feat(console-ai): one-click answer chips for the plan card's questions#1891
Merged
Conversation
When propose_blueprint derives `questionChoices` for a structure-deciding
question (cloud side), the "Proposed plan" card now renders those options as
one-click chips under the question. Clicking a chip sends an answer message via
the existing onSendMessage channel ("For <question>, go with: <option>.") and
lets the agent continue — the no-code "pick, don't type" affordance the gate
was missing.
- mapMessages.detectProposedPlan lifts `questionChoices: [{text, options}]`
defensively (drops malformed / <2-option / no-text entries), matched to a
question by text.
- ChatbotEnhanced renders the chips under each matching question; questions
with no choices stay plain text (type-to-answer). New `planAnswerMessage`
prop builds the localized answer message.
- Localized on both surfaces (AiChatPage t(), ConsoleFloatingChatbot locale;
en + zh planAnswerMessage). Non-breaking: no choices → current behavior.
Verification: 4 new tests (detect lift + malformed-drop + absent; render chips
+ click sends answer; plain-question no-chip) — plugin-chatbot 126 green;
plugin-chatbot + app-shell + i18n type-check clean.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub. |
Contributor
✅ Console Performance Budget
📦 Bundle Size Report
Size Limits
|
os-zhuang
added a commit
that referenced
this pull request
Jul 18, 2026
…ors (#1878) (#2666) Metadata-liveness audit follow-through (umbrella #1878), objectui side. #1891 (naming drift — dashboard title vs label): DashboardGridLayout read only `schema.title`, but the spec's canonical top-level key is `label`, so a spec-compliant dashboard rendered the literal "Dashboard" header. Fall back to `schema.label` (matching DashboardView's title-then-label precedence). #1895 (designer authoring gaps — skill): the skill authoring form exposed only name/instructions/tools. Add editors for the activation-critical fields that were previously hand-edit-only: - `surface` (ask | build | both) - `triggerPhrases` (NL activation phrases) - `triggerConditions` (field / operator / value) — the activation-critical programmatic conditions plugin-dashboard suite green (89 passing); changed files add no type errors. Claude-Session: https://claude.ai/code/session_01LddW4NaQBdf5FTEnBPpnUJ Co-authored-by: Claude <noreply@anthropic.com>
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.
Why
The plan card's structure-deciding questions were free text — the user had to type. Pairs with cloud#416, which makes
propose_blueprintreturnquestionChoicesfor those questions.What
When a question has derived
questionChoices, the Proposed plan card renders them as one-click chips under the question. Clicking sends an answer message via the existingonSendMessagechannel (For "<question>", go with: <option>.) and lets the agent continue (ask the next question or build).detectProposedPlanliftsquestionChoices: [{text, options}]defensively (drops malformed / <2-option / no-text; matched to a question by text).ChatbotEnhancedrenders the chips; questions with no choices stay plain text (type-to-answer). NewplanAnswerMessageprop.planAnswerMessage). Non-breaking: no choices → current behavior, so it's safe regardless of whether/when cloud#416 is deployed.Verification
4 new tests (detect lift + malformed-drop + absent; render chips + click sends answer; plain-question no-chip) — plugin-chatbot 126 green; plugin-chatbot + app-shell + i18n type-check clean (forced).
🤖 Generated with Claude Code