_smart_fee Custom confirmation target
compare_fee_estimates Side-by-side urgency labels + cost for 140vB tx

Mining (2)

Tool Description
get_mining_info Difficulty, hashrate, block size
analyze_next_block Block template: revenue, fee percentiles, top-fee txs

UTXO Set (2)

Tool Description
get_utxo_set_info Total UTXOs, supply, disk size (slow: ~1-2 min)
get_block_count Current block height (fast)

AI Developer Tools (8)

Tool Description
describe_rpc_command Structured help for any RPC command
list_rpc_commands All commands grouped by category
search_blockchain Smart router: auto-detects txid, block hash/height, or address
explain_script Decode script hex into readable opcodes
get_address_utxos Scan UTXO set for an address
validate_address Validate and classify address type (P2PKH/P2SH/P2WPKH/P2WSH/P2TR)
get_difficulty_adjustment Epoch progress, estimated adjustment percentage
compare_blocks Side-by-side block statistics comparison

Lightning (1)

Tool Description
decode_bolt11_invoice Decode a BOLT11 Lightning invoice (no LN node needed)

6 Agent Workflow Prompts

Pre-built multi-step analysis templates that agents can invoke:

Prompt Description
analyze_fee_environment Fee market analysis with send/wait recommendation
investigate_transaction Deep-dive transaction investigation
monitor_mempool_fees Watch for fee drops below threshold
taproot_adoption_report P2TR adoption trends over recent blocks
network_health_report Comprehensive network health assessment
track_transaction Track a tx from mempool to confirmation

7 Resources

Static data endpoints for AI agents:

  • bitcoin://node/status — node summary
  • bitcoin://fees/current — fee estimates
  • bitcoin://mempool/snapshot — mempool analysis
  • bitcoin://connection/status — connection status + troubleshooting hints
  • bitcoin://protocol/script-opcodes — Bitcoin Script opcodes reference
  • bitcoin://protocol/address-types — address type properties and BIPs
  • bitcoin://protocol/sighash-types — signature hash types

Example Configs

See examples/ for ready-to-use config snippets:

What Makes This Different

Feature bitcoin-mcp Competitors
Official MCP Registry Yes No
Tools / Prompts / Resources 35 / 6 / 7 Fewer
Data source Your local node Third-party APIs
No-node fallback Satoshi API remote N/A
Mempool analysis Fee bucketing, congestion, CPFP Basic stats only
Inscription detection Yes No
Pool identification Yes No
SegWit/Taproot metrics Yes No
Next-block prediction Yes No
Agent workflow prompts 6 built-in None
Rate limits None API-dependent
Typed responses Pydantic models Raw JSON

CLI

bitcoin-mcp              # Start MCP server (default)
bitcoin-mcp --version    # Print version
bitcoin-mcp --check      # Test RPC connection and exit

Requirements

  • Python 3.10+
  • Bitcoin Core or Bitcoin Knots with server=1 in bitcoin.conf
  • txindex=1 recommended for transaction lookups

Related

License

MIT

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选