fix(wiki-compose): align BYOK backend with pinned Google compose model#991
Closed
cursor[bot] wants to merge 2 commits into
Closed
fix(wiki-compose): align BYOK backend with pinned Google compose model#991cursor[bot] wants to merge 2 commits into
cursor[bot] wants to merge 2 commits into
Conversation
Wiki Compose orchestrator, research, draft, and web_search now resolve a single fixed ai_models.id (google:gemini-3.5-flash) instead of provider-specific fallbacks. Ingest planner keeps resolveComposeModelId. Co-authored-by: Akimasa Sugai <otomatty@users.noreply.github.com>
PR #990 pins Wiki Compose LLM calls to google:gemini-3.5-flash, but the UI still selected user_anthropic/user_openai from AI settings. Session create succeeded while the first graph node threw BackendProviderMismatchError. Reject non-Google BYOK at session preflight for wiki-compose graphs and resolve the compose backend to user_google or zedi_managed only.
Owner
|
develop には |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Context
Review of PR #990 (
fix(api): pin Wiki Compose LLM to google:gemini-3.5-flash) found a regression when combined with the existing Wiki Compose backend auto-selection (#984 / #986).Bug and impact
Bug: Users with Anthropic or OpenAI BYOK configured in Settings → AI had Wiki Compose sessions created with
user_anthropic/user_openai, but every LLM node now resolvesgoogle:gemini-3.5-flash. The first graph step callscreateZediChatModel, which throwsBackendProviderMismatchError.Impact: Wiki Compose was effectively broken for all non-Google BYOK users (session created, run failed on first LLM call).
Root cause
PR #990 removed provider-aware
resolveComposeModelIdfor Wiki Compose graphs without updating frontend backend selection or server preflight validation.Fix
resolveWikiComposeBackendFromAiSettings— onlyuser_googleBYOK orzedi_managed.assertComposeBackendReady— reject non-user_googleBYOK forwiki-compose/wiki-compose-researchat session create (clear 400).Validation
bun test src/__tests__/agents/core/composeBackendValidation.test.tsbun test src/lib/wikiCompose/resolveComposeBackend.test.tsNote for PR #990
Separate operational risk:
google:gemini-3.5-flashmust exist as an activeai_modelsrow or LLM nodes fail atvalidateModelAccess(web_search has fallback; compose LLM nodes do not).