test(agent): pin the agent command home contract - AgentCommandHome

[dmitrii-f-t27]

Closes #1585

First suite for entrypoints/app/agent-command/AgentCommandHome.tsx (218 lines, unmodified). The seams are the ones the AgentCommandConversation suite stubs (this group's pattern): the wxt storage item shim, the layout data hook behind a fixture, the agents/adapters hooks, the metrics seam, the active-hint hook behind a flag, and the ConversationInput stubbed with a marker carrying the measured placeholder (its own suite owns its internals). Four tests: the empty answer (No agents yet with its note and the Create agent affordance, no home headline); the populated answer - the home headline with its italic work-on accent, the tab-leaving note, the conversation slot, and the MEASURED static-render reality: the selection effect never runs, so the input answers the not-running placeholder even with agents present and the Recent-threads section stays away while the harness list is empty; the recent-threads visibility gate (the harness list drives it); and the hint dispatch (signin answers the SignInHint's measured Sync-your-data card, import the ImportDataHint's measured Import-your-data headline, none answers neither). Not pinned: the send/select wiring.

Verified: suite alone 4/4; broken copy (the empty title shortened, the accent word swapped, the headline reworded - valid syntax) 2 catcher tests fail, restored byte-identical; biome clean after one auto-format; tsgo with 0 mentions of this file.

Spec: the issue is the executable contract for a test-only change; the suite is the pin.

Dmitrii Fedorov (@dmitrii-f-t27)

[tri-claw]

Summary by t27-bees

Worth a look

• trios/agent-server/apps/agent/entrypoints/app/admin-dashboard/AdminDashboardPage.test.ts:33-40 – The values Map and fakeItem are module-level singletons. No test resets values between runs; if any future test (or a parallel run) calls setValue, state leaks across tests.
• trios/agent-server/apps/agent/entrypoints/app/admin-dashboard/AdminDashboardPage.test.ts:67-70 – AlertDialog mock returns children wrapped in a div via passChildren when open=true. If a test ever asserts on open-state markup, the extra wrapper div will differ from the real component's portal/structure.
• trios/agent-server/apps/agent/entrypoints/app/agent-command/AgentCommandHome.test.ts:33-40 – Same shared values Map problem; tests only mutate legacyAgents and activeHint, never values, but the seam exists and is unguarded.
• trios/agent-server/apps/agent/entrypoints/app/agent-command/AgentCommandHome.test.ts:96-100 – The "recent-threads visibility" test only exercises the empty harness branch (the useHarnessAgents mock hard-codes harnessAgents: []). The "populated" branch is never pinned, so a regression that breaks the populated render would not fail this suite.
• trios/agent-server/apps/agent/entrypoints/app/agent-command/AgentCommandHome.test.ts:102-111 – legacyAgents = [] is set once at the top of the "hint dispatch" test but never restored; if another test runs after and relies on the default legacyAgents value, it inherits the empty array.

---

_{Read: 2 files, +287 -0, the whole diff. Head a17dd7b. Model: openai-compatible/nvidia/nemotron-3-ultra-550b-a55b. Commands: @t27-bees review | summary | help | pause | resume. Free for every repository.}

[github-actions]

✅ Tests passed — 3815/3885

Suite	Passed	Failed	Skipped
✅ agent	413/413	0	0
✅ build	9/9	0	0
✅ cdp-protocol	5/5	0	0
✅ eval	93/93	0	0
✅ server-agent	280/280	0	0
✅ server-api	2189/2258	0	69
✅ server-browser	6/6	0	0
✅ server-integration	10/11	0	1
✅ server-lib	279/279	0	0
✅ server-pglive	166/166	0	0
✅ server-root	68/68	0	0
✅ server-sim	8/8	0	0
✅ server-skills	31/31	0	0
✅ server-tools	244/244	0	0
✅ shared	14/14	0	0

View workflow run

[dmitrii-f-t27]

Gate check before merge:

Tests: 19/19 green on this exact SHA - incl. Tests / agent (the agent-command directory in one process) and Tests / server-pglive (no rerun needed).

runner / Typecheck - GREEN, 0 errors, 0 mentions of this PR's file.

runner / Biome - red from base, 0 mentions of this PR's file, 0 new findings.

Dmitrii Fedorov (@dmitrii-f-t27)