Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #36 +/- ##
==========================================
+ Coverage 98.21% 98.24% +0.03%
==========================================
Files 8 8
Lines 615 626 +11
==========================================
+ Hits 604 615 +11
Misses 11 11
🚀 New features to boost your workflow:
|
|
@sbillinge, it's ready for review |
sbillinge
left a comment
There was a problem hiding this comment.
I would rather we try and get them from iCSD or cod as a first cut, than MP
|
@sbillinge. In that case, we don't need a separate tool for this MVP version diffpy.apps. Claude Code itself can download a cif file from COD out-of-box. Should I add a commit to remove this functionality?
I tried MP because it has finer search interface. COD can also work as good if we add a few skills in the future to better instruct the agent to navigate the database. |
|
MP are calculated and not real structures. I think they can work ok but for political reasons it is better for the community if we load structures that were actually solved from data. People use MP because it is free and has an easy to use interface and not because the underlying data are good. |
|
Got it. I will remove this tool and add a minimum skill about COD for user experience. |

What problem does this PR address?
What should the reviewer(s) do?