Repository navigation
Feature/dual engine ai explorer - #1588
Shashankss1205 merged 2 commits into
Conversation
…raphViewer - Ported glowing connection-aware node visualization from the experimental KnowledgeGraph into the main CodeGraphViewer (Classic/Curvy modes). - Deleted redundant UnderstandPage, UnderstandSection, and knowledge-graph directory. - Integrated summary-engine.ts into CodeGraphViewer's side panel to display Architecture Component Role and Dependency Impact. - Cleaned up ComparisonTable.tsx and RepositorySummary.tsx UI. - Fixed TypeScript union error on VisualizationMode.
|
@roshankumar0036singh is attempting to deploy a commit to the shashankss1205's projects Team on Vercel. A member of the Team first needs to authorize it. |
Shashankss1205
left a comment
Shashankss1205
left a comment
There was a problem hiding this comment.
Verified locally: trial-merged onto main, vite build succeeds (3314 modules), and tsc --noEmit reports the same 5 pre-existing errors as main — no new type errors introduced.
Checked the routing change specifically: gating the server path on ?backend= is safe because cli_helpers.py:829 always sets params = {"backend": backend_url} when CGC serves the playground, so the client-side BYOK path only activates in the static deployment — which is exactly the 405 case this fixes.
Note this PR contains #1587's commit as well, so I'm closing that one as superseded. Thanks!
github-project-automation
Bot
moved this from Backlog tasks
to Done
in CGC Progress Board
Resolves CodeGraphViewer.tsx conflicts with the dual-engine AI explorer (CodeGraphContext#1588): both sides added a sidebar mode, so isImpactMode and isAIMode are chained in the ternary rather than replacing each other. Also updates the _FakeDBManager stub for get_driver(graph_name). Co-Authored-By: Claude Opus 5 (1M context) <[email protected]>
Shashankss1205
added
gssoc:approved
2 participants
Feature: Implement Dual-Engine AI Knowledge Explorer for Static Web Mode
This PR addresses the "405 Method Not Allowed" server errors encountered in the static web deployment when using the AI Knowledge Explorer. Previously, the AI Explorer relied entirely on the Neo4j backend to process queries, which is unavailable in the static web export.
This PR introduces a "Dual-Engine Model":
Changes
executeClientAIQueryinsrc/lib/summary-engine.tsto construct the prompt and handle JSON-mode LLM execution.handleAIQuerySubmitinCodeGraphViewer.tsxto intelligently toggle between the backend API fetch and the browser-based client-side execution.