Repository navigation
Pass catalog arrays to jq as files, not arguments - #272
Merged
Merged
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
ArtemSirobaba
approved these changes
Aug 5, 2026
This branch was previously deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The release workflow has failed on every push to main since the catalog reached 28 tools carrying both schema and prompt artifacts:
No release has been cut since, so the catalog is frozen at
release-2026-08-05-105and #271's prompt-doc artifacts cannot reach the signed catalog.The cause is
MAX_ARG_STRLEN, the 128KB cap on a single argv entry, notARG_MAX.ARG_MAXis 2MB on the runner and the whole catalog is ~173KB, which is why the error reads as impossible until you notice the limit applies per argument. The tools array crossed it once tools started carrying schemas and prompts together.Both arrays now reach jq as files via
--slurpfileinstead of--argjson. Temp files go tomktemp -dwith a trap, deliberately not$STAGING, whose contents are uploaded as release assets.Verified by reproducing the failure directly: a 150010-byte single argument fails with the same error under
--argjsonand succeeds under--slurpfile. Every other--argjsonin the script is a number, a single artifact, or bounded by an existing cap (32 schemas, 64 prompts, 256 bundle files), so the final aggregate was the only unbounded one.