The active sources feeding ZABAL Gamez. Each stream keeps its full raw files plus a summary and the ideas it surfaces, organized chronologically by where it comes from. All of it lands in Bonfire - the ZAO memory layer - so any bot can read it and build on top.
A data stream is a live (or upcoming) source of activity that matters to ZABAL Gamez - an interview show, a bootcamp lineage, a bounty, a protocol feature, the workshops themselves. We keep three things together for each: the raw files, a summary, and the ideas worth picking up.
The pipeline reuses the rails the ecosystem already runs on: scripts/pull-data-streams.mjs fetches each live source, writes raw captures, appends to the chronological feed, and emits a dispatch that aggregate-dispatches.mjs merges into the Bonfire graph and push-to-bonfire.mjs pushes to the ZABAL Bonfire. Registry: data-streams.json. Feed: timeline.json.
Loading...
Newest first, color-coded by source.
Loading...
Everything here is machine-readable. Read the registry for what each stream is and its pull config, read the timeline for the chronological feed, or query the Bonfire graph for the streams as entities (label DataStream) and their FEEDS edges into ZABAL Gamez.
To add or refresh a stream: drop it in data-streams.json, then run node scripts/pull-data-streams.mjs. Live sources (GitHub content, Farcaster reads via HAATZ) pull automatically; static sources keep curated timeline entries. Full notes in llms.txt.