Open Hikmah
181 ·GitHub
What is this?
Open Hikmah is an AI-grounded Quran knowledge graph: you search for a verse, drop it on an infinite canvas, and expand it to surface thematic, linguistic (shared Arabic root), and theological connections to other verses.
Open Hikmah is an AI-grounded Quran knowledge graph: you search for a verse, drop it on an infinite canvas, and expand it to surface thematic, linguistic (shared Arabic root), and theological connections to other verses.Is this a mobile app or a website?It's a web app built with Next.js (App Router) and React, so you use it in a browser rather than installing it on a device.
It's a web app built with Next.js (App Router) and React, so you use it in a browser rather than installing it on a device.How do I start exploring verses on the canvas?Search for a verse by direct reference (e.g. 2:255), keyword, or meaning, place it on the infinite canvas, then expand it by Theme, Root Word, or Contrast to reveal related verses; guest access lets you do this without creating an account.
Search for a verse by direct reference (e.g. 2:255), keyword, or meaning, place it on the infinite canvas, then expand it by Theme, Root Word, or Contrast to reveal related verses; guest access lets you do this without creating an account.What do I need to self-host this?A PostgreSQL database with the pgvector extension and Drizzle ORM for migrations, an Anthropic Claude API key (with Gemini as fallback and for embeddings), and Quran Foundation OAuth2 credentials.
A PostgreSQL database with the pgvector extension and Drizzle ORM for migrations, an Anthropic Claude API key (with Gemini as fallback and for embeddings), and Quran Foundation OAuth2 credentials.Can the AI make up connections between verses that aren't real?No — canonical shared Arabic roots and embedding-based semantic similarity identify candidate verses first, and the AI's role is limited to explaining why a connection holds, never to inventing a reference.
No — canonical shared Arabic roots and embedding-based semantic similarity identify candidate verses first, and the AI's role is limited to explaining why a connection holds, never to inventing a reference.Do I need to create an account to use it?No, guest access lets you explore, bookmark, and use the canvas without signing in; creating an account just adds cross-device syncing of bookmarks and workspaces.
No, guest access lets you explore, bookmark, and use the canvas without signing in; creating an account just adds cross-device syncing of bookmarks and workspaces.What's the 99 Divine Names feature?A full Asma'ul Husna section with Arabic script, root morphology, taxonomy, a related verse feed, and contemplative reflections for each of the 99 names.
A full Asma'ul Husna section with Arabic script, root morphology, taxonomy, a related verse feed, and contemplative reflections for each of the 99 names.What languages is the interface available in?English, Turkish, Russian, and Azerbaijani — covering the UI itself, AI-generated connection explanations, and the 99 Names content.
English, Turkish, Russian, and Azerbaijani — covering the UI itself, AI-generated connection explanations, and the 99 Names content.Can I share a canvas I've built with friends?Yes, every canvas state serializes into a single URL you can copy and share, so others see exactly the same graph you built.
Yes, every canvas state serializes into a single URL you can copy and share, so others see exactly the same graph you built.What social features does it have?Daily engagement streaks, a friends leaderboard, head-to-head reading challenges (24h/48h/7d), and @mentions that notify friends when you tag them in verse notes.
Daily engagement streaks, a friends leaderboard, head-to-head reading challenges (24h/48h/7d), and @mentions that notify friends when you tag them in verse notes.What are similar projects?rankme.fast, m_flow, LibreChat, open-glean, WeKnora, LightRAG, Resume-Matcher, OpenExecutive
rankme.fast, m_flow, LibreChat, open-glean, WeKnora, LightRAG, Resume-Matcher, OpenExecutiveWhich topics is it in?vitest, pgvector, drizzle-orm, retrieval-augmented-generation, vector-search, oauth2, knowledge-graph, vector-database, semantic-search, playwright, embeddings, tailwindcss
vitest, pgvector, drizzle-orm, retrieval-augmented-generation, vector-search, oauth2, knowledge-graph, vector-database, semantic-search, playwright, embeddings, tailwindcss