Skip to main content

​
Overview

Goldsky provides curated Bitcoin datasets for blocks and transactions, making it easy to build pipelines for Bitcoin blockchain analysis.

​
Quick Start

Get started with Bitcoin data - choose the dataset that fits your use case: For block data: 
sources:
  bitcoin_blocks:
    type: dataset
    dataset_name: bitcoin.raw.blocks
    version: 1.0.0
    start_at: latest
For transaction data: 
sources:
  bitcoin_transactions:
    type: dataset
    dataset_name: bitcoin.raw.transactions
    version: 1.0.0
    start_at: latest
Available datasets:
  • bitcoin.raw.blocks - Block data including hash, height (number), size, weight, merkle root, timestamp, and transaction count
  • bitcoin.raw.transactions - Transaction data including inputs, outputs (as JSON strings), input/output counts and values, fee, and block metadata

​
Schema Reference

bitcoin.raw.blocks - primary key id
FieldTypeDescription
idSTRINGUnique row id (primary key)
hashSTRINGBlock hash
sizeLONGBlock size in bytes
stripped_sizeLONGBlock size without witness data
weightLONGBlock weight
numberLONGBlock height
versionLONGBlock version
merkle_rootSTRINGMerkle root hash
timestampLONGBlock time (Unix seconds)
nonceSTRINGBlock nonce (hex)
bitsSTRINGDifficulty target in compact form
coinbase_paramSTRINGCoinbase parameter
transaction_countLONGNumber of transactions in the block
bitcoin.raw.transactions - primary key id
FieldTypeDescription
idSTRINGUnique row id (primary key)
hashSTRINGTransaction hash (txid)
sizeLONGTransaction size in bytes
virtual_sizeLONGVirtual size (vsize)
versionLONGTransaction version
lock_timeLONGLock time
block_numberLONGBlock height containing this tx
block_hashSTRINGHash of the containing block
block_timestampLONGBlock time (Unix seconds)
is_coinbaseBOOLEANWhether this is a coinbase tx
indexLONGIndex within the block
inputsSTRINGJSON-encoded array of inputs
outputsSTRINGJSON-encoded array of outputs
input_countLONGNumber of inputs
output_countLONGNumber of outputs
input_valueLONGTotal input value in satoshis
output_valueLONGTotal output value in satoshis
feeLONGTransaction fee in satoshis
inputs and outputs are JSON-encoded strings. Use json_extract or your sink’s JSON functions to parse them downstream.

​
Guide: Track Large Bitcoin Transactions

Monitor transactions with significant BTC movement: 
name: bitcoin-whale-tracker
resource_size: s

sources:
  transactions:
    type: dataset
    dataset_name: bitcoin.raw.transactions
    version: 1.0.0
    start_at: latest

transforms:
  # Filter to large transactions (>1 BTC = 100,000,000 satoshis)
  large_transactions:
    type: sql
    primary_key: id
    sql: |
      SELECT
        id,
        hash,
        block_hash,
        block_number,
        block_timestamp,
        input_value,
        output_value,
        fee,
        _gs_op
      FROM transactions
      WHERE output_value > 100000000

sinks:
  postgres_whales:
    type: postgres
    from: large_transactions
    schema: public
    table: bitcoin_whale_transactions
    secret_name: MY_POSTGRES
    primary_key: id

​
Guide: Monitor Block Production

Track Bitcoin block metadata for network analysis: 
name: bitcoin-block-monitor
resource_size: s

sources:
  blocks:
    type: dataset
    dataset_name: bitcoin.raw.blocks
    version: 1.0.0
    start_at: latest

transforms:
  block_stats:
    type: sql
    primary_key: id
    sql: |
      SELECT
        id,
        hash,
        number,
        timestamp,
        transaction_count,
        size,
        stripped_size,
        weight,
        merkle_root,
        _gs_op
      FROM blocks

sinks:
  postgres_blocks:
    type: postgres
    from: block_stats
    schema: public
    table: bitcoin_blocks
    secret_name: MY_POSTGRES
    primary_key: id

​
Guide: Analyze Transaction Fees

Track Bitcoin transaction fees over time: 
name: bitcoin-fee-tracker
resource_size: s

sources:
  transactions:
    type: dataset
    dataset_name: bitcoin.raw.transactions
    version: 1.0.0
    start_at: latest

transforms:
  fee_analysis:
    type: sql
    primary_key: id
    sql: |
      SELECT
        id,
        hash,
        block_number,
        block_timestamp,
        fee,
        size,
        virtual_size,
        CAST(fee AS DOUBLE) / CAST(size AS DOUBLE) as fee_per_byte,
        _gs_op
      FROM transactions
      WHERE fee > 0

sinks:
  postgres_fees:
    type: postgres
    from: fee_analysis
    schema: public
    table: bitcoin_fee_analysis
    secret_name: MY_POSTGRES
    primary_key: id

​
Performance Tips

Use the most specific dataset for your use case:
  • Block-level analysis? Use bitcoin.raw.blocks
  • Transaction analysis? Use bitcoin.raw.transactions
Apply filters in SQL as early as possible to reduce data volume:
transforms:
  filtered:
    type: sql
    sql: SELECT * FROM transactions WHERE output_value > 100000000
Bitcoin has moderate throughput compared to high-volume chains. Start with small or medium:
resource_size: s  # or m for historical backfills