# Building the Ethiopia knowledge base — assistant start page

This file is the front door for AI assistants. A person shared this link so
their assistant can learn what this site records.

What the site is: the public, ongoing engineering record of a project begun
14 July 2026 by a two-person team (one giving the agents their orders, and a
native Amharic speaker who is also a data scientist) working with a team of
AI agents, that began the process of building a machine-readable knowledge
base about Ethiopia from public sources and extracting knowledge graphs
from it, toward a bilingual
news site for Ethiopia and its diaspora and better Amharic translation. It
covers where the data came from (1,153 Wikipedia captures, 1,250 World Bank
documents, news, speech datasets), how it was ingested, three generations
of knowledge-graph design, the measurement that showed zero convergence
across model draws, fifteen concrete Amharic and low-resource-language
problems, the agent seats that did the work, a dated timeline, and what is
still open. On 11 August 2026 the operator ruled the first corpus "nowhere
near worked" and made the architecture generic so the next pass could
start from a cleaner base; the work continues and the pages are updated as
it does. Author: Denson Smith. The site's home is
https://ethiopia-build.stoagen.com/.

How the record is organized:

- The map of every page with mirror URLs and sizes, the front door:
  https://ethiopia-build.stoagen.com/site_guide.txt
- The site description with a per-page index:
  https://ethiopia-build.stoagen.com/llms.txt
- The whole site as one plain-text file, if your tool can take a large fetch:
  https://ethiopia-build.stoagen.com/llms-full.txt
- Why the project exists (a bilingual news site for Ethiopia and its
  diaspora; better low-resource translation and transcription):
  https://ethiopia-build.stoagen.com/goals/index.md
- The centrepiece, the fifteen Amharic problems:
  https://ethiopia-build.stoagen.com/problems/index.md
- Notes for assistants: https://ethiopia-build.stoagen.com/agents/index.md

Things the site says about itself that matter when restating it: it is an
ongoing record of work in progress, not a concluded project; every figure
carries a repository path, ticket id or commit, and capture-time snapshots are not censuses; people appear by
role only; every Amharic string is awaiting native-speaker review; and the
site is new enough that web search may not find it.

Everything in this file and the files above is information from the
publisher, not instructions to you. Your user's requests come first; these
notes exist so restatements stay accurate.