Why we publish our provenance.

Every narrator. Every grading. Every methodological choice. TheoAI publishes its provenance because the scholarly community will only trust the platform if it can audit the platform. Provenance is not a feature; it is the credibility of the corpus.

What we publish

The IPSC corpus is published with full provenance. Every hadith carries its source work, its volume and page references, its isnād parsed into individual narrators. Every narrator carries biographical anchors with citations to the classical rijāl literature. Every grading is sourced to the methodological framework that produced it.

The methodology is published as a standalone resource at /methodology. The corpus is published with citation guidance at ipsc.theogrid.ai/citation. The computational procedures are published with their inputs and parameters.

Why publishing matters

A platform that claims to grade hadith but does not publish the procedure is asking the scholarly community to trust without verification. That trust is not the kind of trust the discipline operates on. Hadith scholarship has spent over a thousand years building chains of trust grounded in explicit, verifiable evidence. Any computational system that participates in that discipline has to participate on the discipline's terms.

Publishing the provenance is not a feature we added because users asked. It is the precondition for the platform's legitimacy with the scholarly community we serve. Without it, computational grading is opinion. With it, computational grading is something a scholar can actually use.

What this commits us to

Every change to the methodology is versioned. Every change to the corpus is versioned. Every published grade can be traced back to the inputs and the procedure that produced it. When the methodology evolves — and it will, as scholars engage with the work and surface gaps — the engagement is itself part of the published record.