Every answer carries its chain.
A knowledge graph of Qur’an, hadith, and classical fiqh where every node resolves to its source. Query it over HTTP. It returns the evidence and who transmitted it. It does not issue rulings.
Run a query$ curl almizan.page/api/v1/verse/55/9It will not give you a fatwa.
Ask a general-purpose model for a ruling and it will answer fluently, confidently, and without a source. Fluency is not authority.
Al-Mizan returns what the sources say and who said it, then stops. Where scholars differ it returns both positions at equal weight instead of choosing between them.
This follows the International Islamic Fiqh Academy’s Resolution 258 (3/26), and the parallel positions of ICESCO, MUI, Muhammadiyah, and Al-Azhar: AI may assist, but may not issue fatwa.
Built on the field’s identifiers.
Tanzil for Qur’an text. KFGQPC standards. SemanticHadith and SemanticTafsir vocabularies. OpenITI/KITAB identifiers for classical texts.
Every node has a stable, resolvable URI. Two systems cannot contradict each other about a hadith when they use the same identifier for it. Cite Al-Mizan rather than copying it.
Free forever for the texts. Paid for the tooling.
Selling the word of God is not ours to do. Selling graph traversal, export tooling, and verification workflows is honest labour (ujrah). That is what is priced.
- Public
- Free, permanently, no ads. Search, read, trace any chain.
- Researcher
- RM 15/mo. Graph export, private annotations, synthesis.
- Build
- RM 199/mo. API access, higher limits, support.
The Majlis al-Ulama is forming.
Al-Mizan does not decide which position is correct. Qualified scholars do, and we are constituting the council that will.
A verifier reviews a proposed ruling, confirms or rejects the chain of evidence behind it, and signs the result. Every signature is attributed in the record: who verified what, and when.
Scholars constituted: not yet weighed