v3 BetaBlockfrost-compatible APIs, Analytics Store & MCP

Cardano data,
indexed your way.

Yaci Store is a modular, open-source Java indexer for Cardano. Run the ready-made app with Docker, or embed only the stores you need as Spring Boot starters. Serve the data over REST, Blockfrost-compatible APIs, SQL, Parquet or MCP.

Get started View on GitHub

Mainnet · Preprod · Preview · Devnet PostgreSQL (recommended) · MySQL · H2 MIT licensed

yaci-store · mainnetLIVE

Epoch—

Slot—

Epoch progress

Block

Transaction

UTxO

Assets

Metadata

Script

Staking

Governance

Plugins

Epoch and slot are computed live from mainnet genesis. The event stream is illustrative.

How it works

From chain sync to query, one pipeline

Yaci Store follows the chain from any Cardano node, turns every block into typed events, and fans them out to the stores you enable. Plugins filter what gets stored. The results land in your database and, optionally, in analytics-ready Parquet.

Core indexingPluginsAnalytics (v3)AI / MCP (v3)

Features

Everything you need to build on Cardano data

Start with the stores you need and add more later. Every module is a standalone library and a Spring Boot starter, so the indexer grows with your application.

Modular by design

UTxO, blocks, transactions, assets, metadata, scripts, staking, governance and more. Turn each store on or off, or pull it in as its own Spring Boot starter.

Spring Boot starters

dependencies { + implementation 'com.bloxbean.cardano:yaci-store-utxo-spring-boot-starter:<version>' + implementation 'com.bloxbean.cardano:yaci-store-metadata-spring-boot-starter:<version>' + implementation 'com.bloxbean.cardano:yaci-store-assets-spring-boot-starter:<version>' }

Blockfrost-compatible APIs

Paths and response shapes follow the Blockfrost API, so existing clients can point at your own node.

Blockfrost APIs

GET /blocks/latest

Analytics Store

Export indexed data to Parquet or DuckLake catalogs with ACID transactions and time travel, then query it with a sandboxed DuckDB engine.

Analytics Store

POST /api/v1/analytics/query/sql

SELECT epoch, count(*) AS blocks FROM block WHERE epoch >= 500 GROUP BY epoch ORDER BY epoch

ParquetDuckLakeDuckDBcustom exporters

Ready for AI agents

Built-in MCP tools let an agent discover tables, read schemas and run read-only SQL over your analytics data.

Set up the MCP server

How many blocks were produced per epoch since epoch 500?

analytics-list-tables

analytics-describe-table

analytics-execute-sql

Waiting for tools…

Plugins in your language

Filter, transform or react to data with MVEL, SpEL, JavaScript or Python. No Java code or rebuild needed.

Write a plugin

filters:

metadata.save:

- name: "NFT Metadata Filter"

lang: mvel

expression: label == "721"

Ledger state, computed

The ledger-state profile calculates rewards, AdaPot and governance state itself, with no external data source.

Stores

Staking rewards
AdaPot: treasury & reserves
DRep distribution & proposal status

Operate with confidence

Admin UI for sync status and health, Prometheus and Grafana monitoring, rollback handling, pruning and auto-recovery.

Admin UI

Get started

Run it your way

Go from zero to indexing in minutes with Docker, or build a custom indexer with the same modules.

terminal

# Download the Docker distribution from GitHub Releases, then
# point it at a network in config/application.properties:
store.cardano.host=preprod-node.world.dev.cardano.org
store.cardano.port=30000
store.cardano.protocol-magic=1
 
$ ./yaci-store.sh start
$ ./yaci-store.sh logs:yaci-store