Getting started

You have a name. Now let's build from it.

Krio Griot is a research workspace, not a passive tree. This guide walks you from account creation through your first active research session.

Before you begin

What you need to start

You don't need a complete family tree or extensive documentation to start. All you need is one name — even a partial name — and a willingness to follow evidence wherever it leads.

The platform is built around the 1870 break: the point in American records where most enslaved ancestors were first documented under their own names. Everything in Krio Griot is designed to help you work up to that break, through it, and beyond it.

Before you open your first record: write down everything you already know. Names, places, approximate dates, stories your family told. Oral history is evidence — it belongs in the research log from the start.

Step by step

Setting up your account

  1. 01

    Create your account

    Click Start free on the homepage. Enter your name, email address, and a password of at least 8 characters. The Free plan gives you full access to the core research workspace — no credit card needed.

  2. 02

    Add your first person

    Open the app and go to People in the left sidebar. Click Add person. Start with the oldest ancestor you know by name. Enter whatever you have — a full name, a first name only, an alias, or a description like enslaved woman, Mississippi, c. 1840. You can refine it as you find records.

  3. 03

    Record what you know — and how you know it

    Every person record has a source field. Link your knowledge to where it came from: a family story, a death certificate, a census. This discipline — tracking evidence to its origin — is what makes your research defensible and reproducible.

  4. 04

    Open a research question

    Go to Research and click New question. Write the question you are actually trying to answer — Who were the parents of James Perrin, b. c. 1843 Alabama? — not a vague category. Specific questions produce specific searches.

  5. 05

    Run your first AI research session

    From an open research question, click Research this. The AI assistant will identify relevant record sets, suggest search strategies, and help you interpret what it finds. The conversation is logged automatically — every session becomes part of your research record.

  6. 06

    Upload what you already have

    Go to Archives to upload documents — photographs, letters, census images, death certificates. The archive scanner (Basic and Upgrade plans) reads metadata from PDFs and images and links them to the right people automatically.

How the workspace fits together

The five modules

People Your biographical database. Every ancestor, descendant, and connected person. Relationships, dates, places, and sources all live here.
Research Questions you are actively trying to answer, with AI-assisted sessions and a full log of every search and finding.
Archives Your document collection — images, PDFs, transcriptions. Each document can be linked to one or more people in your tree.
DNA Upload match data from AncestryDNA, 23andMe, or MyHeritage. Track matches by shared cM and sort them into working hypotheses.
Collections Curated exhibits you build from your research — for family reunions, memorial projects, or publishing your work for others to see.

Working with the 1870 break

When you hit the wall

At some point your direct-line research will reach 1870 and stop. Names disappear. The 1870 break is the point where formerly enslaved people appear in the federal census under their own names for the first time — but their earlier lives are either undocumented or documented only in records that identified them as property.

Getting through this break is what the platform is built for. The research AI knows the specific record sets most likely to hold post-emancipation evidence: Freedmen's Bureau records, labor contracts, ration rolls, marriage registers, cohabitation registers. It also knows the slave schedules, estate inventories, and plantation records that can carry a name backward into the antebellum period.

The most useful thing you can bring to the 1870 break is a specific county and a specific decade. Start with the 1870 census entry and work backward and forward from the same place.

Next steps

Where to go from here

Read the full documentation for detailed guidance on each module. If you're unsure what the platform can and can't help with, read What I cannot do — it's an honest accounting of the limits of this work.

Questions about your data and how it's stored are answered in the privacy page.