feat: FM-945 Refine Gameplay E2E Automation to Support Dynamic Assessment Trigger and Complete Level Progression

[ashwinnair-chimple]

Changes

• 5 Refine Gameplay E2E Automation to Support Dynamic Assessment Trigger and Complete Level Progression

How to test

• run automated script npm run test:e2e:headed

Ref: FM-945

Summary by CodeRabbit

• Bug Fixes

  • Improved stone interaction reliability by preventing premature “ready” checks and fully resetting stone-loading state after reset.
  • Enhanced assessment and level-completion transitions so progression happens more naturally with fewer timing-related issues.

• Tests

  • Expanded and refactored end-to-end coverage for full gameplay, natural assessment flow, mini-game transitions, and level completion.
  • Updated serial test orchestration and registration ordering to better match real progression timing.
  • Increased level-end waiting robustness and improved verification of awarded stars.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: f3cb2349-cb43-48aa-a230-fdb7b7ae9c22

📥 Commits

Reviewing files that changed from the base of the PR and between a8f2ed7 and 6589f3b.

📒 Files selected for processing (1)

• e2e/helpers/mock-helpers.ts

---

📝 Walkthrough

Walkthrough

The PR adds live gameplay-state helpers, rewires the assessment and level-completion E2E flows to follow natural puzzle progression, and updates stone readiness handling plus mocked assessment config so the tests wait on actual in-game transitions.

Changes

Natural gameplay assessment flow

Layer / File(s)	Summary
Shared gameplay state and helpers e2e/fixtures/game-fixtures.ts, e2e/helpers/game-state-helpers.ts, e2e/helpers/index.ts, e2e/helpers/mock-helpers.ts, src/components/stone-handler/stone-handler.ts	Adds a full gameplay flow state, live puzzle/overlay accessors, correct-stone lookup, mocked assessment config, and stone readiness state updates.
Serial suite wiring e2e/tests/ftm-assessment-survey-flow.spec.ts	Splits assessment registration around the mini-game sequence and switches the suite to the new full-gameplay state.
Natural assessment progression e2e/tests/isolated/tc-009-013-assessment.spec.ts	Rewrites TC_009–TC_012 to drive puzzle advancement, wait for the assessment overlay, answer the question, and export the new registration entry points.
Post-mini-game completion and level end e2e/tests/isolated/tc-009-013-assessment.spec.ts, e2e/tests/isolated/tc-016-level-completion.spec.ts	Completes the remaining puzzles after the mini-game and updates level completion to wait for the end screen and related UI state naturally.

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~60 minutes

Possibly related PRs

• curiouslearning/FeedTheMonsterJS#1921: Adds the assessment-flow gameplay logic and coordinators that the new E2E helpers read from live state.
• curiouslearning/FeedTheMonsterJS#1963: Changes the assessment overlay and combined-mode transition behavior that the new E2E flow now waits on naturally.

Suggested reviewers

• janfb-codev

Poem

A bunny hopped through stones of light,
and waited for the flow just right.
No magic clicks, no forced goodbye—
the stars arrived beneath the sky.
Hop, hop, hooray!

🚥 Pre-merge checks | ✅ 4 | ❌ 2

❌ Failed checks (2 warnings)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 63.64% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.
Tests & Lint & Coverage	⚠️ Warning	E2E tests were updated, but lint couldn’t be verified (no node_modules/eslint here) and there’s no coverage report or 70% threshold in jest config.	Install deps, run npm run lint, and run coverage-enabled tests until a report shows ≥70% (or add a threshold) before merging.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title is concise, descriptive, and matches the PR’s dynamic gameplay E2E automation and level-progression changes.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch FM-945

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

🧹 Nitpick comments (1)

e2e/tests/isolated/tc-009-013-assessment.spec.ts (1)

744-748: 📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low value

registerTests won't run standalone as implied.

TC_013 waits for the first post-mini-game puzzle, but registerTests chains registerTC009_012 directly into _tc013 with no mini-game (TC_014/TC_015) in between — those live in a separate spec. So a direct file run will hang/fail at TC_013's waitForFunction. Consider dropping _tc013 from the "direct run" path, or clarify in the doc comment that this export is only valid when the mini-game runs externally.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@e2e/tests/isolated/tc-009-013-assessment.spec.ts` around lines 744 - 748, The
standalone test entrypoint is misleading because registerTests currently runs
registerTC009_012 and then _tc013 without the intervening mini-game flow that
TC_013 depends on. Update registerTests to either stop before _tc013 for direct
file execution or revise the doc comment to clearly state that the mini-game
stages (TC_014/TC_015) must be executed externally before _tc013 can run.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Nitpick comments:
In `@e2e/tests/isolated/tc-009-013-assessment.spec.ts`:
- Around line 744-748: The standalone test entrypoint is misleading because
registerTests currently runs registerTC009_012 and then _tc013 without the
intervening mini-game flow that TC_013 depends on. Update registerTests to
either stop before _tc013 for direct file execution or revise the doc comment to
clearly state that the mini-game stages (TC_014/TC_015) must be executed
externally before _tc013 can run.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: ef8954c1-7d91-4f38-9f26-511272358caf

📥 Commits

Reviewing files that changed from the base of the PR and between 143b536 and a8f2ed7.

📒 Files selected for processing (7)

• e2e/fixtures/game-fixtures.ts
• e2e/helpers/game-state-helpers.ts
• e2e/helpers/index.ts
• e2e/tests/ftm-assessment-survey-flow.spec.ts
• e2e/tests/isolated/tc-009-013-assessment.spec.ts
• e2e/tests/isolated/tc-016-level-completion.spec.ts
• src/components/stone-handler/stone-handler.ts