test(layout): move adapters, responses, lab, gui into tests/<domain>/ (#3497)

.github/CODEOWNERS

@@ -37,7 +37,7 @@
 # These four files carry every user's request path, including users of no optional
 # subsystem. Optional subsystems (Compatibility Lab and anything added later) register
 # into core-owned slots at activation instead of being imported here; the invariant is
-# enforced by tests/core-lab-boundary.test.ts and designed in
+# enforced by tests/lab/core-lab-boundary.test.ts and designed in
 # devlog/_fin/260814_lab_core_decoupling/.
 # Last-match-wins: this block must stay below /src/server/ to take effect.
 /src/router.ts @lidge-jun

AGENTS.md

@@ -51,7 +51,7 @@ directly or transitively:
 - `src/server/lifecycle.ts`
 - `src/server/responses/core.ts`
 
-`tests/core-lab-boundary.test.ts` enforces this by walking the runtime import
+`tests/lab/core-lab-boundary.test.ts` enforces this by walking the runtime import
 graph and printing the offending chain on failure. It is not a style rule: the
 original violation hid in a six-hop chain
 (`assemble → quota → auth-api → native-main-admission → lifecycle → lab`) where

docs/superpowers/plans/2026-08-06-deepseek-responses-streaming-terminal-repair.md

@@ -26,9 +26,9 @@
 - `src/providers/registry.ts` — declares and resolves the built-in per-model terminal-repair policy.
 - `src/server/responses-terminal-repair.ts` — owns SSE lifecycle tracking, bounded retained state, grace scheduling, and synthetic terminal creation.
 - `src/server/responses/core.ts` — activates the repair before existing transport-specific relay branches.
-- `tests/responses-terminal-repair.test.ts` — unit state-machine and stream-race coverage.
+- `tests/responses/responses-terminal-repair.test.ts` — unit state-machine and stream-race coverage.
 - `tests/deepseek-inbound-wire.test.ts` — end-to-end official DeepSeek wire and HTTP activation.
-- `tests/ws-endpoint.test.ts` — WebSocket event parity for real and repaired terminals.
+- `tests/responses/ws-endpoint.test.ts` — WebSocket event parity for real and repaired terminals.
 - `structure/04_transports-and-sidecars.md` — architectural contract for the provider-scoped streaming repair.
 - `docs/superpowers/specs/2026-08-06-deepseek-responses-streaming-terminal-repair-design.md` — approved design authority; implementation must remain consistent with it.
 
@@ -153,7 +153,7 @@ git commit -m "fix(deepseek): restore Responses upstream streaming"
 
 **Files:**
 - Create: `src/server/responses-terminal-repair.ts`
-- Create: `tests/responses-terminal-repair.test.ts`
+- Create: `tests/responses/responses-terminal-repair.test.ts`
 
 **Interfaces:**
 - Consumes: `ResponsesTerminalRepairPolicy`
@@ -201,7 +201,7 @@ terminal followed by one `[DONE]`.
 Run:
 
 ```bash
-bun test tests/responses-terminal-repair.test.ts
+bun test tests/responses/responses-terminal-repair.test.ts
 ```
 
 Expected: module-not-found failure for
@@ -251,7 +251,7 @@ an item and release every remaining charge during disposal.
 Run:
 
 ```bash
-bun test tests/responses-terminal-repair.test.ts
+bun test tests/responses/responses-terminal-repair.test.ts
 ```
 
 Expected: healthy pass-through and five-second grace completion both pass;
@@ -260,7 +260,7 @@ translator-budget current bytes return to zero after drain.
 - [ ] **Step 5: Commit**
 
 ```bash
-git add src/server/responses-terminal-repair.ts tests/responses-terminal-repair.test.ts
+git add src/server/responses-terminal-repair.ts tests/responses/responses-terminal-repair.test.ts
 git commit -m "feat(responses): repair complete terminal-less streams"
 ```
 
@@ -270,7 +270,7 @@ git commit -m "feat(responses): repair complete terminal-less streams"
 
 **Files:**
 - Modify: `src/server/responses-terminal-repair.ts`
-- Modify: `tests/responses-terminal-repair.test.ts`
+- Modify: `tests/responses/responses-terminal-repair.test.ts`
 
 **Interfaces:**
 - Preserves Task 2 public signatures.
@@ -305,7 +305,7 @@ cancellation, an empty scheduler queue, and
 Run:
 
 ```bash
-bun test tests/responses-terminal-repair.test.ts
+bun test tests/responses/responses-terminal-repair.test.ts
 ```
 
 Expected: the newly added boundary tests fail because Task 2 implements only
@@ -366,7 +366,7 @@ abort/cancel, taint, and candidate completeness immediately before enqueueing.
 Run:
 
 ```bash
-bun test tests/responses-terminal-repair.test.ts tests/sse-failed-tail.test.ts tests/relay-eager.test.ts
+bun test tests/responses/responses-terminal-repair.test.ts tests/responses/sse-failed-tail.test.ts tests/relay-eager.test.ts
 ```
 
 Expected: all tests pass with no duplicate terminal, timer leak, or retained
@@ -375,7 +375,7 @@ budget after teardown.
 - [ ] **Step 5: Commit**
 
 ```bash
-git add src/server/responses-terminal-repair.ts tests/responses-terminal-repair.test.ts
+git add src/server/responses-terminal-repair.ts tests/responses/responses-terminal-repair.test.ts
 git commit -m "fix(responses): fail closed on unsafe terminal repair"
 ```
 
@@ -388,7 +388,7 @@ git commit -m "fix(responses): fail closed on unsafe terminal repair"
 - Modify: `src/server/responses/core.ts:2030-2230`
 - Modify: `tests/deepseek-inbound-wire.test.ts:115-330`
 - Modify: `tests/deepseek-responses-item-id-repair.test.ts`
-- Modify: `tests/ws-endpoint.test.ts`
+- Modify: `tests/responses/ws-endpoint.test.ts`
 
 **Interfaces:**
 - Consumes: `providerModelResponsesTerminalRepair(...)`
@@ -432,7 +432,7 @@ WebSocket handling. Assert the client receives progressive delta frames, one
 Run:
 
 ```bash
-bun test tests/deepseek-inbound-wire.test.ts tests/ws-endpoint.test.ts
+bun test tests/deepseek-inbound-wire.test.ts tests/responses/ws-endpoint.test.ts
 ```
 
 Expected: requests now carry `stream:true` from Task 1 but no provider-scoped
@@ -482,7 +482,7 @@ streaming resolver still returns `false`.
 Run:
 
 ```bash
-bun test tests/deepseek-inbound-wire.test.ts tests/ws-endpoint.test.ts tests/responses-state.test.ts tests/deepseek-responses-item-id-repair.test.ts tests/deepseek-reasoning-replay.test.ts
+bun test tests/deepseek-inbound-wire.test.ts tests/responses/ws-endpoint.test.ts tests/responses/responses-state.test.ts tests/deepseek-responses-item-id-repair.test.ts tests/deepseek-reasoning-replay.test.ts
 ```
 
 Expected: progressive output precedes terminal, both transports close once,
@@ -491,7 +491,7 @@ item ids are stable, and continuation state retains reasoning/function output.
 - [ ] **Step 6: Commit**
 
 ```bash
-git add src/server/responses/core.ts tests/deepseek-inbound-wire.test.ts tests/deepseek-responses-item-id-repair.test.ts tests/ws-endpoint.test.ts
+git add src/server/responses/core.ts tests/deepseek-inbound-wire.test.ts tests/deepseek-responses-item-id-repair.test.ts tests/responses/ws-endpoint.test.ts
 git commit -m "fix(deepseek): repair terminal-less Responses streams"
 ```
 
@@ -560,7 +560,7 @@ git commit -m "docs(deepseek): describe streaming terminal repair"
 Run:
 
 ```bash
-bun test tests/responses-terminal-repair.test.ts tests/deepseek-inbound-wire.test.ts tests/ws-endpoint.test.ts tests/sse-failed-tail.test.ts tests/relay-eager.test.ts tests/responses-item-id-repair.test.ts tests/deepseek-responses-item-id-repair.test.ts tests/deepseek-reasoning-replay.test.ts tests/responses-state.test.ts
+bun test tests/responses/responses-terminal-repair.test.ts tests/deepseek-inbound-wire.test.ts tests/responses/ws-endpoint.test.ts tests/responses/sse-failed-tail.test.ts tests/relay-eager.test.ts tests/responses/responses-item-id-repair.test.ts tests/deepseek-responses-item-id-repair.test.ts tests/deepseek-reasoning-replay.test.ts tests/responses/responses-state.test.ts
 ```
 
 Expected: 0 failures.

docs/superpowers/plans/2026-08-09-routed-computer-use-browser.md

@@ -229,7 +229,7 @@ git commit -m "docs(codex): explain routed local tool access"
 Run:
 
 ```bash
-bun test tests/responses-parser.test.ts tests/bridge.test.ts tests/multi-agent-compat.test.ts
+bun test tests/responses/responses-parser.test.ts tests/adapters/bridge.test.ts tests/multi-agent-compat.test.ts
 ```
 
 Expected: custom declaration, streaming `custom_tool_call`, exact freeform input, and output replay remain green. Add no production protocol change unless one of these tests proves a real gap.
@@ -239,7 +239,7 @@ Expected: custom declaration, streaming `custom_tool_call`, exact freeform input
 Run:
 
 ```bash
-bun test tests/responses-terminal-repair.test.ts tests/deepseek-inbound-wire.test.ts tests/deepseek-responses-item-id-repair.test.ts tests/passthrough-abort.test.ts
+bun test tests/responses/responses-terminal-repair.test.ts tests/deepseek-inbound-wire.test.ts tests/deepseek-responses-item-id-repair.test.ts tests/responses/passthrough-abort.test.ts
 ```
 
 Expected: progressive deltas, strict terminal repair, item IDs, and cancellation pass.
@@ -249,7 +249,7 @@ Expected: progressive deltas, strict terminal repair, item IDs, and cancellation
 Run:
 
 ```bash
-bun test tests/vision/vision-sidecar-e2e.test.ts tests/vision/vision-anthropic.test.ts tests/vision/vision-cache.test.ts tests/vision/vision-fail-closed.test.ts tests/catalog-vision-sidecar-modalities.test.ts tests/openai-responses-passthrough.test.ts
+bun test tests/vision/vision-sidecar-e2e.test.ts tests/vision/vision-anthropic.test.ts tests/vision/vision-cache.test.ts tests/vision/vision-fail-closed.test.ts tests/catalog-vision-sidecar-modalities.test.ts tests/responses/openai-responses-passthrough.test.ts
 ```
 
 Expected: captions replace raw image parts in passthrough bodies, empty references do not consume captions, and partial/failure paths omit pixels safely.
@@ -259,7 +259,7 @@ Expected: captions replace raw image parts in passthrough bodies, empty referenc
 Run:
 
 ```bash
-bun test tests/images/plan.test.ts tests/images/synthetic-tool.test.ts tests/images/z-handler-activation.test.ts tests/images/loop-reasoning-replay.test.ts tests/responses-image-gen-repair.test.ts
+bun test tests/images/plan.test.ts tests/images/synthetic-tool.test.ts tests/images/z-handler-activation.test.ts tests/images/loop-reasoning-replay.test.ts tests/responses/responses-image-gen-repair.test.ts
 ```
 
 Expected: image tool planning, activation, alias restoration, replay, and result repair pass.

docs/superpowers/plans/2026-08-12-cl10-public-evidence-implementation.md

@@ -49,14 +49,14 @@
 - Create: `src/lab/public/project.ts`
 - Create: `src/lab/public/index.ts`
 - Modify: `src/lab/index.ts`
-- Test: `tests/lab-public-evidence.test.ts`
+- Test: `tests/lab/lab-public-evidence.test.ts`
 
 **Interfaces:**
 - Produces: `PublicEvidenceBundleUnsignedV1`, `PublicEvidenceRecordV1`, `PublicEvidenceSubjectV1`, `PublicRouteRegistryManifestV1`, `projectPublicEvidence()`, `validatePublicEvidenceBundle()`.
 
 - [ ] **Step 1: Write RED tests** for closed-schema rejection, deterministic public IDs/day buckets, protocol/route/task subject discrimination, exact route allowlist, private-route `not_exportable`, IC-only incident refs, no local ID leakage, and secret/PII canaries.
 - [ ] **Step 2: Run focused test and verify expected RED failures.**
-  Run: `bun test tests/lab-public-evidence.test.ts`
+  Run: `bun test tests/lab/lab-public-evidence.test.ts`
 - [ ] **Step 3: Implement minimal closed public types/registry/validator/projector.**
   Public identities use domain-separated SHA-256 over JCS public-safe bytes. The registry manifest is repo-owned, versioned, digested, and cannot be supplied by an imported bundle as trust authority.
 - [ ] **Step 4: Run focused test and verify GREEN.**
@@ -67,7 +67,7 @@
 - Create: `src/lab/public/signature.ts`
 - Create: `src/lab/public/storage.ts`
 - Modify: `src/lab/paths.ts`
-- Test: `tests/lab-public-evidence.test.ts`
+- Test: `tests/lab/lab-public-evidence.test.ts`
 
 **Interfaces:**
 - Produces: `getOrCreatePublicPublisher()`, `signPublicEvidenceBundle()`, `verifyPublicEvidenceBundle()`, `writePublicEvidenceBundle()`, `readPublicEvidenceBundle()`.
@@ -82,7 +82,7 @@
 **Files:**
 - Create: `src/lab/public/revocation.ts`
 - Create: `src/lab/public/community.ts`
-- Test: `tests/lab-public-evidence.test.ts`
+- Test: `tests/lab/lab-public-evidence.test.ts`
 
 **Interfaces:**
 - Produces: `PublicEvidenceRevocationV1`, `verifyPublicEvidenceRevocation()`, `importCommunityBundle()`, `listCommunityBundles()`.
@@ -97,8 +97,8 @@
 **Files:**
 - Modify: `src/lab/ledger/purge.ts`
 - Modify: `src/lab/paths.ts`
-- Test: `tests/lab-public-evidence.test.ts`
-- Test: `tests/lab-evidence-ledger.test.ts`
+- Test: `tests/lab/lab-public-evidence.test.ts`
+- Test: `tests/lab/lab-evidence-ledger.test.ts`
 
 **Interfaces:**
 - Consumes: existing `purgeSensitiveEvidence()` and `purgeActions: export`.
@@ -114,7 +114,7 @@
 **Files:**
 - Modify: `src/cli/lab.ts`
 - Modify: `src/server/management/lab-routes.ts`
-- Test: `tests/lab-public-evidence.test.ts`
+- Test: `tests/lab/lab-public-evidence.test.ts`
 - Test: relevant Lab CLI/management tests discovered in repository.
 
 **Interfaces:**
@@ -147,7 +147,7 @@
 **Files:**
 - Modify docs only if validation findings require factual updates.
 
-- [ ] **Step 1:** Run `bun test tests/lab-public-evidence.test.ts tests/lab-evidence-ledger.test.ts`.
+- [ ] **Step 1:** Run `bun test tests/lab/lab-public-evidence.test.ts tests/lab/lab-evidence-ledger.test.ts`.
 - [ ] **Step 2:** Run `bun x tsc --noEmit`.
 - [ ] **Step 3:** Run `bun run privacy:scan`.
 - [ ] **Step 4:** Run relevant Lab query/ledger/CLI/GUI tests.

docs/superpowers/plans/2026-08-13-cl10-deep-review-hardening.md

@@ -26,7 +26,7 @@
 ### Task 1: Add adversarial RED coverage
 
 **Files:**
-- Create: `tests/lab-public-deep-review-regressions.test.ts`
+- Create: `tests/lab/lab-public-deep-review-regressions.test.ts`
 - Modify: `tests/ci-workflows.test.ts`
 
 **Interfaces:**
@@ -135,7 +135,7 @@
 - Modify: `src/lab/public/operator.ts`
 - Modify: `src/cli/lab.ts`
 - Modify: `src/server/management/lab-routes.ts`
-- Modify: `tests/lab-public-surfaces.test.ts`
+- Modify: `tests/lab/lab-public-surfaces.test.ts`
 
 **Interfaces:**
 - Produces: preview/export results that expose exclusion indices/reasons and `stored.created` only, never local event IDs or filesystem paths.

docs/superpowers/specs/2026-08-06-deepseek-responses-streaming-terminal-repair-design.md

@@ -227,7 +227,7 @@ The implementation is expected to touch these responsibility boundaries:
   wrapper.
 - `src/server/responses/core.ts`: resolve the provider policy and wrap the SSE
   body before transport-specific relay branches.
-- `tests/responses-terminal-repair.test.ts`: unit state-machine coverage.
+- `tests/responses/responses-terminal-repair.test.ts`: unit state-machine coverage.
 - `tests/deepseek-inbound-wire.test.ts`: end-to-end wire, progressive delivery,
   repair composition, and WebSocket/HTTP activation.
 - Existing relay and item-id tests only where an explicit integration contract

gui/src/pages/claude-manual-env.ts

@@ -1,7 +1,7 @@
 /**
  * Pure manual-env builder for the Claude Code page (devlog
  * 260720_claude_authmode_persist/020): extracted from ClaudeCode.tsx so the
- * copy-paste shell block is directly unit-testable (tests/claude-manual-env.test.ts).
+ * copy-paste shell block is directly unit-testable (tests/gui/claude-manual-env.test.ts).
  */
 import { AUTO_COMPACT_WINDOW_DEFAULT } from "./claude-code-types";
 

gui/src/pages/dashboard-shared.ts

@@ -241,7 +241,7 @@ export function visionTimeoutPatch(timeoutMs: number): SidecarPatch {
 
 /**
  * Dashboard names for the runtime timeout contract in `src/vision/timeout-bounds.ts`.
- * Pinned by `tests/vision-sidecar-timeout-bounds.test.ts`.
+ * Pinned by `tests/gui/vision-sidecar-timeout-bounds.test.ts`.
  */
 export const VISION_TIMEOUT_MS_DEFAULT = DEFAULT_VISION_TIMEOUT_MS;
 export const VISION_TIMEOUT_MS_MAX = MAX_VISION_TIMEOUT_MS;

gui/src/pages/integrations/integration-tabs.ts

@@ -4,7 +4,7 @@
  * A separate module rather than exports on Integrations.tsx, because a file that
  * exports both a component and constants breaks React fast refresh
  * (react/only-export-components). These need to be importable: they are the only
- * client lists in the GUI that neither tests/integrations-invariants.test.ts
+ * client lists in the GUI that neither tests/gui/integrations-invariants.test.ts
  * compares nor the compiler forces, so a client added everywhere else still gets
  * no tab and nothing fails. gui/tests/integrations-tab-coverage.test.ts stands in
  * that gap and reads them from here.

gui/tests/integrations-tab-coverage.test.ts

@@ -6,7 +6,7 @@ import { INTEGRATION_TAB_HASHES } from "../src/app-routing";
 /*
  * The gap this closes.
  *
- * tests/integrations-invariants.test.ts compares five client lists, and the
+ * tests/gui/integrations-invariants.test.ts compares five client lists, and the
  * per-page label maps are Record<FileIntegrationClientId, TKey> so the compiler
  * forces those. TABS and FILE_CLIENTS are neither: they are a plain array and a
  * plain Set, so a client added everywhere else still gets no tab and nothing

gui/tests/models-workspace-panels.test.tsx

@@ -1,7 +1,7 @@
 /**
  * Models tab workspace — mounted behaviour.
  *
- * The routing helpers are unit-tested at `tests/models-workspace-tabs.test.ts`. This file
+ * The routing helpers are unit-tested at `tests/gui/models-workspace-tabs.test.ts`. This file
  * exists because those assertions cannot see the failures that actually happened here:
  * a component-level early return that unmounted the whole tab tree while the catalog
  * loaded, and a disabled resource that swapped the combo editor for an empty state and

scripts/openai-provider-option-final-gates.ts

@@ -42,14 +42,14 @@ export async function runGateSequence(plan: GateSpec[], deps: GateDeps): Promise
 }
 
 const focusedTests = [
-  "tests/openai-provider-option.test.ts",
-  "tests/openai-provider-option-migration.test.ts",
-  "tests/openai-provider-option-startup.test.ts",
-  "tests/openai-provider-option-e2e.test.ts",
-  "tests/openai-provider-option-tooling.test.ts",
+  "tests/adapters/openai/openai-provider-option.test.ts",
+  "tests/adapters/openai/openai-provider-option-migration.test.ts",
+  "tests/adapters/openai/openai-provider-option-startup.test.ts",
+  "tests/adapters/openai/openai-provider-option-e2e.test.ts",
+  "tests/adapters/openai/openai-provider-option-tooling.test.ts",
   "tests/provider-registry-parity.test.ts",
-  "tests/provider-payload.test.ts",
-  "tests/codex-account-mode-state.test.ts",
+  "tests/gui/provider-payload.test.ts",
+  "tests/gui/codex-account-mode-state.test.ts",
   "tests/routing/router.test.ts",
   "tests/codex-routing.test.ts",
   "tests/server-auth.test.ts",
@@ -70,7 +70,7 @@ export function finalGatePlan(root: string, evidenceDir: string, unitRoot = dirn
   const env = { ...process.env, OCX_EVIDENCE_DIR: evidenceDir } as Record<string, string>;
   const unitPath = relative(root, unitRoot);
   return [
-    { name: "openai-provider-option-e2e", command: ["bun", "test", "tests/openai-provider-option-e2e.test.ts"], cwd: root, env },
+    { name: "openai-provider-option-e2e", command: ["bun", "test", "tests/adapters/openai/openai-provider-option-e2e.test.ts"], cwd: root, env },
     { name: "provider-option-focused", command: ["bun", "test", "--isolate", ...focusedTests], cwd: root, env },
     { name: "isolated-runtime-smoke", command: ["bun", "scripts/openai-provider-option-runtime-smoke.ts", "--unit-root", unitPath, "--evidence-dir", evidenceDir], cwd: root },
     { name: "live-key-status", command: ["bun", "scripts/openai-provider-option-runtime-smoke.ts", "--check-live-key", "--unit-root", unitPath, "--evidence-dir", evidenceDir], cwd: root },
@@ -91,8 +91,8 @@ export function finalGatePlan(root: string, evidenceDir: string, unitRoot = dirn
     {
       name: "scoped-diff-check",
       command: ["git", "diff", "--check", "--", "README.md", "readme/README.ko.md", "readme/README.zh-CN.md", "structure",
-        "docs-site/src/content/docs", "devlog/_chase/_model", "tests/openai-provider-option-e2e.test.ts",
-        "tests/openai-provider-option-tooling.test.ts", "tests/fixtures/openai-provider-option-migration-child.ts",
+        "docs-site/src/content/docs", "devlog/_chase/_model", "tests/adapters/openai/openai-provider-option-e2e.test.ts",
+        "tests/adapters/openai/openai-provider-option-tooling.test.ts", "tests/fixtures/openai-provider-option-migration-child.ts",
         "scripts/openai-provider-option-runtime-child.ts", "scripts/openai-provider-option-runtime-smoke.ts",
         "scripts/openai-provider-option-evidence-scan.ts", "scripts/openai-provider-option-final-gates.ts",
         "scripts/openai-hardening-live-policy.ts", "scripts/openai-hardening-runtime-env.ts",