feat(core): bundle Stack Overflow survey demo data

[sparkplug604]

Summary

This PR makes the Core demo feel more real and useful on first run.

Instead of generating a placeholder Markdown source, praxis demo core now uses a small bundled Stack Overflow Developer Survey dataset. The demo ingests aggregate 2024/2025 AI and tooling survey data, indexes it, and shows an explained search result with trust, freshness, source IDs, and capture IDs.

Using a real public dataset makes it easier to understand what Praxis is doing: turning source material into searchable, traceable, explainable agent context.

What Changed

• Added a small aggregate Stack Overflow Developer Survey demo dataset.
• Included source provenance with URLs, access date, source year, license notes, byte counts, and checksums.
• Added build notes explaining how the aggregate files were derived.
• Updated praxis demo core to ingest the bundled dataset.
• Included the dataset in packaged builds.
• Added smoke test coverage for the explained search output.

Notes

The bundled dataset contains aggregate rows only. Raw respondent-level survey rows are not committed.

The freshness value in the demo reflects when Praxis captured or checked the source, not when the original survey data was published.

Validation

• Ran praxis demo core from a clean temp root.
• Ran the focused demo smoke test.
• Ran compile checks for the touched command modules.
• Built a wheel and confirmed the demo dataset files are included.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
praxis	Error		Jun 15, 2026 5:34am