Ancient Near East Research

What it does

Ancient Near East Research provides an academic research toolkit inside an agent environment. It enables retrieval of Hebrew Bible passages (with vocalization and commentaries) via the Sefaria API, access to cuneiform databases (CDLI and ORACC) for Akkadian/Sumerian primary sources, and progressive web discovery: broad neural search (Omnisearch/Exa), page extraction and PDF parsing (Firecrawl), and stealth extraction patterns for bot-protected academic sites (Obscura). The skill bundles scripts for fetching texts, compiling quotations, and OCRing scanned PDFs with a Mistral-based pipeline.

When to use it

Use this skill for literature reviews, comparative philological queries, citation collection, and academic source extraction where canonical texts (Hebrew, Ugaritic, Akkadian) or hard-to-find papers are required. Triggers include Hebrew quotations, requests to 'find papers' or 'search JSTOR/Google Scholar', PDF extraction, and any query referencing Sefaria, CDLI, ORACC, or ANE topics (Minoan, Ugaritic, Tehom/Tiamat comparisons).

What's included

• Scripts: fetch_sefaria.py, fetch_cuneiform.py, compile_quotations.py, mistral_ocr_pdf.py, ane_stealth_fetch.sh
• References: curated passages, domain lists, and extraction patterns in the references/ directory
• Instructions: Includes examples for Sefaria API calls, cuneiform P-number lookups, and a tiered escalation ladder for discovery and extraction.

Compatible agents

Designed for Claude Code style agents and CLI-driven agent frameworks that can run bundled Python scripts and shell helpers. Works well with agents that can call external scripts and handle structured academic output.