Repository navigation
feat: add chat-first experience with adaptive visual answers - #92
Merged
Merged
Conversation
jerelvelarde
marked this pull request as ready for review
October 9, 2026 14:49
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The app now opens directly to chat and selects a presentation for each request: text for direct answers, A2UI for basic tables, and sandboxed interactive UI for charts, diagrams, calculators, and maps. Jev makes the typed routing decision; the answer model defaults to
chat-latest.Chat preserves the reader's position during generation, reveals answer blocks as they enter view, and shows copy controls on hover or keyboard focus. Follow-up interactions share the chat thread, and provider/bridge failures surface retryable errors.
The coastal-trip demo uses live USGS tiles and credited Wikimedia photos. Numbered pins drop into place over six seconds with pause/resume, replay, card selection, reduced-motion behavior, and standalone HTML export support. Connections are illustrative, not verified driving directions. The restored Pitch, Roll & Yaw starter generates a Three.js airplane with labeled axes, animated controls, sliders, and reset.
Documentation and repository links now use
CopilotKit/OpenIntelligentUI. Local setup creates placeholders for the required OpenAI and TypeSafe keys. The Open Generative UI runtime API names and deployment service names remain unchanged; the standalone MCP package follows the rename on main asopen-intelligent-ui-mcp. The README follows the OpenDots layout with a tracked engineer CTA, the launch film, and four feature clips. Uploads remain in an unposted issue draft; media loads when signed in but anonymous asset requests currently return 404.The merge preserves main’s branding rename and fixes Jev filtering for the runtime-injected
render_a2uitool. A2UI guidance now uses the actual structured tool arguments; regression tests cover renderer filtering in the real agent graph.Validation
f8d496e: frozen pnpm/uv installs, lint, explicit app TypeScript check, production build, 139 JavaScript tests and 93 Python tests passed. The routing regression failed before the fix and passes after it.git diff --checkpasses.Remaining validation
Local Docker build could not run: Docker Desktop reports that it is unable to start. Existing CI does not build Docker images, so container-image validation remains outstanding; the application production build passed. Container-image validation remains outstanding; the user explicitly requested merging this PR before the separate website attribution change.
Deployment
Set
OPENAI_API_KEYandTYPESAFE_API_KEYon the agent service and verify access tochat-latestandjev-latest. Run a deployed provider-backed smoke test after release. This PR does not deploy the changes.