SpecStream for Supabase

What you get

• Upserts on entity_idEach record is one row in web_records, keyed by entity_id. The fields column holds values by label, and sources holds the page and read time for each.
• A table that keeps upSave the request on a schedule, hourly to yearly. Each monitor.found delivery carries only the records that are new or changed since the run before.
• Realtime to your appThe table joins the supabase_realtime publication. Add a select policy and each row the function writes reaches your app as it lands.

Set it up

1. 1Create the web_records table in the SQL editor. Row level security starts on, so only the function's service role can write to it.
2. 2Deploy the specstream Edge Function with Supabase's JWT requirement off. It turns away any delivery whose Specstream-Signature doesn't match.
3. 3Register the function's URL as a SpecStream webhook for find.completed and monitor.found, and save its secret with supabase secrets set.

The full setup guide has every step, with what to do if something doesn’t arrive.

• AirbyteComing soonYour lists and their records, in any Airbyte destination
• FivetranComing soonLIST, RECORD, and RECORD_VALUE tables in your warehouse
• dltComing soonRecords into DuckDB, BigQuery, Snowflake, or Postgres
• Meltanotap-specstream from PyPI, into any Singer target