Docs
⌘ K

6 minutes

Using Substreams.dev to Find Substreams Packages

Use Substreams.dev⁠ and take full advantage of its core features to discover ready-to-use Substreams packages for the data you want.

Overview

This guide explains how to use Substreams.dev⁠ (the Substreams Registry) to quickly discover, inspect, and consume Substreams packages. Where Graph Explorer helps you find published Subgraphs, Substreams.dev is the storefront for the Substreams ecosystem: it lets you browse curated data collections, inspect individual packages and their modules, explore what’s available per chain, and learn from real-world projects.

When you visit Graph Explorer, you can also access the link to explore Substreams⁠.

The goal of this page is to help data consumers answer a single question: “Is there already a Substreams package that produces the data I need, and how do I start streaming it?” If you’re ready to build your own package instead, see the Substreams Quick Start.

Prerequisites

  • Basic familiarity with Substreams and the idea that a package (.spkg) is a precompiled binary describing the data to extract from a chain.
  • The Substreams CLI⁠ installed if you plan to inspect or run packages locally (substreams gui, substreams run, substreams info).
  • An API key to stream on-chain data. Substreams is permissionless — you can obtain a key here⁠ without providing personal information.
  • A GitHub account if you plan to log in, create an authentication token, or publish your own package.

Substreams.dev is organized into four primary areas, each answering a different discovery question. The sections below walk through each one in the order you’d typically use them.

AreaDiscovery question it answers
Datasets”What kind of data am I looking for?” (browse by use case)
Packages”Which specific package produces it, and what does it output?”
Chains”What’s available on the network I care about?”
Projects”How have others put this data to use?”

Datasets

Datasets are curated collections of Substreams packages, grouped so you can start from the kind of data you want rather than a package name. (Referred to here by their current site label, “Datasets.”)

The Datasets⁠ area is the best starting point when you know the type of on-chain activity you want but not the exact package. Instead of searching by keyword, you pick a data domain and move into a vetted set of packages that produce it.

Datasets are organized by use case, including categories such as:

  • DEX / AMMs (e.g., swap feeds)
  • Lending
  • Liquid Staking
  • Protocol Staking
  • Perps / Derivatives (e.g., position monitoring)
  • Prediction Markets
  • Governance
  • Oracles
  • Bridges
  • Stablecoins
  • NFTs
  • Core Data

Each dataset card typically surfaces:

  • The category name and whether it is Featured
  • The number of curated packages available in that category
  • The blockchains covered (such as Ethereum, Solana, Polygon, Arbitrum, Optimism, and Base)
  • Typical outputs or use cases (for example, “swap feeds” or “position monitoring”)
  • A status indicator (such as “In progress”) when a collection is still being assembled

How to use it: Select a data domain, review the curated package recommendations inside it, then choose how you want to consume the data — package-level CLI, JavaScript, local execution, or a deployment path through the Substreams x The Graph Market⁠ workflow. This is the fastest route from “I need DEX swaps” to a package that produces them.

Packages

The Packages⁠ area is the core of the registry. A Substreams package is a precompiled .spkg binary that defines exactly what data is extracted and transformed from a chain — conceptually similar to the mapping logic in a traditional Subgraph.

Package listings can be sorted and browsed by:

  • Last uploaded — the most recently published packages
  • Featured — highlighted, recommended packages
  • Most downloaded — the most widely used packages
  • Most used and Alphabetical — additional ordering options when searching

Each package entry generally shows its name, creator, target network, version, published date, and download count. This metadata helps you quickly gauge maturity and adoption before committing to a package.

Click into any package to reach its dedicated page, where you can:

  • View package metadata — name, version, documentation, and target network
  • Inspect the module graph (DAG): every module, its kind (map, store, or blockIndex), output types, and update policies
  • See how modules depend on one another (dependsOn / dependedBy) and the input chain for each module (source blocks, other maps, stores in get/deltas mode, and params)
  • Review the protobuf output types and proto files the package produces, so you know the exact shape of the data
  • Copy ready-to-run commands to stream or deploy the package

Inspecting a package from the CLI. Once you have a package’s .spkg URL, you can inspect and run it directly without downloading anything first:

# See a package's modules, output types, and metadatasubstreams info https://spkg.io/creator/package-v1.0.0.spkg# Explore results interactively in a terminal UIsubstreams gui https://spkg.io/creator/package-v1.0.0.spkg <module_name># Stream output from a specific modulesubstreams run -e mainnet.eth.streamingfast.io:443 https://spkg.io/creator/package-v1.0.0.spkg <module_name> -t +1000

Tip: You can also inspect packages, read their module DAGs, and generate sink commands programmatically with the Substreams Search MCP, which lets AI agents search the registry and analyze packages for you.

Chains

The Chains⁠ area lets you discover data by starting from a network rather than a use case. It’s the right entry point when your question is “what Substreams data can I get on this chain?”

The Chains view organizes blockchain families and surfaces, for each network:

  • The available endpoints used to stream data (for example, “3 networks / 12 endpoints” for Ethereum)
  • The block model for the chain (how blocks and their contents are represented)
  • The packages available for that network

Networks span several ecosystems, including Ethereum, Solana, BNB Chain, Arbitrum, Polygon, and Near, with additional chains supported across the broader Substreams ecosystem. Selecting a chain filters discovery down to what’s available there, so you can move straight from a network choice to a compatible package and its streaming endpoint.

Projects

The Projects⁠ area showcases real-world implementations — practical examples of how teams have used Substreams packages to power applications, pipelines, and data products.

Use Projects to:

  • See end-to-end examples of packages in production, from discovery through to a live data sink
  • Understand common patterns for wiring packages into applications and databases
  • Get ideas for how to combine packages in a more composable way

Projects are especially useful once you’ve found candidate packages in Datasets or Packages and want to see how similar data has been consumed and sinked in practice.

Consuming Package Data

After you’ve found a package that fits your needs, you choose how to consume the data. Substreams supports multiple sinks:

  • Subgraph — configure an API to meet your data needs and host it on The Graph Network.
  • SQL Database⁠ — send the data to a Postgres or Clickhouse database.
  • Direct Streaming⁠ — stream data directly into your application.
  • PubSub⁠ — publish data to a PubSub topic.

Many package pages will help you identify the right sink. If a package has an embedded sink configuration, it can produce ready-to-run install, setup, and run commands with the correct network endpoint; if not, you can still check whether its modules output sink-compatible types (such as DatabaseChanges) and wire up a sink yourself.

Publishing Your Own Package

If you can’t find a package that produces the data you need, you can build and publish one. Substreams packages are written in Rust, and once compiled to an .spkg you can share them with the community through the registry. See Publishing a Substreams Package for the full flow, and the Substreams Quick Start to start building.

Additional Resources