Write the knowledge base
Agentic retrieval needs a corpus worth exploring, so you start by growing the 8 policies into a knowledge base of 52 documents in 5 sections.
Study first
- Learn why WorkBot chunks by heading. Each
##section becomes one chunk, stored with its document's title and section, so a search hit is a place a person could open. Anthropic's post on contextual retrieval shows what goes wrong without that context, with a chunk that no longer says which company or quarter it describes. - Google's technical writing lesson on organizing large documents. Its advice on task-based headings, each with a short introduction under it, applies to every section you write.
Build
- Write 52 documents as
content/knowledge/<section>/<slug>.md, in 5 sections:policies,manuals,it,faqandpeople. Move the 8 policies from Day 4 intopolicies, and give each document front matter (title,owner,updated) and##headings. - Add a
CLAUDE.mdin the folder that names the sections and the front matter. It also says that each heading must stand on its own, because headings are what search returns. - Turn
src/lib/policies.tsinto a thin view of thepoliciessection, so the Day 4 pages keep working.
Done when
- The folder holds 52 documents in 5 sections, and the Day 4 policy pages still list and render the 8 policies.


