Skip to content

perf(apex-debugger): shrink bundle via esbuild ESM resolution of effect - W-23313202#7692

Merged
mshanemc merged 5 commits into
developfrom
sm/W-23313202-2-shrink-apex-debugger-bundle-via-esbuil
Jul 8, 2026
Merged

perf(apex-debugger): shrink bundle via esbuild ESM resolution of effect - W-23313202#7692
mshanemc merged 5 commits into
developfrom
sm/W-23313202-2-shrink-apex-debugger-bundle-via-esbuil

Conversation

@mshanemc

@mshanemc mshanemc commented Jul 8, 2026

Copy link
Copy Markdown
Contributor

What does this PR do?

Summary

  • Applies the node-build-only effect ESM conditions override (from W-23293744, PR perf(org-browser): resolve effect ESM to shrink bundle (POC) - W-23293744 #7675) to salesforcedx-vscode-apex-debugger's esbuild.config.mjs, so esbuild resolves effect's ESM build and tree-shakes most of the unused fast-check dependency pulled in via effect/Schema.
  • Extracts the shared effectEsmConditions constant into scripts/bundling/effect.mjs and refactors salesforcedx-vscode-org-browser to consume it too, removing the duplicated literal (kept opt-in per package, not baked into nodeConfig — see ADR).
  • Adds docs/adr/0021-effect-esm-condition-override-scope.md recording the shared-opt-in-constant decision and the promotion criterion for folding it into nodeConfig later.
  • Result: node dist/index.js 5,655,306B -> 4,424,609B (-21.8%); fast-check bytesInOutput 236,897B -> 47,642B (-79.9%). Output stays CJS; effect now resolves dist/esm/* instead of dist/cjs/*.

Plan

.claude/plans/W-23313202.md

Reviewer notes

  • ADR review finding: apex-debugger is single-target, so ADR 0013 (multi-target scoping) doesn't apply here, and the shared effectEsmConditions extraction is correctly wired into both packages' wireit files lists. No violation found, no code change needed.

Test plan

  • salesforcedx-vscode-apex-debugger has no Playwright/e2e suite (jest only, which CI already runs). Runtime confidence for the effect CJS->ESM resolution change rests on org-browser's existing Playwright coverage of the identical mechanism (PR perf(org-browser): resolve effect ESM to shrink bundle (POC) - W-23293744 #7675) plus this package's jest suite.
  • Manually smoke-test the bundled dist/index.js on desktop: launch an Apex debug session and confirm the extension activates without module-resolution errors (effect ManagedRuntime/Schema paths load correctly).

What issues does this PR fix or reference?

@W-23313202@

Functionality Before

Apex debugger's desktop bundle resolved effect's CJS build, pulling in the full fast-check dependency via effect/Schema's unconditional require.

Functionality After

Apex debugger's desktop bundle resolves effect's ESM build (via shared effectEsmConditions), letting esbuild tree-shake ~80% of fast-check out; dist/index.js shrinks ~21.8%.

🤖 Generated by auto-build pipeline. Original WI: https://gus.lightning.force.com/lightning/r/ADM_Work__c/a07EE00002dxZMaYAM/view

mshanemc and others added 5 commits July 8, 2026 00:11
@mshanemc @claude
chore: plan for W-23313202
e0493db 
Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>
@mshanemc @claude
docs(adr): scope effect-esm esbuild condition override - W-23313202
a21dcd0 
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
@mshanemc @claude
refactor(bundling): share effect-esm conditions constant - W-23313202
05fed22 
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
@mshanemc @claude
perf(apex-debugger): resolve effect ESM to shrink bundle - W-23313202
368c7f3 
Node dist/index.js: 5,655,306B -> 4,424,609B (-21.8%).
fast-check bytesInOutput: 236,897B -> 47,642B (-79.9%).
effect resolves dist/esm/* (was dist/cjs/*); output stays CJS.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
@mshanemc @claude
fix: exclude dist/*-metafile.json from apex-debugger VSIX
7f7469d 
Matches the exclusion every other metafile-producing package has;
prevents the 1.75MB dist/node-metafile.json from shipping in the VSIX.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
@mshanemc mshanemc marked this pull request as ready for review July 8, 2026 08:14
@mshanemc mshanemc requested a review from a team as a code owner July 8, 2026 08:14
@mshanemc mshanemc requested a review from kylewalke July 8, 2026 08:14
@mshanemc

mshanemc commented Jul 8, 2026

Copy link
Copy Markdown
Contributor Author

/ai-auto approve

peternhale
peternhale approved these changes Jul 8, 2026
View reviewed changes

@peternhale peternhale left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Peer-approved on behalf of @mshanemc per /ai-auto approve

@mshanemc mshanemc merged commit 792ec72 into develop Jul 8, 2026
68 of 69 checks passed
@mshanemc mshanemc deleted the sm/W-23313202-2-shrink-apex-debugger-bundle-via-esbuil branch July 8, 2026 14:22
mshanemc added a commit that referenced this pull request Jul 9, 2026
@mshanemc @claude
@daphne-sfdc @svc-idee-bot @jonnyhork @dependabot
perf(apex): resolve effect ESM to shrink bundle - W-23313207 (#7717)
ab84252 
* refactor(apex): migrate langServer namespace read to ProjectService - W-23291527 (#7677)

* refactor(apex): read project namespace via ProjectService not SalesforceProjectConfig - W-23291527

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* docs(skills): rescan external + core-extension-api consumers - W-23291527

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor(core): remove dead SalesforceProjectConfig export + class - W-23291527

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor(core): remove orphaned error_parsing_sfdx_project_file i18n key

Sole consumer (SalesforceProjectConfig.handleError) was deleted in this branch.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>

* perf(ci): background npm install during release-artifact download - W-23195719 (#7681)

* chore: plan for W-23195719

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* perf(ci): background npm install during release download in publish workflows - W-23195719

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor: address review findings - W-23195719

- revert stray package-lock.json drift unrelated to the CI-workflow change
- compress verbose validate-vsix rationale bullet in plan doc

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>

* fix: get rid of old broken command SFDX: Execute Anonymous Apex with Currently Selected Text + rename debug command to SFDX: Debug Anonymous Apex with Editor's Selected Text for consistency - W-23329171 (#7680)

* fix: get rid of old broken commands SFDX: Execute Anonymous Apex with Currently Selected Text and SFDX: Debug Anonymous Apex in salesforcedx-vscode-apex-replay-debugger because this is already handled in Apex Log extension

* fix: restore SFDX: Debug Anonymous Apex command because it's used by the Debug code lens

* fix: rename debug command for consistency

* fix: map debug Anonymous Apex command to sf.anon.apex.debug.delegate for code lens

* fix: can debug highlighted text in Anonymous Apex files

* refactor(org): migrate sf.org.logout.all to services - W-23069610 (#7624)

* chore: plan for W-23069610

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* chore: plan for W-23069610

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* refactor(org): migrate sf.org.logout.all to services - W-23069610

Replace the SfCommandlet/ParametersGatherer orgLogoutAll with an Effect
orgLogoutAllCommand registered via registerCommandWithLayer. The picker +
confirm fold inline; AuthRemover.removeAuth wraps in Effect.tryPromise (typed
OrgLogoutError) and already covers alias/config unset, so the redundant
all-path helpers are dropped. Config/state refresh composes as an Effect
(updateConfigAndStateAggregatorsEffect) instead of bouncing through the async
wrapper. Default path keeps its shared helpers with a follow-up TODO.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test(org): cover orgLogoutAllCommand effect - W-23069610

Rewrite the selected-logout suite (renamed orgLogoutAll.test.ts) around the new
Effect command: picker selection drives removeAuth per username + config refresh;
empty selection and declined confirm are caught as no-op success; no-project
fails the precondition; removeAuth rejection surfaces OrgLogoutError. Drop the
stale selectOrgsForLogout mock from the default-path suite.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test(org): e2e real logout removes auth via throwaway org - W-23069610

Add a real-logout step to orgPickers.desktop: confirm the modal (not Escape) for
a dedicated throwaway org and assert removal durably via `sf org list --json`,
plus a secondary UI check that the picker no longer lists it. The cancel steps
keep MINIMAL_ORG_ALIAS authed. New createThrowawayOrg helper mirrors
createMinimalOrg (fixed pre-seeded alias so requireOrgInCI is never reached);
orgE2E.yml pre-seeds + reaps logoutThrowawayOrg. Fixture opts into
window.dialogStyle custom so the confirm button is clickable.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(org): apply review findings to orgLogout - W-23069610

- hoist AuthRemover.create() out of the per-org forEach (one shared instance;
  each create() reloads ConfigAggregator + reads all org files)
- wrap showQuickPick in Effect.tryPromise -> typed OrgLogoutError, not a defect
- remove redundant removeOrgAliases/checkIsCurrentTargetOrg/doUnsetTargetOrg
  block + helpers from OrgLogoutDefault (removeAuth already unsets config+aliases)
- jest: add scratch-confirm success + fail-fast short-circuit tests
- e2e: expectOrgLoggedOut uses explicit Date.now() deadline loop with clear timeout

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* docs(org): apply concise findings to W-23069610 plan - W-23069610

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(org): fix TS errors in orgLogout tests and playwright spec - W-23069610

- OrgLogoutDefault constructor is now no-arg; rewrite orgLogout.test.ts to match
- Move Effect.Effect cast outside .pipe() in orgLogoutAll.test.ts
- Use literal 'Logout' instead of missing packageNls.org_logout_scratch_logout key

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(org): update orgUtil.test.ts for Effect-based updateConfigAndStateAggregators - W-23069610

setUsernameOrAlias now delegates to updateConfigAndStateAggregators() which routes
through getOrgRuntime().runPromise(Effect). Assert runPromise called + write-before-runPromise
ordering instead of testing internal effect steps that run inside the mocked runtime.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(e2e): recreate consumed throwaway logout org on CI retry - W-23069610

The real-logout step removes auth for logoutThrowawayOrg, so any Playwright
retry or the sequential --last-failed step re-runs setup with the org already
gone, tripping requireOrgInCI. Drop the CI gate from createThrowawayOrg so a
miss recreates the org (CI has dev-hub auth); the pre-seed only saves attempt 1
a create. Also bump the secondary 'picker no longer lists throwaway org'
assertion timeout to 15s to absorb picker-refresh lag after logout.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(e2e): capture sf org create scratch failure body for throwaway org - W-23069610

The throwaway-org create failed on CI with a bare 'Command failed: sf org
create scratch …' and no diagnostic — execAsync drops the CLI --json error
body (scratch limit, dev-hub auth, timeout). Add runScratchOrgCreate in
shared.ts: on failure it persists raw stdout/stderr to
~/.sf/vscode-spans/org-create-failure-<ts>.txt (CI copies that dir into
test-results) and rethrows an error including the body. Wire it into the
throwaway setup (the only create that actually runs on CI).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(e2e): create force-app dir for throwaway org create - W-23069610

sf org create scratch validates packageDirectories paths exist on disk;
sfdx-project.json declared force-app but it was never created, causing
MissingPackageDirectoryError on CI when the org is re-created on retry.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(services): clear StateAggregator before listAllAuthorizations - W-23069610

AuthInfo.listAllAuthorizations reads via a cached StateAggregator whose
orgs.readAll only adds to its configs map and never evicts entries for auth
files deleted out of process. A logged-out org therefore lingered in the
cache and the org pickers kept listing it (orgPickers e2e: throwaway org
still listed after logout). Reset the instance to force a fresh disk read.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(org): unset local target-org on logout via AuthRemover projectPath + skipCache - W-23069610

Logout left the local target-org set because AuthRemover.init resolved
local config from process.cwd() and trusted a stale process-cached
ConfigAggregator, so removeAuth's unsetConfigValues never matched the
current target-org.

Pass projectPath (from the workspace) + skipCache: true at both
AuthRemover.create sites (orgLogoutAllCommand and OrgLogoutDefault) so
config is resolved dir-explicitly and re-read from disk before unset.

Requires forcedotcom/sfdx-core#1309 (adds those options); CI will fail
until a @salesforce/core release includes it.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(org): clear defaultOrgRef in-process on default-org logout - W-23069610

OrgLogoutDefault.run relied on the async config-file watcher to clear the
in-process defaultOrgRef after removeAuth. Span evidence from the failing
apex-testing clearOnLogout e2e (all desktop platforms) shows no
getConnection/clearDefaultOrgRef fires post-logout within the 60s window, so
TargetOrgRef keeps the logged-out org and the apex-testing tree never clears.

removeAuth (authRemover.js) only writes config.json on disk; it never touches
the in-process ref. Restore the deterministic in-process clear (present on
develop as doUnsetTargetOrg, dropped when this path was migrated) by calling
ConfigService.unsetTargetOrg() after removeAuth — it unsets local target-org,
invalidates the aggregator, and clears defaultOrgRef synchronously.

Test: restore the org-runtime layer harness and assert unsetTargetOrg runs on
success and is skipped when removeAuth rejects.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(org): only clear target-org ref when logged-out org was the target - W-23069610

Review feedback: the previous commit cleared defaultOrgRef unconditionally,
which would wipe the current target-org even when logging out a different,
non-target org (logout picks any authed org).

Restore develop's guard: check ConfigService.isCurrentTargetOrg(username,
aliases) BEFORE removeAuth (afterward the config key is already unset), and
only clearTargetOrgRef() when the logged-out org was the target. Thread the
resolved aliases into OrgLogoutDefault so alias-matching works. Add a jest case
asserting the non-target path leaves the ref intact.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* docs(flakiness-review): scan all branches, not just develop (#7682)

* fix(flakiness-review): scan all branches, not just develop

The primary E2E suites (coreE2E.yml + every "* E2E (Playwright)" workflow)
use `push: branches-ignore: [main, develop]`, so they NEVER run on develop —
only on PR/feature branches. The scan's `gh run list --branch develop` filter
therefore excluded the entire coreE2E signal: a 5-day scan saw 11 runs when
~1300 E2E runs (100+ failures) actually existed across 17 workflows.

Changes:
- Enumerate E2E workflows via `gh workflow list`, then query each separately.
  A single `gh run list --limit N | filter` truncates: this repo produces
  ~1000 runs/5d and bot workflows dominate the cap before the name filter runs.
- Bound the window server-side with `--created >=CUTOFF` (macOS/GNU date).
- All run filtering is deterministic jq in ONE pipeline (drop develop/main
  startup-failure noise, compute hadRetries, keep only failures + retry-masked
  passes): 1300 raw -> ~140 actionable. The scan agent returns the output
  verbatim instead of reasoning over 1300 rows (which stalled it at ~1M tokens).
- Capture `headBranch` per run (RUNS_SCHEMA) to attribute failures to branch.
- Relabel metrics Part-1 counts as "actionable runs" since successes are now
  filtered upstream.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* chore(review-diff): denylist backlog-grooming, work-item-sequencing, dependabot-alerts

These are operational/planning skills, not code-review lenses.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>

* chore(deps): bump @salesforce/core from 8.32.0 to 8.32.2 W-23349530 (#7684)

* chore(deps): bump @salesforce/core from 8.32.0 to 8.32.2

* chore(deps): pin @salesforce/core to ^8.32.2 across packages

* docs(flakiness-review): parse real Playwright artifacts, cap retryRate, thread runIds - W-23334955 (#7685)

* chore: plan for W-23334955

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix(flakiness-review): parse playwright report.json via jq; retryRate as retried/total

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix(flakiness-review): thread runIds from parse through testMetrics to clusters

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix(flakiness-review): per-leg grouping, distinct-run counts, shared report.json decode

Group metrics by (testName, workflowName) so OS/suite matrix legs stay
distinct; count distinct runIds for totalRuns/retry/fail so per-report
duplicates (and <run-id>-extra dirs) cannot inflate counts. Make retried
mutually exclusive with unexpected failures. Extract a single shared
decode_report recipe reused by Collect metrics / Cluster failures /
Download artifacts, replacing the dead results.json globs and the python3
span-parse heredoc with jq.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>

* fix: correct handling of log levels to be able to debug Anonymous Apex - W-23339705 (#7683)

* fix: explicit debug levels in DebuggingHeader + fix regex to handle cases where Visualforce is the last category

* test: add a new E2E test for debugging Anonymous Apex

* fix: remove dead getOrgApiVersion + FlagParameter exports from utils-vscode - W-23355254 (#7697)

* chore: plan for W-23355254

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* fix: remove dead getOrgApiVersion + FlagParameter exports from utils-vscode - W-23355254

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* perf(apex-debugger): shrink bundle via esbuild ESM resolution of effect - W-23313202 (#7692)

* chore: plan for W-23313202

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* docs(adr): scope effect-esm esbuild condition override - W-23313202

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor(bundling): share effect-esm conditions constant - W-23313202

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* perf(apex-debugger): resolve effect ESM to shrink bundle - W-23313202

Node dist/index.js: 5,655,306B -> 4,424,609B (-21.8%).
fast-check bytesInOutput: 236,897B -> 47,642B (-79.9%).
effect resolves dist/esm/* (was dist/cjs/*); output stays CJS.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix: exclude dist/*-metafile.json from apex-debugger VSIX

Matches the exclusion every other metafile-producing package has;
prevents the 1.75MB dist/node-metafile.json from shipping in the VSIX.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* chore(soql): @salesforce/core devDependency (type-only) - W-23354450 (#7688)

* chore: plan for W-23354450

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* chore(soql): @salesforce/core devDependency (type-only) - W-23354450

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>

* refactor(replay-debugger): remove direct dependency on salesforcedx-vscode-core - W-23336192 (#7693)

* chore: plan for W-23336192

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor(replay-debugger): remove direct dependency on salesforcedx-vscode-core - W-23336192

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor: apply medium/low review findings - W-23336192

- fix misleading RED/GREEN test comments; describe actual _doInitialize id-gating
- dedupe WorkspaceContextUtil stub via shared factory
- trim plan prose to fragments
- drop stale apex-replay-debugger row from external-consumers skill

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix(e2e): drop core from replay-debugger fixture extension dirs - W-23336192

Core is no longer an extensionDependency and its VSIX is no longer built by
wireit, so VSIX resolution failed with "Expected exactly 1 VSIX in
packages/salesforcedx-vscode-core/ but found 0". The metadata extension's own
fixtures deploy via useMetadataExtensionCommands without core installed, so it
is not needed at runtime.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>

* ci: run check:knip in validatePR - W-23277534 (#7691)

* chore: plan for W-23277534

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* ci: run check:knip in validatePR - W-23277534

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor: apply review findings - W-23277534

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* refactor(apex-testing): extract execution service from ApexTestController - W-23165784 (#7637)

* chore: plan for W-23165783

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* refactor(apex-testing): add ApexTestTreeService skeleton (Refs + typed errors + layer) - W-23165783

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor(apex-testing): discovery/restore/populate -> ApexTestTreeService - W-23165783

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor(apex-testing): delegate ApexTestController state to tree service - W-23165783

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test(apex-testing): add ApexTestTreeService fresh-layer tests - W-23165783

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(apex-testing): apply review findings to discovery service - W-23165783

- discoverTests: catchAll -> mapError (keep typed error channel)
- restore path: catchAll -> catchTag -> RestoreResultsError; flatten inline gen
- restore scan: sequential for/tryPromise -> Effect.forEach; use FsService.stat
- populate: Promise.all -> Effect.all unbounded; drop mutable counter via flat forEach
- reset: drop Effect.all/flatMap indirection; for-loop -> forEach
- sessionMethodIds delete loop -> forEach
- reorder restoreResultsBody before restorePreviousResults
- narrow service surface: drop dead Refs + populate sub-steps from return
- inline sortByMtimeAscending; remove dead wrappers + testController export
- clearResults: add withSpan tracing

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test(apex-testing): apply review findings to tests + plan - W-23165783

- move sortUrisByMtimeAscending test to sortHelpers.test.ts (live helper)
- drop redundant getters test (covered by reset identity assertion)
- assert exact error message in discovery-failure test
- plan: trim duplicated TrySubjectIdError/skills prose (concise)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(apex-testing): widen run helper R type in ApexTestTreeService tests - W-23165783

The service internally yields ExtensionProviderService (via restoreResultsBody)
but declares dependencies:[] so TypeScript infers the full R union. The run helper
type was too narrow (ApexTestTreeService only); cast to satisfy the provided layer.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* chore: plan for W-23165783

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* refactor(apex-testing): restore apply-loop -> Effect.forEach + per-item URI - W-23165783

Replace the for...await inside a single outer tryPromise with Effect.forEach
(concurrency:1, oldest-first) and per-item tryPromise so a failing URI surfaces
in its own RestoreResultsError. Add uri:Schema.String to RestoreResultsError and
populate it at every scan/apply construction site; annotate the restore logWarning
with uri. Add fresh-layer test asserting the offending URI surfaces and ordering
is preserved.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(apex-testing): apply critical/high review findings - W-23165783

- RestoreResultsError.uri now optional; drop empty-string placeholder on
  the pre-scan NoDefaultOrgError catch (no file URI to attribute)
- migrate surviving console.debug/error in shell catch branches to
  Effect.logWarning/logDebug/logError via the package runtime
- clearResults: replace Effect.catchAll with catchTags over the typed
  error union instead of erasing the error types
- notifyDiscoveryFailure/restore notification: use vscode.window.show*Message
  directly instead of the legacy notificationService wrapper

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor(apex-testing): apply medium/low review findings - W-23165783

- test: spy on vscode.window.show*Message (notification migration); drop
  the notificationHelpers mock
- test: comment baseLayer fresh-instance isolation and why the restore
  test builds its own layerWithFs
- plan: trim duplicated Context/Template prose per concise style

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor(apex-testing): address PR review on discovery service extraction - W-23165783

- reuse getMessageFromError from effect-ext-utils; drop local errorToString
- drop over-wide DiscoveryFailure union; narrow notifyDiscoveryFailure param
- collapse SuiteRetrievalError construct-then-catch into a single catchAll
- doDiscover catchTags -> catchAll; fix latent leak of ServicesExtensionNotFoundError/
  InvalidServicesApiError/NoWorkspaceOpenError from the restore path (now non-fatal catchAll)
- convert tree-map getters + getTestClassName/clearAllSuiteChildren to module-level
  functions, removing 6 class-methods-use-this suppressions; update callers + tests
- add JSDoc to ApexTestDiscoveryService; drop redundant layer-merge narration comments

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* docs(doc-maintenance): cut per-item list/merge narration, prefer JSDoc on definition - W-23165783

Codifies the extensionProvider Layer.mergeAll review finding: one-line-per-arg
narration that restates a referenced symbol belongs in a JSDoc on the symbol's
own definition (hover surfaces it), not as a line comment at the use site.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* docs(skills): add i18n-messages skill - placeholder semantics + style guide - W-23234337 (#7633)

* chore: plan for W-23234337

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* docs(skills): add i18n-messages skill - placeholder semantics + style guide - W-23234337

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor: medium/low review findings

- concise SKILL.md scope bullets, reuse + per-surface phrasing
- note bare-string skips format() (message.ts:37 guard)
- sync plan line refs to actual i18n.ts (placeholder 72/78, Delete 74, Logout 75, tree 103)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor(org): migrate sf.org.delete.username to services - W-23069611 (#7619)

* chore: plan for W-23069611

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor(org): migrate sf.org.delete.username to Effect + TerminalService - W-23069611

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test(org): cover orgDeleteUsernameCommand service migration - W-23069611

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test(org): e2e orgDeleteUsername asserts real delete + picker refresh - W-23069611

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(org): cancellation aborts delete loop; hoist services; export OrgDeleteFailedError - W-23069611

Apply review findings to orgDeleteUsernameCommand:
- only TerminalServiceError is caught per-org (Either.left); UserCancellationError
  from withCancellableProgress now propagates and aborts the loop instead of being
  swallowed by Effect.either and surfacing a spurious error toast. Fixes the inverted
  deleteOne doc comment.
- resolve PromptService/TerminalService once, capture deleteOne by closure (no per-org re-yield)
- channel.showChannel (Effect-native) instead of legacy channelService singleton
- Effect.forEach over results uses { discard: true }
- subcommand uses org.orgType directly; orgTypeLabel helper removes duplicated mapping
- export OrgDeleteFailedError + @ExportTaggedError
- new jest test exercises cancellation-aborts-loop (was untested: stub was identity passthrough)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test(org): fix orgDeleteUsername e2e timeout budget + backoff - W-23069611

- reduce waitForOrgGone to 120s (delete completes in seconds) so create -w 10 (600s)
  + this no longer exceed test.setTimeout; bump setTimeout to 900s for headroom
- exponential backoff (1s->10s) in waitForOrgGone instead of fixed 5s poll

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* docs(plan): tighten W-23069611 wording; note src+test co-modification exception - W-23069611

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor(org): accumulate org-delete failures with Effect.partition - W-23069611

Replace hand-rolled Either-wrapping + catchTag + re-loop accumulation with
Effect.partition. Move the cancellable progress to wrap the whole loop so a
Cancel interrupts (and aborts) rather than being bucketed per-org as a typed
failure; partition's per-element Effect.either passes the interrupt through.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(apex-testing): correct @ExportTaggedError knip tags - W-23069611

NoDefaultOrgError is only used in-package (TS4023 export) so it needs the tag;
the four codeCoverageService errors are imported by colorizer.test.ts so their
tags were redundant. Resolves knip Unused-exports + Tag-hints failures.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* docs(effect): add error-accumulation (partition/validateAll) guidance to skill

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

* chore: plan for W-23165784

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor(apex-testing): typed errors for payload + temp-folder failures - W-23165784

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor(apex-testing): extract execution service from ApexTestController - W-23165784

Move run/debug/result-processing region (runTests/debugTests/executeTests/
updateTestResults/clearStaleTagsForTests/onResultFileCreate + cacheSingleSelection/
getTempFolder) into ApexTestExecutionService. lastProcessedResultFile becomes a
service-owned Ref<Option<URI>>. ChannelService/FsService reached via api.services
(class values are type-only exports cross-package; ChannelService runtime instance is
ChannelServiceLayer('Apex Testing'), not .Default). Sentinel emitted on run path only.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test(apex-testing): unit tests for ApexTestExecutionService - W-23165784

Add apexTestExecutionService.test.ts (sentinel emit once on run path / none on debug,
runAllTestsInOrg branch, PayloadBuildError + TestTempFolderError surfaces, stale-tag
clear, lastProcessedResultFile Ref dedupe, empty-suite path, debug org-only filter).
Rewire testController.test.ts mock runtime to provide ApexTestExecutionService.Default +
a stub ChannelService; move debug-path coverage into the service suite.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* test(apex-testing): e2e gate Explorer run-path sentinel - W-23165784

Gate the net-new Explorer run-path 'Ended SFDX: Run Apex Tests' sentinel in the
testExplorer tree-item class-run step (clear Apex Testing channel before the run, assert
the sentinel after). Also harden apexTestExecutionService.test.ts types so it compiles
under test/tsconfig (ts-jest isolatedModules had masked the casts).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix: critical/high review findings - W-23165784

- testController.incrementalUpdate: log non-fatal error, not bare catch
- runTests: extract debug/execute branch to named Effect.fn
  (runTestPipeline) for span visibility; forEach+push -> Array.from spread
- testController.clearResults: Effect.fn over Effect.gen+withSpan

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* refactor: medium/low review findings - W-23165784

- i18n apex_test_results_restored_message: %s -> %d for count slot;
  drop String() wrapper at call site
- restorePreviousResults: Effect.promise for showInformationMessage
  response (acted on) instead of Effect.sync + floating .then
- constants RESULT_MAX_AGE_MS: derive from Duration.days(1), no raw ms
- doDiscover: catchTags(DiscoveryError, PackageResolutionError) over catchAll
- docs/skill/plan: tighten run-on sentences (concise style)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(e2e): selectOutputChannel ensures output panel is open - W-23165784

Intervening UI (test run, maximized panel toggle) can collapse the output
panel between selectOutputChannel calls. The bare 5s waitFor for visibility
then timed out at testExplorer.headless.spec.ts:129. Reuse ensureOutputPanelOpen
to (re)open the panel instead of assuming it is visible.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

* fix(e2e): restore Test Results tab after sentinel check to unblock method-run step - W-23165784

The class-run step selects the Apex Testing output channel to assert the run-path
sentinel, which swaps the bottom panel away from Test Results. The following
single-method step then asserts /Pass Rate/i (rendered in Test Results) without
re-activating the tab, so on the slower ubuntu/windows runners it hung the full
TEST_RUN_TIMEOUT (10m) and pushed the job past the 60m limit. Restore the Test
Results tab after the sentinel assertion.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix(e2e): confirm suite quick-pick with Enter so run-suite executes - W-23165784

selectSuiteInQuickPick relied on selectQuickInputOptionByTyping's synthetic DOM
click to accept the pick. On slower CI runners (ubuntu/windows) that click only
highlights the row without accepting it, so the Run/Add Apex Test Suite command
never fires. The Apex Testing output channel then stays empty until the 600s
TEST_RUN_TIMEOUT, and three such retries blow the 60m job budget -> job cancelled.
Trace confirmed: sf.apex.test.suite.run span never emitted on ubuntu. Press Enter
to confirm the pick, guarded so it only fires while still on the suite prompt.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix(e2e): gate suite-run output on progress toast, not empty channel - W-23165784

The async org run appends nothing to the Apex Testing channel until it
completes, so polling channel content burned the whole test budget on an
empty channel while the run was still healthy — timing out on slower
ubuntu/windows runners. Wait for the run progress toast to appear then
clear before asserting output, mirroring the passing testExplorer path.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix: move apex test execution errors out of views layer - W-23165784

High review finding: utils/payloadBuilder.ts imported error schemas from
views/, the only utils->views edge. Errors are pure Schema.TaggedError with
no vscode/view dependency; relocate to utils/ so the dependency direction is
views->utils like the rest of the package.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* refactor: apply medium/low apex-testing review findings - W-23165784

- payloadBuilder: buildAsyncPayload via Effect.tryPromise (typed failure, not
  a defect) and flatten the async IIFE into per-branch Effects; empty payload
  fails with PayloadBuildError via filterOrFail (no optional payload field).
- executeTests span: wrap the actual debug/execute work in withSpan('apexTestRun')
  instead of the trailing annotate-only (dead) span.
- getTempFolder: catchTags(NoDefaultOrgError, NoWorkspaceOpenError) instead of
  catchAll, matching apexTestRunCodeAction.
- testController: route incremental-update catch through Effect.logWarning
  instead of console.warn; revert clearResults Effect.fn IIFE back to
  Effect.gen + withSpan (effectFnIife).
- trim verbose ExecutionContext/service doc comments; numerals in plan doc.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* fix(e2e): gate single-method Explorer run on output channel - W-23165784

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix(e2e): gate Explorer single-method run on sentinel + Passed decoration - W-23165784

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor(apex-testing): consolidate stale-tag helpers + split runTests - W-23165784

Address PR #7637 review comments:
- Consolidate inverse stale-tag ops into ApexTestTreeService: markStaleTags
  (add + propagate) and clearStaleTags (remove + propagate), sharing the
  classHasStaleMethod/suiteHasStaleClass propagation predicates. Deletes the
  shell applyStaleTags and the execution service clearStaleTagsForTests.
- DiscoveryContext: drop applyStaleTags callback; add staleTag + getSuiteToClasses.
- Split runTests scope-narrowing/suite loops into pure helpers
  (narrowToStaleMethods, resolveUnloadedSuites, expandSuitesToMethods, coveredClassNames).
- Reframe the ChannelService/FsService api.services-bridge comment as the
  apex-testing convention (not a workaround).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* refactor(apex-testing): address PR review comments

- write* helpers → Effects surfacing FsServiceError (drop console.error
  swallow + tryPromise wrappers); callers catchAll/logError so a write
  failure stays non-fatal, matching writeAndOpenTestReport
- affectedClasses derivation → functional map/flatMap/filter + effect/HashSet
  (clearStaleTags, restoreResultsBody); mutation sweep stays a forEach
- executeTests + runTestPipeline 8 positional params → options objects

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* perf(review-plan): speed up plan-adversary via warm-start prompt + evidence budget - W-23355256 (#7696)

* chore: plan for W-23355256

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* perf(review-plan): prime plan-adversary with plan-referenced code pointers

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* perf(plan-adversary): cap redundant evidence greps

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

* refactor: apply review findings — drop prime agent + diff machinery

Fold plan-adversary warm-start into a prompt directive; remove the
redundant haiku prime pass, PRIME_SCHEMA, `base` arg, and diff/truncation
plumbing (the adversary sees plan citations on its own first Read; the
pre-Build diff is always empty and pointers to to-be-created files were
dead Reads). Tighten plan doc to concise style.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* ci: change closePendingReleaseIssues.yml to grab issues from the changelog instead of checking the "pending release" tag - W-23236874 (#7704)

* refactor(org): add ConfigService.getTargetOrg, migrate org-ext off OrgConfigProperties - W-23354710 (#7695)

* chore: plan for W-23354710

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* feat(services): add ConfigService.getTargetOrg - W-23354710

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor(org): migrate org-ext off OrgConfigProperties to ConfigService.getTargetOrg - W-23354710

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix: add getTargetOrg to ConfigService mocks to fix test compile break - W-23354710

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor: address review findings - dedupe config reads, add coverage - W-23354710

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix: add getTargetOrg to org-browser ConfigService mock - W-23354710

Interface now requires getTargetOrg since Phase 1 added it to ConfigService.

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* chore(deps): bump json-stream-stringify from 3.1.6 to 3.1.7 (#7700)

Bumps [json-stream-stringify](https://github.com/Faleij/json-stream-stringify) from 3.1.6 to 3.1.7.
- [Commits](Faleij/json-stream-stringify@v3.1.6...v3.1.7)

---
updated-dependencies:
- dependency-name: json-stream-stringify
  dependency-version: 3.1.7
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* chore(deps): bump @salesforce/types from 1.7.1 to 1.8.0 (#7699)

Bumps [@salesforce/types](https://github.com/forcedotcom/wsdl) from 1.7.1 to 1.8.0.
- [Commits](https://github.com/forcedotcom/wsdl/commits)

---
updated-dependencies:
- dependency-name: "@salesforce/types"
  dependency-version: 1.8.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* chore(deps): bump @vscode/test-web from 0.0.78 to 0.0.81 (#7698)

Bumps [@vscode/test-web](https://github.com/microsoft/vscode-test-web) from 0.0.78 to 0.0.81.
- [Release notes](https://github.com/microsoft/vscode-test-web/releases)
- [Changelog](https://github.com/microsoft/vscode-test-web/blob/main/CHANGELOG.md)
- [Commits](microsoft/vscode-test-web@v0.0.78...v0.0.81)

---
updated-dependencies:
- dependency-name: "@vscode/test-web"
  dependency-version: 0.0.81
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* refactor(org): migrate sf.org.logout.default to services - W-23069609 (#7689)

* chore: plan for W-23069609

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* refactor(org): migrate sf.org.logout.default to services - W-23069609

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* test(org): cover orgLogoutDefaultCommand effect - W-23069609

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* test(org): e2e default-org logout removes auth - W-23069609

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix: review findings for org logout default - W-23069609

- no-default-org branch uses vscode.window.showInformationMessage (void)
  instead of legacy notificationService
- extract shared createAuthRemover helper; both logout commands delegate
- inline checkIsCurrentTargetOrg/clearTargetOrgRef single-call wrappers
- fix stale removeAuth doc comment

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* docs(plan): apply concise style to W-23069609 plan

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* feat(org): gate orgDisplay sensitive-info behind modal confirm - W-23230633 (#7690)

* chore: plan for W-23230633

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* feat(org): gate orgDisplay sensitive-info behind modal confirm - W-23230633

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* test(org): drive orgDisplay modal confirm in e2e - W-23230633

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* test(org): reuse shared clickModalDialogButton in e2e specs

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* docs(org): fix stale orgDisplay modal-gate comments and plan numeral

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* fix(org): split orgDisplay warning into message + detail for line breaks - W-23230633

VSCode modal collapses newlines in the message param; only the detail param
preserves line breaks. Add optional detail to PromptService.confirmOrThrow and
split the sensitive-info warning into a short headline plus a multi-paragraph
detail.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* docs(apex): add ADR to migrate telemetry to Effect spans/logs - W-23348766 (#7694)

* chore: plan for W-23348766

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* docs(apex): add ADR to migrate telemetry to Effect spans/logs - W-23348766

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor: fix review findings (citation line range, plan wording)

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* docs(apex): trim ADR-0002 to decision-record altitude - W-23348766

Cut per-event call-site line numbers, allowlist internals, and downstream
attribute-key constraints (all downstream-WI material) per ADR-FORMAT.md.
Kept decision, spans-only rationale, annotateRootSpan export rule, and the
load-bearing apexLSPLog span-vs-attribute choice.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>

* perf(apex-oas): shrink bundle via esbuild ESM resolution of effect - W-23313204 (#7707)

* chore: plan for W-23313204

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* perf(apex-oas): resolve effect ESM to shrink bundle - W-23313204

* fix: apply medium/low review findings for apex-oas bundle

- add dist/*-metafile.json to .vscodeignore (exclude 1MB debug metafile from VSIX)
- add ../../scripts/bundling/effect.mjs to wireit vscode:bundle files (cache invalidation)
- trim repeated caveats and multi-sentence bullets in plan doc

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* perf(apex-log): shrink bundle via esbuild ESM resolution of effect - W-23313203 (#7703)

* chore: plan for W-23313203

Co-Authored-By: Claude Haiku 4.5 <noreply@anthropic.com>

* perf(apex-log): resolve effect ESM to shrink bundle - W-23313203

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor: apply review findings (wireit files, concise plan)

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Haiku 4.5 <noreply@anthropic.com>

* perf(apex): resolve effect ESM to shrink bundle - W-23313207

* fix: review findings - add effect.mjs to apex vscode:bundle wireit files

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* refactor: review findings - dedupe vscode-uri blast-radius prose in plan

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

---------

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
Co-authored-by: Daphne Yang <139700604+daphne-sfdc@users.noreply.github.com>
Co-authored-by: Release Bot <svc_idee_bot@salesforce.com>
Co-authored-by: Jonny Hork <jhork@salesforce.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
@mshanemc mshanemc mentioned this pull request Jul 9, 2026
Merged
6 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@peternhale peternhale peternhale approved these changes
@kylewalke kylewalke Awaiting requested review from kylewalke kylewalke is a code owner automatically assigned from forcedotcom/ide-experience

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@mshanemc @peternhale