Follow Discord
Sweep 22 Oct 2026 · 04:32Z Build v2.1.308 527 read Stable v2.1.308 Latest v2.1.308 Next v2.1.308 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.246 ·

Artifact publish limits, symlinked directories, and type-owned file errors

Artifact publishing counts file_path in its limit, handles symlinked directories, and names reserved-file errors.

You'll notice Improvements
JSON All of v2.1.246
You'll noticeTier: how much it should matter to you
2Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
ArtifactsArea: what it touches
ImprovementsKind: in v2.1.246,
ImprovementsSection of the release
What

Publishing a version of an artifact now counts the single file_path entry alongside the files list when it checks the per-version file limit, resolves paths correctly when your working directory is a symlink, and gives specific advice when you try to publish files the artifact type reserves for itself. The over-limit and reserved-file failures both name what went wrong instead of falling back to a generic retry message.

Details
  • The over-limit error now reads "file_path and files list N entries (removals included)", where previously only the files list and removals were counted toward the total.
  • A typed path is resolved against both the working directory and its real path, so publishing from a symlinked directory no longer produces a wrong publish base.
  • When the server marks a file error with the kind type_owned_path, meaning the path is one the artifact type owns, the message names the offending files and tells you to leave them out of file_path and files, rather than reporting an unlabelled failure.
  • The reserved-file message depends entirely on the server reporting that kind on the publish response; without it the older generic error still shows.
  • Applies to every artifact publish with no flag, setting, or environment variable to gate it.
Evidence

` file_path and files list , type_owned_path`

Strings lifted out of the shipped bundle, so the claim above can be checked against them.

See this entry in the whole of v2.1.246 →

Feedback