Isn't it better to expose the `query` and `. --update` commands to the CLI client?

[duartemvix]

I've been looking to try this and had a few thoughts on how it could be better applied, but please bear in mind that I haven't read it fully and might miss a few nuances. That said, I've thought that:

• Isn't it better to have a query command as a CLI command rather than a skill?

Simply loading the entire skill might not be necessary if I only want to consume the graph.

As well as having a graphify . --update for updating after changes would be easier to instruct to an agent to call.

[safishamsi]

Implemented in v0.3.8. graphify query "<question>" now works as a CLI command — loads graphify-out/graph.json, scores nodes by terms, runs BFS (or --dfs), and prints the subgraph as text. Flags: --dfs, --budget N, --graph <path>.

[duartemvix]

Awesome. Well done.