Repository navigation
Conversation
Introduce a plugin-based model routing step (plugin trigger "chat.model") that can override the target provider/model for LLM streams. Import ProviderV2 and resolve the routed model and corresponding user.model, then use that model throughout logging, provider/auth lookups, runtime selection, providerOptions, and message transforms. Emit routedFrom/reason metadata when routing occurs. Also add helpers (currentUserMessageText, partText) to extract the current user's message text for routing.
Expose a new "chat.model" plugin hook in the plugin interface to allow routing of the model used for an LLM request (returns an optional routed model and reason). Add tests in packages/opencode/test/session/llm.test.ts that cover routing behavior: preserving the original model when no route, failing when routed to an unknown model, passing routed model to chat.headers and chat.params, and an end-to-end routing plugin file written to disk (routing-plugin.ts + opencode.json). Also add test helpers (pluginLayer, llmLayerWithPlugin), import pathToFileURL, and wire up request/response assertions for the routed model and modified params/headers.
This was referenced Jun 3, 2026
Contributor
|
Automated PR Cleanup Thank you for contributing to opencode. Due to the high volume of PRs from users and AI agents, we periodically close older PRs using automated criteria so maintainers can focus review time on the most active and community-supported contributions. This PR was closed because it matched the following cleanup criteria:
PRs created within the last month are not affected by this cleanup. If you believe this PR was closed incorrectly, or if you are still actively working on it, please leave a comment explaining why it should be reopened. A maintainer can review and reopen it if appropriate. Thanks again for taking the time to contribute. |
|
@JGuck , do you have another pr? |
3 tasks done
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.
Issue for this PR
Closes:
Addresses / partially addresses:
Type of change
What does this PR do?
This adds a request-scoped
chat.modelplugin hook that runs before provider/model/auth resolution. A plugin can use it to replace the selected{ providerID, modelID }for the current LLM request.The selected session model is not mutated. Later request hooks, including
chat.paramsandchat.headers, receive the routed model. If a plugin routes to an unknown model, the existing provider lookup fails before sending the LLM request.The hook also receives the current user message
{ id, text }, so routing plugins can make a simple decision from the prompt without requiring full conversation history.This intentionally does not update the model picker, persist routed models across turns, or change subagent inheritance behavior.
How did you verify your code works?
bun typecheckinpackages/pluginbun typecheckinpackages/opencodebun test test\session\llm.test.ts --timeout 30000inpackages/opencodeScreenshots / recordings
Not a UI change.
Checklist