Anatomy of the .claude/ Folder

Re:Anatomy of the .claude/ Folder by ijustlovemath on Friday March 27, 2026 @10:50PM (#47551073)

Even the most complex distributed systems can be understood with the context windows we have. Short of 1M+ loc, and even then you could use documentation to get a more succinct view of the whole thing.

[ Parent | on HN ]

Re:Anatomy of the .claude/ Folder by zwaps on Saturday March 28, 2026 @03:20AM (#47552354)

This really doesn’t pan out in practice if you work a lot with these models

And also we know why: effective context depends on inout and task complexity. Our best guess right now is that we are often between 100k to 200k effective context length for frontier, 1m NIHS type models

[ Parent | on HN ]

Re:Anatomy of the .claude/ Folder by braebo on Saturday March 28, 2026 @05:24PM (#47558224)

Agreed, not to mention the m additional cost of chats with more context.

[ Parent | on HN ]