Home

Glintstone Wiki

Glintstone unifies fragmented cuneiform research data — CDLI, ORACC, eBL, ePSD2, OGSL — so scholars and volunteers can search, browse, and analyze ~353k ancient artifacts in one place.

This wiki is the long-form documentation surface. Code-level knowledge lives in the main repo under .claude/skills/gs-*/ (project skills) and CLAUDE.md (durable rules).

---

Start here

• Assyriology 101 — 5-minute domain primer for new engineers.
• Engineer Onboarding — Day 1–5 ramp-up.
• Ideas — exploratory notes; not yet promoted to GitHub issues.

Getting Started

• The App — using app.glintstone.org.
• The API — hitting api.glintstone.org.
• With Claude (MCP) — using Glintstone tools inside Claude.

Overview

• Attribution
• Open Record
• Report an Error
• What's Missing

Data Model

Narrative docs about the schema and ingestion pipeline.

• Data Sources
• Data Quality
• Data Issues
• Import Pipeline Guide
• ML Integration
• Citation Pipeline Summary

Reference — Data Model

The 5-layer schema, with table-level detail per layer.

• Overview
• Artifacts — Layer 1
• Texts — Layer 2
• Linguistic Analysis — Layer 3
• Lexical Resources — Layer 4
• Scholarly Context — Layer 5

Reference — API

• Overview
• Artifacts
• Search & Agentic
• Dictionary
• Collections
• Composites
• Stats & Meta

Reference — MCP

• Overview
• semantic_search
• summarize_artifact
• interpret_token
• submit_correction

Opportunities

• Overview
• Annotation Ecosystem
• Citation Resolution
• Knowledge Graph
• Trust Infrastructure

Personas

• Early Users
• Hobbyist

The full interactive persona pages remain in-repo at docs/personas/*.html.

Project

• Contributing
• Roadmap
• Team
• Open Questions

Research

Point-in-time research reports. Treat as historical snapshots, not living docs.

• Cuneiform Linguistics Synthesis
• Ecosystem Report
• Academic Workflows
• Curriculum Structures
• BabyLemmatizer Integration
• Schema Pipeline Examples
• Marketing Positioning
• Translation History — Gemini
• Translation History — Grok