Purpose Build a trustworthy GitHub Actions pipeline that verifies pull requests, creates immutable release artifacts, uses least privilege, caches safely, and controls deployment concurrency.
Deliverable .github/workflows/ci.yml plus release/deploy workflows that build once, tag by commit SHA, and gate promotion.
Before you touch the code Open the previous phase checkpoint, run the repository's verification commands, and make sure Git is clean. A guided lab is useful only when you know which change introduced a failure. Write down the current browser/API/database behavior you expect before modifying anything.
Architecture trace pull request -> deterministic install -> lint/typecheck/test/build -> trusted main -> immutable artifact -> environment gate -> deployment
Read this flow left to right, then right to left. For each arrow, say what data crosses the boundary, which side trusts the other, and what evidence would prove that this boundary works. This habit prevents framework APIs from becoming disconnected memorization.
Step 1 — Create CI triggered by pull_request with read-only contents permission and explicit concurrency cancellation. Do the smallest version of this step first. After the change, stop and verify it before continuing. Use terminal output, HTTP responses, browser DevTools, tests, or database inspection depending on the boundary. If the result differs from your prediction, do not add more code; explain the mismatch and fix the earliest failing assumption.
Step 2 — Set up Node and pnpm, then install with --frozen-lockfile so manifests cannot drift from the lockfile. Do the smallest version of this step first. After the change, stop and verify it before continuing. Use terminal output, HTTP responses, browser DevTools, tests, or database inspection depending on the boundary. If the result differs from your prediction, do not add more code; explain the mismatch and fix the earliest failing assumption.
Step 3 — Run lint/typecheck/tests/build as separate clear steps or jobs with understandable failure ownership. Do the smallest version of this step first. After the change, stop and verify it before continuing. Use terminal output, HTTP responses, browser DevTools, tests, or database inspection depending on the boundary. If the result differs from your prediction, do not add more code; explain the mismatch and fix the earliest failing assumption.
Step 4 — Add dependency/package cache only after correctness works; key it from lockfile inputs. Do the smallest version of this step first. After the change, stop and verify it before continuing. Use terminal output, HTTP responses, browser DevTools, tests, or database inspection depending on the boundary. If the result differs from your prediction, do not add more code; explain the mismatch and fix the earliest failing assumption.
Step 5 — Create trusted-main release workflow that tags an image/artifact with the commit SHA rather than latest only. Do the smallest version of this step first. After the change, stop and verify it before continuing. Use terminal output, HTTP responses, browser DevTools, tests, or database inspection depending on the boundary. If the result differs from your prediction, do not add more code; explain the mismatch and fix the earliest failing assumption.
Step 6 — Add environment protection/permissions for deployment and ensure stale production runs are not silently canceled mid-release. Do the smallest version of this step first. After the change, stop and verify it before continuing. Use terminal output, HTTP responses, browser DevTools, tests, or database inspection depending on the boundary. If the result differs from your prediction, do not add more code; explain the mismatch and fix the earliest failing assumption.
Checkpoint When the lab works, inspect git diff. Remove accidental debugging output and unrelated edits. Run type checking/tests/build commands relevant to this phase. Commit the checkpoint with a message that describes the capability added, not merely the files changed.
Explain-back Without looking at the lesson, explain why each layer in the architecture trace exists, what this phase added to TaskFlow, and what would break if the new responsibility were moved to the wrong layer. If you cannot explain the boundary, repeat the trace before moving on.
Guided Build Lab matters because it changes a real boundary in TaskFlow. The goal is to understand the inputs, owner, observable output, and failure modes before adding more abstraction.
Restate the goal of “Guided Build Lab” as one observable behavior. Write the success signal before touching code.
Trace the boundary map: route data → UI state → component tree → interaction → accessible feedback. Mark the one boundary this lesson changes.
Inspect the current repository state and record the exact files, route, table, or runtime involved. Avoid creating an abstraction before you can name the existing owner.
Implement the smallest vertical slice. Add one coherent change, run the narrowest verification command, and read the output before continuing.
Connect the slice to TaskFlow’s phase artifact: .github/workflows/ci.yml plus release/deploy workflows that build once, tag by commit SHA, and gate promotion.. Verify behavior from the outside as a user or API client would.
Create one controlled failure that is safe to reproduce. Predict the symptom first, capture evidence, locate the earliest broken boundary, then fix only that cause.
Close the reference and rebuild or explain the core change from memory. Record what you would monitor or test before trusting the same change in production.
if (query.isPending) return <TaskSkeleton />;
if (query.isError) return <TaskError retry={query.refetch} />;
if (!query.data.length) return <EmptyTasks />;
return <TaskTable tasks={query.data} />;Do not copy this as a finished answer. Predict what each boundary does, type the smallest useful slice, run it, then compare the result with your prediction.
Return to the nearest observable boundary: input, network, process log, database, or rendered UI. Compare actual evidence with the predicted path.
Change one variable at a time and reproduce the original failure after the fix; otherwise you have correlation, not diagnosis.
Without copying the final source, create a second small scenario that applies “Guided Build Lab” to a different TaskFlow resource or input. Write the acceptance criteria first, implement it, and provide the evidence that each criterion is satisfied.
Create CI triggered by pull_request with read-only contents permission and explicit concurrency cancellation.
Use the smallest change that demonstrates this lesson. Stop after it works and capture evidence.Using pull_request_target with untrusted checkout/secrets
Predict the symptom before causing the failure. Then locate the earliest boundary where reality diverges from expectation.Why does this responsibility live here, what input does it trust, what output proves success, and what would go wrong if another layer owned it?
If your explanation is vague, repeat the boundary trace before moving on.Close the reference. Reproduce the core idea from memory or explain the exact request/data flow aloud. Then check your diff or notes for accidental shortcuts that bypass validation, authorization, error handling, or verification earned in earlier phases.