More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Jewish Genealogy Toolkit (one-time): Russian Empire & USSR

    1

    *One-time purchase: you get the full English edition as it is today (version 0.5.2, November 2026).

    $49

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Jewish Genealogy Toolkit (one-time): Russian Empire & USSR

    Jewish Genealogy Toolkit (one-time): Russian Empire & USSR

    Example session with this skill installed

    I want to find my great-grandfather. Family story: Moishe Feigel, born around 1880 "near Kamenets-Podolsky", a tailor, came to New York about 1906. I have his US naturalization papers. Where do I start?

    • Read your context and instructions
    • Compiled the jewish genealogy toolkit

    What we know so far

    FactSourceStatus
    Moishe Feigel, born c. 1880family storyunverified
    "near Kamenets-Podolsky"family storytown or district?
    tailor, arrived NY c. 1906naturalization papersdocument: read it

    First question

    Which town is Moishe from, and what were his father's name and age?
    Stop when the town is written in a US document and the same man appears in a record from that town.

    Plan, cheapest first

    1. Declaration of Intention: town and date of birth, word by word.
    2. Passenger list: "last residence" and the relative left behind.
    3. Place for 1906: district, province, Polish and Yiddish names, today's name.
    4. JewishGen, 1875/1897 census indexes: Feigel, Feigl, Faigel, Feygel, Fejgel.
    5. Archive scans on Wikimedia Commons: check the index against the scan.

    Project rules

    • Quote and archive reference for every claim.
    • "Not found" only with stated coverage (source, pages, spellings).
    • Logins and captchas stay with you.

    Next step: send me a photo or transcription of the Declaration of Intention, and I will read the place of birth word by word.

    Names in this example are fictional. The agent reports nothing as found until it has seen the document.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    One-time purchase: you get the full English edition as it is today (version 0.5.2, November 2026). For every future update, choose the subscription listing of the same pack, "Jewish Genealogy Toolkit (English): Russian Empire & USSR". A Russian edition is also available from this author.

    Who this is for

    • You are applying for citizenship by descent (Poland, Lithuania, Germany, Austria, Romania, Israel and others) and need a documented chain from an ancestor to you, not a family legend.
    • Your Jewish family came from the Russian Empire (Ukraine, Bessarabia, Belarus, Lithuania, Poland) and settled in the US, Israel or Europe. You know a town "near Kamenets" and a name that changed in every document.
    • You are a genealogist or advanced hobbyist who wants an AI agent that follows research discipline instead of guessing.

    The problem

    Records from this region are scattered across Ukrainian, Russian and Israeli archives, US immigration files and volunteer indexes. Indexes are incomplete, names change spelling between Yiddish, Russian, Polish and English, and borders moved every few decades. An AI agent left alone will happily "find" what you hoped for, with an invented archive reference.

    What your agent does with this pack

    • Starts properly: an intake interview, project files, one research question at a time, a stop condition for each.
    • Finds the right place and source: the town, district and province for the given year; archive files on Wikimedia Commons; Soviet-era records (war, evacuation, repression, civil registry); the road from a US or Israeli record back to the shtetl.
    • Searches the databases well: JewishGen, FamilySearch microfilms, community surname indexes, Yandex Archive Search, the National Library of Israel's historical Jewish press, with all spelling variants and a control query for every zero.
    • Reads the documents: full-resolution archive scans with bundled scripts (extract the original image, map a volume without an index), revision lists 1795–1858 chained by the "age at the last revision" column, Hebrew and Yiddish columns, old family photos.
    • Covers the Jewish history of the region: pogrom victim and relief lists 1918–1922, Jewish cemeteries and headstones, Ashkenazi DNA matches under endogamy.
    • Delivers a result: evidence levels and negative proof, reconstruction of whole families, a citizenship-by-descent dossier, a family history relatives will actually read.
    • Works as a team: briefs that make a research agent's report checkable, and an optional five-role setup (historian, archive reader, open-sources researcher, fact-checker, supervisor).

    Why it does not make things up

    • Every claim comes with a quote from the document and its archive reference.
    • "Not found" counts only with stated coverage: which source, which pages, which spellings, index or the file itself.
    • Uncertain readings are marked as uncertain, not smoothed over.
    • An index is never treated as proof: the agent goes to the scan.

    Good to know

    • Prompts in any language; the agent answers in yours.
    • Captchas, logins, payments and messages always stay with you; the agent never sends requests to archives on your behalf.
    • Optional Python 3 for the bundled scripts (pypdf, Pillow, numpy for scans). Without Python the skills still work as methods.

    Author

    Anna Shefel, family-history researcher working on Jewish families of Podolia, Bessarabia and Moscow. The pack grew out of my open guide "Finding Ancestors: a practical guide": https://shefelchana.github.io/genealogy-guide-ru/ (source: https://github.com/shefelchana/genealogy-guide-ru).

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Research & Analysis.

    The pack is a set of methods plus five small Python scripts; it contains no database and no private archive access. Some sites need a free account in your own name (JewishGen, FamilySearch, Yandex Archive Search) and you sign in yourself. Reading handwriting depends on scan resolution; the agent says when a page cannot be read. Closed programs (Spain and Portugal Sephardic routes) are flagged, not pursued. One-time purchase = current version, no automatic updates.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed today

    Accounts
    JewishpogromsAmhazikaronDuckarchiveWikimediaGenealogyindexerJ Roots
    Optional
    Python 3 (bundled scripts; pypdf, Pillow, numpy for scans)

    Network access is used only by three optional scripts that query public genealogy sites. Sign-ins stay with you; the agent never enters passwords.

    Permissions

    Terminal / Shell
    Network Access
    Read Files
    Write Files

    Allowed Hosts

    jewishpogroms.info
    pogrom.amhazikaron.org
    inspector.duckarchive.com
    commons.wikimedia.org
    genealogyindexer.org
    forum.j-roots.info

    Shell and Python run the bundled scripts (scan extraction, full-text search). Read/Write Files: project notes and downloaded scans in the user's folder. Browser/Network: public archive and database pages (Wikimedia Commons, JewishGen, genealogyindexer.org, pogrom databases, forum.j-roots.info).

    Listed1 day ago
    Updatedtoday

    What's inside

    Frequently Asked Questions

    Revision lists 1795–1858, metrical books (births, marriages, deaths), the 1875 and 1897 census lists, family and recruit lists, Soviet civil registry, war, evacuation and repression records, pogrom victim and relief lists 1918–1922, cemetery records and headstones, and US naturalization and passenger records as the bridge back to the shtetl. Coverage is deepest for Ukraine (especially Podolia), Bessarabia and the rest of the Pale of Settlement.

    No paid subscriptions are required. Some sites need a free account in your own name (JewishGen, FamilySearch, Yandex Archive Search); you sign in yourself and the agent never enters passwords. Many sources need no account at all: archive scans on Wikimedia Commons, archive.org, genealogyindexer.org.

    It searches every period spelling of a surname (Yiddish, Russian, Polish and English forms, plus OCR losses such as a missing first letter), uses the Beider–Morse and Daitch–Mokotoff indexes with their known limits, treats double Jewish given names as one name, and maps traditional to Soviet names (Moishe to Mikhail, Khaim to Efim). A spelling match is never treated as proof without a second convergence: age, patronymic, place or date.

    26 skills in one pack, each with its own SKILL.md, plus reference files, templates and five Python scripts: archive scan page extraction, volume mapping, full-text search in genealogyindexer.org, pogrom database search and a forum post index.

    A one-time purchase gives you the pack as it is at the time of purchase (version 0.5.2, November 2026). If you want every future update delivered automatically, choose the subscription listing of the same pack instead.

    Yes. orchestrating-genealogy-agents shows how to brief research agents so their reports can be checked, how to split work without overlap, and an optional five-role team (historian, archive reader, open-sources researcher, fact-checker, supervisor). Everything also works with a single agent.

    New in Research & Analysis