--- title: AI-assisted genealogy | daily.dev description: A personal experiment using Claude Code as an autonomous research agent to build a genealogy tree, growing it to over 600 individuals and 12 generations in... canonical: https://daily.dev/posts/ai-assisted-genealogy-9jp15nvxk twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: AI-assisted genealogy | daily.dev og:description: A personal experiment using Claude Code as an autonomous research agent to build a genealogy tree, growing it to over 600 individuals and 12 generations in... og:url: https://daily.dev/posts/ai-assisted-genealogy-9jp15nvxk og:image: https://api.daily.dev/og/posts/9jP15nvxK.png og:image:alt: AI-assisted genealogy og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # AI-assisted genealogy **[A Java geek](https://daily.dev/sources/frankel)** · 10 min read · 8 upvotes · 1 comments ## Summary A personal experiment using Claude Code as an autonomous research agent to build a genealogy tree, growing it to over 600 individuals and 12 generations in under a month. Details a git-based workflow storing data in the GEDCOM 5.5.1 format, with skills for site navigation and record transcription, grammar-check hooks to keep the GEDCOM file valid, and serial subagents for long-running overnight research sessions. Also covers visualizing the resulting tree with Topola Genealogy Viewer hosted privately on Cloudflare Pages, plus practical genealogy lessons learned (birth vs marriage certificates, French civil record access rules, regional naming quirks). ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What file format should I use to store genealogy data I'm building with an AI assistant? GEDCOM 5.5.1 is a practical choice, since it is broadly supported by visualization and desktop tools even though FamilySearch released a newer GEDCOM 7.0 in 2021. GEDCOM structures data by indentation level using leading numbers rather than spaces, with an INDI record representing each individual and nested tags for names, birth details, sources, and citations. _Developers experimenting with AI agents on personal data projects follow workflow write-ups like this on daily.dev._ ### How can I visualize a GEDCOM file as an interactive family tree without exposing private data publicly? Topola Genealogy Viewer renders an interactive tree directly from a static GEDCOM 5.5.1 file and can be wired into a build process such as a GitHub workflow. Hosting the generated site on Cloudflare Pages keeps it private via OTP-based authentication on an allowed email list, unlike GitHub Pages which is public by default; the free Cloudflare Pages tier allows up to 500 builds a month. _Anyone weighing GEDCOM visualization or private static-site hosting options can compare approaches on daily.dev._ ### How do I keep an AI coding agent from drifting during long unattended overnight research sessions? Run the work through a subagent to keep the main agent's context clean, use serial rather than parallel subagents to avoid hitting quota limits mid-session, and have the subagent commit to a worktree branch that the main agent later reviews and merges. This structure, used successfully with Claude Code on Claude Max, prevents drift over long autonomous runs. _Developers tuning agent orchestration for long-running autonomous tasks track patterns like this on daily.dev._ ## Community discussion Top comments from developers on daily.dev. **@agustinbarrientos** · 0 upvotes > How did you separate records for living relatives before sending scans through the model? Keeping the viewer private doesn't cover what leaves the machine during transcription. ## Similar posts on daily.dev - [Updating a side project with AI in 275 commits](https://daily.dev/posts/updating-a-side-project-with-ai-in-275-commits-esedmwnfc) · Ben Hoyt · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#git](https://daily.dev/tags/git), [#claude-code](https://daily.dev/tags/claude-code) [View this post on daily.dev](https://daily.dev/posts/ai-assisted-genealogy-9jp15nvxk) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"AI-assisted genealogy","url":"https://daily.dev/posts/ai-assisted-genealogy-9jp15nvxk","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/ai-assisted-genealogy-9jp15nvxk"},"datePublished":"2026-08-30T16:18:42.957Z","dateModified":"2026-08-30T16:19:12.359Z","description":"A personal experiment using Claude Code as an autonomous research agent to build a genealogy tree, growing it to over 600 individuals and 12 generations in...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/45a30a1e0e84ae982a0157f47b934056?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/45a30a1e0e84ae982a0157f47b934056?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"A Java geek","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"A Java geek","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/b7f3fdbf443347ea84705578a195085b","url":"https://daily.dev/sources/frankel"},"commentCount":1,"discussionUrl":"https://daily.dev/posts/ai-assisted-genealogy-9jp15nvxk","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":8},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":1}],"keywords":"ai-agents,git,claude-code","timeRequired":"PT10M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"A Java geek","item":"https://daily.dev/sources/frankel"},{"@type":"ListItem","position":3,"name":"AI-assisted genealogy"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/ai-assisted-genealogy-9jp15nvxk","comment":[{"@type":"Comment","text":"How did you separate records for living relatives before sending scans through the model? Keeping the viewer private doesn’t cover what leaves the machine during transcription.","datePublished":"2026-08-31T21:38:55.582Z","url":"https://daily.dev/posts/9jP15nvxK#c-QmMG5JDMe","author":{"@type":"Person","name":"Agustin Barrientos","url":"https://daily.dev/agustinbarrientos","image":"https://media.daily.dev/image/upload/s--5ayxQnqn--/f_auto/v1788281802/avatars/avatar_wQYYVe5Tbj0NJ7C7qPoa8?_a=BAMAMicg0"}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/ai-assisted-genealogy-9jp15nvxk#faq","mainEntity":[{"@type":"Question","name":"What file format should I use to store genealogy data I'm building with an AI assistant?","acceptedAnswer":{"@type":"Answer","text":"GEDCOM 5.5.1 is a practical choice, since it is broadly supported by visualization and desktop tools even though FamilySearch released a newer GEDCOM 7.0 in 2021. GEDCOM structures data by indentation level using leading numbers rather than spaces, with an INDI record representing each individual and nested tags for names, birth details, sources, and citations. Developers experimenting with AI agents on personal data projects follow workflow write-ups like this on daily.dev."}},{"@type":"Question","name":"How can I visualize a GEDCOM file as an interactive family tree without exposing private data publicly?","acceptedAnswer":{"@type":"Answer","text":"Topola Genealogy Viewer renders an interactive tree directly from a static GEDCOM 5.5.1 file and can be wired into a build process such as a GitHub workflow. Hosting the generated site on Cloudflare Pages keeps it private via OTP-based authentication on an allowed email list, unlike GitHub Pages which is public by default; the free Cloudflare Pages tier allows up to 500 builds a month. Anyone weighing GEDCOM visualization or private static-site hosting options can compare approaches on daily.dev."}},{"@type":"Question","name":"How do I keep an AI coding agent from drifting during long unattended overnight research sessions?","acceptedAnswer":{"@type":"Answer","text":"Run the work through a subagent to keep the main agent's context clean, use serial rather than parallel subagents to avoid hitting quota limits mid-session, and have the subagent commit to a worktree branch that the main agent later reviews and merges. This structure, used successfully with Claude Code on Claude Max, prevents drift over long autonomous runs. Developers tuning agent orchestration for long-running autonomous tasks track patterns like this on daily.dev."}}]} ```