feat(plugins): add authenticated model catalog filter - #4766
Draft
linonetwo wants to merge 3 commits into
Draft
Conversation
|
This pull request targeted The base branch has been automatically changed to |
This was referenced Aug 3, 2026
Closed
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Why
Frontend auth plugins can allow or deny GET /v1/models, but cannot make the returned catalog match the authenticated caller's provider/model ACL. Returning the global registry leaks unavailable model names and breaks client model discovery; denying the endpoint prevents normal OpenAI-compatible clients from discovering their allowed models.
Our concrete deployment uses CPA native api-keys as the single identity source and a frontend-auth plugin for per-key provider/model grants. It needs model discovery to return exactly the granted model subset without introducing a second gateway or key system.
What
Adds an optional, backwards-compatible ModelCatalogFilter plugin capability and RPC method model.catalog_filter. Filters run after frontend authentication, receive request/access metadata plus the current model list, and chain in plugin priority order. Filter errors fail the model endpoint closed with 503 rather than falling back to the unfiltered global catalog.
Both OpenAI model handlers use the hook. Existing plugins and configurations are unchanged.