Data source not in the catalog? Here is how to connect it without waiting for a vendor roadmap

The problem of needing a connector that doesn't exist

Every connector catalog has its limits. At some point, the data team needs a source that simply isn't there: an in-house tool without any documented public API, or a niche vendor too small for any ingestion platform to prioritize.

Up until now, you had three options, and none of them were good. Wait for the connector to land on a vendor's roadmap, with no timeline and no guaranteed priority. Build the entire ingestion pipeline from scratch, with all the maintenance that entails: pagination, rate limiting, schema evolution, backfills, alerting, all rebuilt from the ground up and maintained by your team forever. Or, settle for an all-in-one platform just because it promises to cover that specific use case, compromising on the rest of your stack.

The worst-case scenario never changes: it's the home-grown pipeline that runs smoothly in month one and becomes a permanent maintenance project by month six.

What is Custom Connector?

Custom Connector allows you to point Erathos to a GitHub repository containing your connector's code. Erathos builds the code, keeps it synced with every push, and lists the result in your workspace catalog with a Custom badge, right alongside the native connectors.

This isn't DIY data ingestion. It's writing your extraction logic once in your repo and running it within the same infrastructure that already handles scheduling, incremental syncs, observability, and alerting for every native Erathos connector.

How it works

1. Install the Erathos GitHub App on the account or organization that owns the connector's code.
2. Tell Erathos which repository to watch.
3. Every push to the watched repository automatically triggers a rebuild of the connector image.
4. After the first successful build, the connector appears in your data sources catalog with a Custom badge, and setup is as simple as configuring any native connector.

Prerequisites

• Admin access to the GitHub account or organization that owns the repository you want to watch.
• A repository containing a valid Erathos connector project, following the development guide.

Why this isn't custom-built data ingestion

Writing the extraction logic doesn't mean taking on the operational overhead again:

Automated builds and syncs. Every push to the watched repo rebuilds and syncs the connector, meaning no parallel deploy pipelines to maintain.

Incremental syncs by default. The same engine that powers native connectors runs your custom connector.

Out-of-the-box observability and alerting. Run times, record volumes, and error contexts are visible for every sync, with automated webhook and alert notifications sent to email, Slack, or Discord.

Your team writes the source integration logic. Erathos remains responsible for everything that happens after it goes into production.

Get started now

Create your Erathos account and connect your proprietary SaaS or any source outside the catalog in just a few steps. Once the first build succeeds, Erathos handles the rest: scheduling, incremental syncing, alerting, and observability, just like any native connector.

Having an internal tool without an off-the-shelf connector shouldn't force you to choose between waiting on a vendor's roadmap or building entire ingestion infrastructure from scratch.

Check out the complete connector development guide and documentation at docs.erathos.com/connectors/custom-connectors.