Agent:doc-updater & command: update-codemaps #311
-
|
There is an agent called doc-updater, and also a command called update-codemaps. Upon closer inspection, you'll notice that doc-updater also generates codemaps, but the files it produces are different from those generated by update-codemaps. When should I use doc-updater, and when should I use update-codemaps? Would there be conflicts between the docs they generate? |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment
-
|
Short version: use So the intended split is:
They are not meant to create two competing doc systems. If both write the same codemap file back-to-back, the last writer wins, so I would avoid running them independently on the same scope without review. I agree the distinction should be documented more explicitly. |
Beta Was this translation helpful? Give feedback.
Short version: use
/update-codemapswhen you specifically want token-lean architecture maps refreshed. Usedoc-updaterwhen docs in general have drifted; that agent may call/update-codemapsas one step, but it also updates README/guides and other documentation.So the intended split is:
/update-codemapsdoc-updaterThey are not meant to create two competing doc systems. If both write the same codemap file back-to-back, the last writer wins, so I would avoid running them independently on the same scope without review.
I agree the distinction should be documented more explicitly.