LLM wiki: your new tool for work

Turns scattered sources into a connected knowledge base that an agent maintains

You will learn how to build a personal wiki with a coding agent—a connected set of pages about your field that the agent writes and maintains itself.

What you will be able to do with it:

  • Prepare for a discussion—gather everything the collection knows about the topic, including the points where sources disagree.
  • Analyze a hypothesis—test your idea against the material you have gathered and see what supports it and what contradicts it.
  • Add your own note—record your own conclusion so that it feeds into every later answer the wiki gives.
  • Survey a new area—build a picture of an unfamiliar topic and see what the collection is missing on it.

By the end, the wiki will merge every new source with everything already recorded, register disagreements between sources with attribution of who claims what, keep a separate list of what the collection lacks, and answer questions with links to the original material.

We will work with a collection of articles about product management, but you can build the wiki on your own content from the start—then you will get a working tool rather than a training example.

By the end of this case you'll have a working product and a set of reusable skills:

working with a coding agent as the editor of a knowledge base: the schema in CLAUDE.md, protocols for ingesting a source, querying and linting
a connected markdown wiki in Obsidian: page types, cross-links, the graph of connections, the index and the log
setting sources against each other: recording disagreements, auditing connectivity, listing the gaps in the collection
versioning the wiki in GitHub and running cloud agent sessions from your phone

How do I get access to the program?

This case is part of the GoPractice “Build with AI” subscription. You can start right now.