xr843/fojin

Buddhist Digital Text Platform — 10,500+ texts, 613 sources, trilingual cross-canon, AI Q&A (RAG), knowledge graph, full-text search

What it solves

FoJin provides a unified, verifiable platform for querying and studying Buddhist scriptures. It solves the problem of Buddhist texts being scattered across disparate databases (like CBETA, SuttaCentral, and 84000) with different interfaces and languages, making it difficult for users to find specific passages or compare versions across different canons (Chinese, Pali, and Tibetan).

How it works

FoJin uses Retrieval-Augmented Generation (RAG) over a corpus of 670K+ embedded passages from 612 data sources. The system, featuring an AI assistant named "XiaoJin," retrieves relevant texts using a combination of vector similarity (pgvector + HNSW), keyword reranking, and root-sutra recall. To ensure accuracy and prevent hallucinations, it employs a deterministic citation whitelist and verbatim-quote verification. It also features a trilingual cross-canon alignment system that uses LLMs to verify chunk-level parallels between different language versions of the same sutra.

Who it’s for

  • Researchers and Academics: Those needing citable, verifiable sources and cross-canon parallel reading for Buddhist studies.
  • Students of Buddhism: Individuals seeking grounded answers to scriptural questions without having to manually search multiple databases.
  • AI Developers: Those who want to integrate verified Buddhist knowledge into other AI assistants via the provided MCP server.

Highlights

  • Verifiable AI Q&A: Answers include clickable citations that link directly to the source text, with anti-hallucination guards.
  • Trilingual Parallel Reading: LLM-verified alignments between Chinese, Pali, and Tibetan canons, allowing side-by-side comparison of passages.
  • MCP Server Integration: A Model Context Protocol (MCP) server allows external AI assistants (like Claude or ChatGPT) to access FoJin's verified corpus.
  • KNOWLEDGE GRAPH: An interactive map and graph featuring 110K+ entities and 22K+ teacher-student lineage chains.
  • Extensive Dictionary Support: Integration of 39 authoritative Buddhist dictionaries with over 747K entries.

Related

  • Project
    xr843/Master-skill

    Master‑skill is an open‑source AgentSkill package that lets LLM assistants adopt verified Buddhist master personas. Each master answers queries using real canonical texts (CBETA, BDRC, SuttaCentral, etc.) with citations, enforced by hard‑gate ethics rules and a two‑stage fidelity test suite. Install via `npx master-skill install …`, then invoke slash commands like `/master-huineng` in Claude Code, Cursor, Gemini CLI, or the web chat at fojin.app. The project provides 15 pre‑packaged masters, cross‑tradition comparison tools, offline fallbacks, and a Rust desktop manager.

  • Dispatch
    Cross Canon: Implementing the Bible as a RAG Database

    Cross Canon is a Retrieval-Augmented Generation (RAG) implementation that indexes the Bible to allow for semantic search and conversational querying of scripture.

  • Project
    aa0101181514/tw-legal-rag

    tw‑legal‑rag is an open‑source Python CLI that connects to a hosted semantic‑search service containing 22 million+ Taiwanese judgments, statutes and administrative rulings. It lets you search with natural language, fetch exact case numbers, bundle results (including excerpts, case‑history and citation whitelists) into a JSON file, and run a deterministic citation‑check against LLM‑generated answers. The tool itself never calls an LLM and works without API keys, serving as a retrieval‑augmented generation (RAG) front‑end for legal AI applications.

  • Project
    dosco/graphjin

    A compiler and runtime that provides AI agents with a single governed graph over databases, warehouses, files, and source code via GraphQL and MCP.

  • Project
    KKKKhazix/human-writing

    A framework and toolset for AI agents to eliminate "AI-style" Chinese writing by enforcing human-like rhythms, factual substance, and structural checks.