feat(questionnaires): demo-polish — surface agentic intelligence + smooth operator flow#59
Open
JohnD-EE wants to merge 1 commit into
Open
feat(questionnaires): demo-polish — surface agentic intelligence + smooth operator flow#59JohnD-EE wants to merge 1 commit into
JohnD-EE wants to merge 1 commit into
Conversation
…ooth operator flow Make the already-built questionnaire capabilities visible in the demo and sharpen the operator journey, plus fix a demo-blocking capability-registration gap found during runtime verification. Intelligence made visible: - Contradiction callout: the chat now branches on warning.code and renders a tasteful "I noticed something" notice for flagged contradictions (F4.3 data already streamed to the client; the chat rendered every warning identically). - Progress bar: a slim weighted-coverage bar in the session lifecycle strip, driven by the SessionStatusView.completion.coverage already fetched. - Rationale preview: the model's reasoning previews one-line in the collapsed answer-panel row (full text still on expand). Operator flow: - Preview as respondent: one-click link from the admin detail page to the no-login /q surface for launched anonymous versions (disabled with guidance otherwise) — reuses the existing anonymous-session machinery. - Review & Launch: a draft's primary action surfaces the launch-gate checklist (mirroring assertLaunchable) and enables Launch only when all criteria pass. Demo seed: the Northwind seed now runs anonymously with contradiction flagging on so both surfacings work out of the box (drops name/email profile capture). Fix: buildTurnInvokers now calls registerBuiltInCapabilities() before dispatching. The live turn loop dispatches capabilities directly (not via the orchestration chat handler / agent-call executor, where the platform registers), so on a fresh process serving only questionnaire traffic the handler map was empty and every dispatch returned unknown_capability. Idempotent. Docs: F9.2 runbook marked shipped; P9 (and the P0–P9 build) marked complete. Verified live against the dev server + OpenAI: coverage advanced 0→0.33, a genuine contradiction fired over SSE, and both admin affordances rendered. npm run validate clean; 190 questionnaire unit tests pass. Co-Authored-By: Claude Opus 4.8 <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
1 participant
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.
Summary
Demo-polish pass: make the already-built questionnaire capabilities visible in the demo and sharpen the operator journey, plus fix a demo-blocking capability-registration gap found during runtime verification.
Intelligence made visible
warning.codeand renders a tasteful "I noticed something" notice for flagged contradictions. (F4.3 already streamed this data to the client; the chat had been rendering every warning identically.)SessionStatusView.completion.coveragealready fetched.Operator flow
/qsurface for launched anonymous versions (disabled with guidance otherwise); reuses the existing anonymous-session machinery.assertLaunchable) and enables Launch only when all criteria pass.Demo seed
The Northwind seed now runs anonymously with contradiction flagging on, so both surfacings work out of the box (drops the name/email profile capture).
Fix
buildTurnInvokersnow callsregisterBuiltInCapabilities()before dispatching. The live turn loop dispatches capabilities directly (not via the orchestration chat handler / agent-call executor, where the platform registers), so on a fresh process serving only questionnaire traffic the handler map was empty and every dispatch returnedunknown_capability. Idempotent.Docs
F9.2 runbook marked shipped; P9 (and the P0–P9 build) marked complete.
Verification
Verified live against the dev server + OpenAI: coverage advanced 0→0.33, a genuine contradiction fired over SSE, and both admin affordances rendered.
npm run validateclean; 190 questionnaire unit tests pass.🤖 Generated with Claude Code