feat(catalog): publish ultra effort on Codex 0.144+ captures - #32
Merged
architectds merged 1 commit intoAug 19, 2026
Merged
Conversation
Gate max at 0.138.0 (pre-0.138 clients fail the whole catalog on it) and ultra at 0.144.0 (matches bundled native catalogs; withholds on 0.138-0.143 builds that may still reject it upstream). Native merge only - curated models never invent ultra.
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Native GPT models (e.g.
gpt-5.6-sol) advertise anultrareasoning rung in bundled Codex catalogs from ~0.144 onward, but ModelDock was filtering it out at every capture version. This publishesultraonly when backward-compatible.Backward compatibility
captured_with)maxultramodel_catalog_jsonwith a closed enum ending atxhigh; a singlemaxanywhere makes the client exit 1 and publish no models.maxvia an open enum.ultrais gated separately at 0.144 to match when bundled native catalogs started advertising it and to avoid picker rungs on 0.138–0.143 stacks that may still reject it upstream.Curated/routed models are unchanged — only native merge entries that already carry
ultracan publish it.Verification
node --test test/catalog.test.mjs— version-gated ladder tests for old, mid, and current capturescodex exec -m gpt-5.6-solwithmodel_reasoning_effort="ultra"through ModelDock gateway returns 200 and"effort":"ultra"in rolloutScope
Two files only:
src/catalog.mjs,test/catalog.test.mjs. No gateway, auth, or private deployment changes.