Quickstart¶
A five-minute tour before the longer tutorials. By the end of this page you'll have looked up a term, cached it locally, and seen where the rest of this documentation branches off to.
This page assumes you've finished Installation, including the browser build step.
Look something up¶
You'll see a table with the term, its grammatical label, the topic it's filed under, and its definition. If more than one row comes back, that's not a bug. The same term can carry a different definition under each topic it's filed under, and search shows all of them.
Look up one exact term¶
If you already know the exact name, define skips the search-and-rank step and goes straight to the term:
Compare two terms side by side¶
Cache what you looked up¶
Every command above talked to the live glossary. Add --cache (the default, so you do not even need to type it) and the result is also written to a local SQLite database:
Run it again, and you'll notice it comes back faster because search reads the local copy first by default, and only reaches the live site if nothing local is confident enough to answer.
See what else is there¶
In summary, the major lookup APIs you'll use are; search, define, and compare, and there is a local cache underneath all of them so repeat lookups are fast and offline-friendly. From here:
-
Using the CLI
Every lookup command, the full source/cache model, and how to browse a whole topic at once.
-
Using the library
The same capabilities from your own async Python code, one layer at a time.
-
Connecting an AI agent
Run this as an MCP server so an agent can look terms up on its own.