feat(cli): add experimental stack list by jgoux · Pull Request #6509 · supabase/cli

jgoux · 2026-09-07T23:36:25Z

supabase experimental stack list discovers managed stacks across branches and worktrees, including stopped and unconfigured stacks. It reports persisted identity, runtime, and desired lifecycle in a stable order, with text and structured output.

This PR builds on experimental stack status and reads the managed registry without loading project configuration or contacting stack owners.

github-actions

Superseded by a newer AI review

🤖 AI Review

Only Claude's independent review was available; Codex's review did not complete. All eight Claude findings were verified and confirmed. No major or critical defects were identified.

Findings

Severity	Location	Category	Sources	Claim
🟡 MINOR	`apps/cli/src/commands/experimental/stack/list/list.handler.ts:61`	`error-handling`	claude	One unreadable registry entry aborts the entire listing, and the resulting error does not identify the affected stack directory.
🟡 MINOR	`apps/cli/src/commands/experimental/stack/list/list.handler.ts:56`	`consistency`	claude	The legacy output-flag rejection omits an actionable suggestion, causing text output to show the generic --debug troubleshooting hint instead.
🟡 MINOR	`apps/cli/src/commands/experimental/stack/list/list.handler.ts:43`	`maintainability`	claude	renderRuntime handles StackRuntime with a non-exhaustive fallback, so a new runtime variant would silently render its kind rather than cause a compile-time failure.
⚪ NIT	`apps/cli/src/commands/experimental/stack/list/list.handler.ts:21`	`naming`	claude	compareCodepoint is misnamed because JavaScript string relational comparison orders UTF-16 code units rather than Unicode code points.
⚪ NIT	`apps/cli/src/commands/experimental/stack/list/list.command.ts:11`	`telemetry`	claude	Passing empty flags and config records to withLegacyCommandInstrumentation has no effect.
⚪ NIT	`apps/cli/src/commands/experimental/stack/list/list.handler.ts:46`	`readability`	claude	The complete multi-line stack display is embedded in one long template expression, making user-visible output harder to review and modify.
⚪ NIT	`apps/cli/src/commands/experimental/stack/list/list.integration.test.ts:161`	`test-quality`	claude	A single test combines two independent scenarios: legacy output-flag rejection and registry-read failure.
⚪ NIT	`apps/cli/src/commands/experimental/stack/list/list.integration.test.ts:42`	`test-quality`	claude	mutationCalls also counts read-only findStack and inspectStack calls, so its name misstates what the assertion measures.

Stats

Claude findings: 8 · Codex findings: 0 · Confirmed: 8 · Refuted: 0 · Uncertain: 0

Models: claude-opus-5 + gpt-5.6-sol · Trigger: auto · Workflow run

This review runs once per PR. A maintainer can request another with a /ai-review comment.

github-actions · 2026-09-08T12:48:34Z

Supabase CLI preview

npx --yes https://pkg.pr.new/supabase/cli/supabase@edeccb4273a62bdd4e9316aaf6cf45353ec4e4cd

Preview package for commit edeccb4.

jgoux · 2026-09-08T13:01:28Z

/ai-review

github-actions

🤖 AI Review

Only Claude's independent review was available; the Codex pass did not complete. Three findings are confirmed: one minor telemetry classification issue and two test-quality issues. Two findings are refuted because fail-closed discovery is deliberate existing behavior and the empty JSON case adds no distinct branch beyond existing coverage.

Findings

Severity	Location	Category	Sources	Claim
🟡 MINOR	`apps/cli/src/commands/experimental/stack/list/list.handler.ts:34`	`error-handling`	claude	All stack-discovery failures are classified as invalid configuration, including registry filesystem and permission failures.
🟡 MINOR	`apps/cli/src/commands/experimental/stack/list/list.integration.test.ts:172`	`test-coverage`	claude	The two error tests can pass with an unexpected error type because their substantive assertions are conditional on instanceof LegacyExperimentalStackListError without asserting that type.
⚪ NIT	`packages/stack/src/public/effect-stack.integration.test.ts:1015`	`test-quality`	claude	The new failure-inspection block duplicates the immediately preceding StackStateInvalidError assertion.

Refuted findings (kept for transparency, not posted as review comments)

packages/stack/src/public/EffectStack.ts:1101 (reliability): A single unreadable or unsupported registry entry aborts the entire list operation and prevents healthy stacks from being returned.
Refuted: The propagation is deliberate fail-closed behavior, not introduced by this change. trusted/packages/stack/src/public/EffectStack.ts:1089-1091 already propagates every read failure except the explicitly recoverable remnant case, and trusted/packages/stack/src/public/effect-stack.integration.test.ts:967-987 explicitly verifies that durable missing-state remnants remain fail-closed. The PR preserves that contract while adding stack-id context.
apps/cli/src/commands/experimental/stack/list/list.integration.test.ts:152 (test-coverage): There is no dedicated test for an empty registry using JSON output.
Refuted: There is no distinct empty-JSON implementation branch to leave untested. apps/cli/src/commands/experimental/stack/list/list.handler.ts:77-78 sends every non-text result through the same output.success call, already exercised by the structured-output test at :116-136, while the empty-array result is independently exercised at :152-159.

Stats

Claude findings: 5 · Codex findings: 0 · Confirmed: 3 · Refuted: 2 · Uncertain: 0

Models: claude-opus-5 + gpt-5.6-sol · Trigger: manual · Workflow run

This review runs once per PR. A maintainer can request another with a /ai-review comment.

github-actions · 2026-09-08T13:59:59Z

+const mapStackError = (error: StackDiscoveryError) =>
+  new LegacyExperimentalStackListError({
+    reason: "invalid-config",
+    message: error.message,
+    suggestion:
+      "Inspect the managed stack registry under $SUPABASE_HOME/managed/stacks or ~/.supabase/managed/stacks.",
+    cause: error,
+  });


🟡 MINOR · error-handling · source: claude

All stack-discovery failures are classified as invalid configuration, including registry filesystem and permission failures.

Evidence: apps/cli/src/commands/experimental/stack/list/list.handler.ts:34-41 always assigns reason "invalid-config". packages/stack/src/public/EffectStack.ts:1089-1097 converts failures from exists/readDirectory into StackStateInvalidError, so platform I/O failures reach that classification.

Suggested fix: Preserve enough platform-error information in the stack API to distinguish registry I/O failures, then give those failures a non-configuration actionability and targeted filesystem guidance.

github-actions · 2026-09-08T13:59:59Z

+            if (Option.isSome(error) && error.value instanceof LegacyExperimentalStackListError) {
+              expect(error.value.message).toContain("legacy -o/--output flag");
+              expect(error.value.suggestion).toContain("--output-format");
+              expect(error.value[ErrorActionabilityId]).toEqual(actionability.provideFlags);
+            }


🟡 MINOR · test-coverage · source: claude

The two error tests can pass with an unexpected error type because their substantive assertions are conditional on instanceof LegacyExperimentalStackListError without asserting that type.

Evidence: apps/cli/src/commands/experimental/stack/list/list.integration.test.ts:169-177 and :199-208 assert only that some error exists before conditionally entering the expected-type branch.

Suggested fix: Assert toBeInstanceOf(LegacyExperimentalStackListError) before narrowing in both tests.

github-actions · 2026-09-08T13:59:59Z

+        if (Exit.isFailure(listed)) {
+          const error = Cause.findErrorOption(listed.cause);
+          expect(Option.isSome(error)).toBe(true);
+          if (Option.isSome(error)) {
+            expect(error.value).toBeInstanceOf(StackStateInvalidError);
+            if (error.value instanceof StackStateInvalidError) {
+              expect(error.value.stackId).toBe(orphanId);
+              expect(error.value.message).toContain(orphanId);
+            }
+          }
+        }


⚪ NIT · test-quality · source: claude

The new failure-inspection block duplicates the immediately preceding StackStateInvalidError assertion.

Evidence: packages/stack/src/public/effect-stack.integration.test.ts:1009-1014 already extracts the failure and asserts StackStateInvalidError; :1015-1025 repeats both operations before checking stackId and message.

Suggested fix: Remove the older assertion block and retain the new narrowed block containing the additional checks.

jgoux requested a review from a team as a code owner September 7, 2026 23:36

jgoux mentioned this pull request Sep 7, 2026

feat(cli): add experimental stack logs #6510

Open

github-actions Bot reviewed Sep 8, 2026

View reviewed changes

jgoux force-pushed the feat/experimental-stack-list branch from 92e35d8 to a3e8e9b Compare September 8, 2026 05:31

jgoux force-pushed the feat/experimental-stack-list branch from a3e8e9b to a0e49b1 Compare September 8, 2026 06:02

jgoux added 8 commits September 8, 2026 14:22

feat(cli): add experimental stack list

a1946b6

fix(cli): refine experimental stack list reporting

7d36677

test(cli): fix list container fixture

664e029

test(cli): strengthen stack list ordering assertion

77028b2

test(cli): require both stack sort entries

88add78

fix(cli): address experimental stack list review

f4553d7

fix(stack): satisfy effect lint in stack discovery

ca4f2b2

test(cli): add list operation to stack service fixtures

edeccb4

jgoux force-pushed the feat/experimental-stack-list branch from a0e49b1 to edeccb4 Compare September 8, 2026 12:32

github-actions Bot reviewed Sep 8, 2026

View reviewed changes

Navigation Menu

Uh oh!

feat(cli): add experimental stack list - #6509

feat(cli): add experimental stack list#6509
jgoux wants to merge 8 commits into
feat/experimental-stack-statusfrom
feat/experimental-stack-list

jgoux commented Sep 7, 2026

Uh oh!

github-actions Bot left a comment •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-actions Bot commented Sep 8, 2026

Uh oh!

jgoux commented Sep 8, 2026

Uh oh!

github-actions Bot left a comment

Uh oh!

github-actions Bot Sep 8, 2026

Uh oh!

github-actions Bot Sep 8, 2026

Uh oh!

github-actions Bot Sep 8, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Conversation

jgoux commented Sep 7, 2026

Uh oh!

github-actions Bot left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

🤖 AI Review

Findings

Stats

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-actions Bot commented Sep 8, 2026

Supabase CLI preview

Uh oh!

jgoux commented Sep 8, 2026

Uh oh!

github-actions Bot left a comment

Choose a reason for hiding this comment

🤖 AI Review

Findings

Stats

Uh oh!

github-actions Bot Sep 8, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions Bot Sep 8, 2026

Choose a reason for hiding this comment

Uh oh!

github-actions Bot Sep 8, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

github-actions Bot left a comment •

edited

Loading