Goldsky

Polymarket data in your own database

Goldsky streams Polymarket trades, positions and profit and loss into your database, with the full history included. Trading desks and analysts query it like any other table.

Talk to our team

Read the guide

At a glance

Order fills with price, size and fee
Positions and realized P&L by wallet
Full history in the same tables
Webhooks for the wallets you follow

Trusted by trading firms and blockchain teams

Skip the Polymarket indexer

Polymarket runs on the Polygon blockchain. Goldsky decodes its contracts into tables of trades and positions, so your team starts from the data.

Nothing to decode
Your engineers connect to Goldsky’s Polymarket datasets. Nobody has to write and maintain code that reads Polymarket’s contracts.
Analysis in SQL
Fills and positions are written to Postgres or ClickHouse, where analysts can work out volume, open interest and order flow.
One table for everyone
Trading, research and risk query the same trade table, with live and historical rows together.

Polymarket datasets: fills, orders and positions

Four datasets cover trading and positions. Goldsky streams them wherever you need them.

Order fills

A row each time an order is partly or fully filled.

Matched orders

A row per matched taker order, for a simpler view of each trade.

Balances and positions

The outcome tokens each wallet holds, with average entry price and realized profit and loss.

History included

Pipelines stream new activity and include past data, so backtests and live dashboards read the same tables.

Webhooks

Filter to the markets or wallets you follow and send each match to an endpoint your systems listen on.

Destinations

Postgres, ClickHouse, Kafka, object storage or a webhook.

Building a Polymarket indexer vs. using Goldsky datasets

Table 1. Building a prediction market indexer in-house vs. using maintained Goldsky datasets.
Concern	Building it yourself	With Goldsky
Venue contract upgrades	A rewrite of your decoders, and a gap in your history, each time	Goldsky publishes datasets for the new contracts, as it did for Polymarket v2
History	Your indexer has to backfill it from the blockchain	Full history included from the start
Where the data goes	Whatever you build a connection for	Postgres, ClickHouse, Kafka, object storage or a webhook
Who holds it	You, along with the system you have to maintain	Every row is written to your own database

Built for regulated environments

SOC 2 Type II, dedicated and on-premises deployment, and access controls for banks, payment companies and trading firms.

SOC 2 Type II

Goldsky maintains SOC 2 Type II compliance.

Dedicated infrastructure

Single-tenant infrastructure in Goldsky-managed cloud accounts, with static IPs for allowlisting, on Enterprise.

On-premises and BYOC

Run the full stack on-prem or in your own cloud, so data and compute stay inside your boundary.

Access controls

Role-based access on every plan, plus SAML SSO and audit logs on Enterprise.

Complete, validated data

Every block is checked for completeness before delivery, and daily validation repairs any gaps.

Data for your compliance program

Transfer and identity data for MiCA, Travel Rule and BSA/AML: Goldsky supplies it, your team owns the program.

Deliver on your timeline

Banks, payment companies and trading firms use Goldsky to get onchain products live in weeks. Our engineers work with yours from the first call to production.

Scoping call

We learn what your digital asset product needs and give you a delivery estimate.

Proof of concept

Our engineers and yours build a working version on your own data.

Kickoff and go live

We launch it together and check the numbers against your systems.

Ongoing support

24/7 support in a shared channel with the engineers who run your infrastructure.

Step 1
Scoping call
We learn what your digital asset product needs and give you a delivery estimate.
Step 2
Proof of concept
Our engineers and yours build a working version on your own data.
Step 3
Kickoff and go live
We launch it together and check the numbers against your systems.
Ongoing
Ongoing support
24/7 support in a shared channel with the engineers who run your infrastructure.

Book a scoping call

Build with Polymarket data today

Copy the prompt into your AI coding agent and build a working version today.

Open in Claude Code

I'm building on Polymarket data (order fills, positions, PnL) using Goldsky.

## Set up your tooling first

```bash
npx skills add goldsky-io/goldsky-agent
```

That installs to Claude Code, Cursor, Codex, Windsurf and 30+ other agents. Then connect Goldsky's MCP server at `https://docs.goldsky.com/mcp` however your tool does it — per-client setup is at https://docs.goldsky.com/mcp-server

Finally, use the `auth-setup` skill to install the Goldsky CLI if it's missing, and confirm I'm logged in before you start.

## What I'm building

Follow the Goldsky guide at https://docs.goldsky.com/compose/guides/build-a-prediction-market

Help me:

1. Stream order fills and matched trades from Polymarket into our own database
2. Maintain running per-user balances and PnL from that stream
3. Keep full history available from day one, not just new activity
4. Route filtered fills into a Compose task so we can act on them automatically

## How to work

- Confirm dataset names yourself rather than assuming them
- Write the files, and create any Goldsky secrets the pipeline needs
- Validate locally — `goldsky turbo validate` for pipelines, a local task call for Compose
- Ask me for anything you can't determine: contract addresses, endpoints, thresholds, which database we're writing to
- Put credentials straight into Goldsky secrets, never into this chat

**Stop before deploying anything live**, then show me the commands to run.

Get Polymarket data into your stack

Bring the markets you follow and where you want the data to a 30-minute call and leave with a concrete plan.

Talk to our team

Read the guide

Frequently asked questions

Can we still use the public Polymarket subgraphs?

Not reliably. Polymarket moved to new (v2) contracts on April 28, 2026 and stopped using subgraphs, so the public endpoints now return incomplete data. Goldsky’s v2 datasets replace them.

Where can Goldsky deliver the data?

Postgres or ClickHouse for queries, Kafka or object storage for other systems, or a webhook when you want to react to single events.

Who holds the data?

A copy sits in a database your team runs. The pipeline writes each row there, history included.

Is Goldsky SOC 2 audited?

Goldsky maintains SOC 2 Type II compliance, with the report available under NDA.

Tokenized assets Turbo Pipelines Compose Polymarket data on Goldsky