fix(console): stop inferring confirmation actions from chat text - #3185
Open
AbigailDeng wants to merge 3 commits into
Open
fix(console): stop inferring confirmation actions from chat text#3185AbigailDeng wants to merge 3 commits into
AbigailDeng wants to merge 3 commits into
Conversation
AbigailDeng
enabled auto-merge
August 4, 2026 09:24
Codecov Report✅ All modified and coverable lines are covered by tests. @@ Coverage Diff @@
## dev #3185 +/- ##
=======================================
Coverage 84.56% 84.56%
=======================================
Files 1503 1503
Lines 140015 140015
Branches 17180 17180
=======================================
+ Hits 118397 118400 +3
+ Misses 13753 13750 -3
Partials 7865 7865
Flags with carried forward coverage won't be shown. Click here to find out more. 🚀 New features to boost your workflow:
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
The Chat page inferred business state from arbitrary assistant prose. Any text
containing words such as
confirm,approval,确认, or同意could create afrontend-only
needs_confirmationstate, renderConfirm and create, and sendthe hard-coded prompt
Confirm. Please create it now.. This is incorrect whenthe assistant is asking the user to choose a Team or provide another value.
This PR:
needs_confirmation/creatingstate path;completed_textand presents them asReady for reply, while preserving structured Studio target handling;normal composer after authoritative Conversation reconciliation;
mistaken for the current product contract.
This is frontend-only. It does not change backend APIs, chat history contracts,
or approval/intervention transport. Existing typed approval action wiring is a
separate concern tracked by #3119 and is not introduced or changed here.
Affected Paths
apps/aevatar-console-web/src/pages/chat/index.tsxapps/aevatar-console-web/src/pages/chat/chatTypes.tsapps/aevatar-console-web/src/pages/chat/index.test.tsxapps/aevatar-console-web/src/locales/en-US.tsapps/aevatar-console-web/src/locales/zh-CN.tsapps/aevatar-console-web/docs/prd/2026-07-01-chat-top-entry-mvp-prd.mdapps/aevatar-console-web/docs/prototypes/2026-07-01-chat-top-entry-prototype.htmlTest Coverage
All changed application paths are covered by the existing Chat route integration
suite. The regression case uses the reported Team-selection semantics and proves
that:
Confirm and createandNeeds confirmationare absent after the final turnand after authoritative reconciliation unlocks the composer;
0does not unlock continuation;minimumStateVersion.Tests: 137 -> 137 test files (+0 new files; 1 existing route integration case
updated). Test coverage audit: 4/4 changed behavior paths covered (100%).
Pre-Landing Review
Four informational findings were fixed before push:
Completedclarification status withReady for reply/可继续回复;Testing, security, performance, maintainability, API-contract, design, and
adversarial reviews found no branch-introduced blocker. Adversarial review also
identified existing typed-approval wiring and Studio target validation risks on
dev; this PR does not modify those paths.Design Review
Design Review (lite): 2 findings, both fixed. The erroneous warning card is
removed, clarification stays in the transcript, and the conversation status now
communicates that the user may reply.
Eval Results
No prompt-related files changed; evals were skipped.
Scope Drift
Scope Check: CLEAN. The change is limited to the frontend Chat behavior,
regression coverage, locale cleanup, and directly affected historical docs.
Plan Completion
No plan file was detected. The reported frontend behavior and recurrence
prevention work are complete.
Verification Results
No browser verification plan was present. Deterministic route integration and
static/build checks were used instead.
Test Plan
pnpm --dir apps/aevatar-console-web exec jest --runInBand --runTestsByPath src/pages/chat/index.test.tsx --testNamePattern 'keeps Team selection in the composer until reconciliation observes a positive Conversation watermark'pnpm --dir apps/aevatar-console-web tscpnpm --dir apps/aevatar-console-web exec biome lint src/pages/chat/index.tsx src/pages/chat/chatTypes.ts src/pages/chat/index.test.tsx src/locales/en-US.ts src/locales/zh-CN.tspnpm --dir apps/aevatar-console-web buildbash tools/ci/test_stability_guards.shbash tools/docs/lint.shgit diff --check origin/dev...HEADThe complete frontend test suite was not run, in accordance with the repository's
incremental unit-test policy.