First-run lands in Myra's DM; mix DMs and channels in the sidebar

DESIGN.md

@@ -16,11 +16,12 @@ Top to bottom:
 
 1. **Brand row** — logo mark and a "New workbench" button (`+`) that
    opens `/new`.
-2. **Bench list** — **Agents**, then **Channels**. Agent rows are
-   `kind: chat` DMs (one per agent). Channel rows are `kind: workbench`
-   rooms. Search is built into the list itself. Nothing page-scoped
-   ever renders in this body; it lists conversations, not product
-   sections.
+2. **Bench list** — one recency list of conversations, pins first.
+   Agent DMs (`kind: chat`, one per agent) mix with channels
+   (`kind: workbench` rooms). Not two labeled empty sections, and not
+   a list titled Workbenches. Search is built into the list itself.
+   Nothing page-scoped ever renders in this body; it lists
+   conversations, not product sections.
 3. **Footer rail** — Mission Control is pinned above the rail as its own
    row. The first-run rail below it is Routines, Files, Skills, Agents, in
    that order (CL-6765). Insights and Evals join that rail only when
@@ -60,13 +61,15 @@ consumed by both). A route that gets renamed or relocated leaves a redirect
 behind at its old path — old links and bookmarks always land somewhere
 real, never a 404.
 
-**`/new` is the shipped create surface** (`NewWorkbenchPickerRoute`). The
-primary act is a prompt: say what the channel should do, or pick a
-named-template shortcut underneath. Blank `+` / prompt mint an empty
-channel and invite nobody. Named templates mint that same empty
-channel, then invite existing agents (including Myra as a participant,
-never as the mint host). The sidebar `+` and an empty-bench land on
-`/` both open this route — there is no Describe door.
+**`/new` is the shipped create surface** (`NewWorkbenchPickerRoute`),
+not home. The primary act is a prompt: say what the channel should do,
+or pick a named-template shortcut underneath. Blank `+` / prompt mint
+an empty channel and invite nobody. Named templates mint that same
+empty channel, then invite existing agents (including Myra as a
+participant, never as the mint host). The sidebar `+` opens this
+route. First-run after credential does not: `/` hops to Myra's one
+DM (`openAgentDm` / find-or-reopen). There is no parallel Myra home
+route and no Describe door.
 
 **`/inbox` is gone as a page** (CL-6151). The path stays as a redirect
 home so old links still resolve; it is not a live groups inbox.

PRODUCT.md

@@ -15,8 +15,10 @@ opens a multi-principal room (`kind: workbench`). There is no separate
 "project" or "space" object sitting above the conversation; the
 conversation is the unit of work.
 
-This is why the sidebar is Agents then Channels, not one recency list
-titled Workbenches. The same Interchange agent can sit in its DM and in
+This is why the sidebar is one recency list of conversations (pins
+first) — agent DMs mixed with channels — not two labeled sections, and
+not a list titled Workbenches. That noun is the product name and the
+mint verb. The same Interchange agent can sit in its DM and in
 many channels; product reopens or invites, and does not clone the
 definition or mint a sibling instance per room. See
 [docs/GLOSSARY.md](docs/GLOSSARY.md) for the full term mapping and
@@ -34,43 +36,48 @@ team spaces); within a bench, they open, create, and work in workbenches.
 Workbench is intentionally not a multi-pane IDE. The product surface is one
 column at a time:
 
-- **A sidebar of Agents then Channels** lists conversations in the
-  selected bench. Agent rows are DMs (`kind: chat`); channel rows are
-  rooms (`kind: workbench`). There is no one recency list titled
+- **A sidebar of conversations** lists agent DMs and channels together
+  by recency, pins first, in the selected bench. Agent rows are DMs
+  (`kind: chat`); channel rows are shared rooms (`kind: workbench`).
+  There are not two labeled empty sections, and the list is not titled
   Workbenches.
 - **"New workbench" always creates.** It opens `/new` — the shipped
-  prompt-primary picker — never a picker of existing things to join.
-  Plus mints an empty channel. Nobody is auto-hosted.
+  prompt-primary picker — never home, and never a picker of existing
+  things to join. Plus mints an empty channel. Nobody is auto-hosted.
 - **Agents are principals, not templates.** Opening Sales opens Sales —
   the one 1:1 tenant with that agent. The same agent can sit in its DM
   and in many channels. Product reopens or invites; it does not clone
-  the definition or mint a sibling instance per room. Myra is an agent
-  row, not a special home slot.
+  the definition or mint a sibling instance per room. Myra is the
+  first-run guide in her DM, not a special home slot and not a
+  parallel home route.
 - The active workbench occupies the main column; a contextual panel beside
   it carries account-wide surfaces (approvals, recent activity) that stay
   visible regardless of which workbench is open.
 
 ## First run
 
 A brand-new account is walked through login and a credential connect, then
-lands on the create surface rather than an empty shell:
+lands in Myra's one DM rather than `/new` or an empty shell:
 
 1. **Login.**
 2. **Credential** — connect a model provider (one-click OAuth for
    supported providers, or a pasted API key); see `packages/onboarding`.
-3. **Create** — `/` hops an empty bench (zero workbenches) to `/new`, the
-   prompt-primary picker (`apps/web/src/pages/new-workbench-picker.tsx`).
-   A prompt box is the primary act: typing a goal and submitting mints an
-   empty channel and sends that text as the first message; blank plus
-   invites nobody. Named-template rows underneath mint that same empty
-   channel, then invite existing principals (including Myra as a
-   participant, never as mint `definitionId`) — one-click shortcuts, not
-   a kind-then-Create second step. There is no Describe door and no
-   `describe-first-workbench.tsx`.
-
-A bench that already has one or more workbenches skips create and lands
-on `workbenches[0]` (see `apps/web/src/pages/home-page.tsx`). Myra is an
-agent row, not a home slot.
+3. **Myra's DM** — `/` hops an empty bench to that agent's one DM
+   (`openAgentDm` / find-or-reopen). There is no parallel Myra home
+   route. `/new` stays the create door (sidebar `+`), not this hop.
+
+Create stays on `/new` (`apps/web/src/pages/new-workbench-picker.tsx`):
+a prompt box is the primary act: typing a goal and submitting mints an
+empty channel and sends that text as the first message; blank plus
+invites nobody. Named-template rows underneath mint that same empty
+channel, then invite existing principals (including Myra as a
+participant, never as mint `definitionId`) — one-click shortcuts, not
+a kind-then-Create second step. There is no Describe door and no
+`describe-first-workbench.tsx`.
+
+A bench that already has one or more workbenches skips first-run and
+lands on `workbenches[0]` (see `apps/web/src/pages/home-page.tsx`). Myra
+is the first-run guide in her DM, not a home slot.
 
 The shell's first-run destinations stay small on purpose (CL-6765):
 Mission Control is pinned above the footer rail; the rail itself is
@@ -158,13 +165,15 @@ as an explicit absence, never a fabricated zero. See `packages/insights`.
 User-facing surfaces (UI, docs, support) use exactly these nouns:
 
 - **Workbench** — the product name, and the mint verb ("New workbench").
-  A workbench is a conversation tenant: a DM or a channel.
-- **Agent** — a coworker identity in the sidebar's Agents list. Opening
-  the row reopens that agent's one DM. Never "template."
+  A workbench is a conversation tenant: a DM or a channel. Do not title
+  the sidebar list Workbenches.
+- **Agent** — a coworker principal. Opening the row reopens that agent's
+  one DM. Never "template." Myra is the first-run guide in her DM.
 - **DM** — the one 1:1 conversation with an agent. Never cloned by a
   second open.
-- **Channel** — a multi-party room. Plus mints an empty one; nobody is
-  auto-hosted. Named templates invite existing agents into that room.
+- **Channel** — a shared room between people and agents. Plus mints an
+  empty one; nobody is auto-hosted. Named templates invite existing
+  principals into that room.
 - **Bench** — the shared team scope a person signs into and switches
   between; shown in the bench switcher, never called a "workspace" or
   "org" in copy.

apps/web/src/pages/home-page.tsx

@@ -1,28 +1,24 @@
 // Default land: `/` is a hop onto the most-recent existing workbench or
-// `/new`, never a parallel Myra home. A bench that already has one or
+// Myra's one DM, never the picker. A bench that already has one or
 // more workbenches lands in the first listed row (`workbenches[0]`) in
-// the main stage. A brand-new bench with zero workbenches has nothing to
-// land in yet, so this hop sends it to the guided create surface
-// (`NewWorkbenchPickerRoute`, CL-6342) instead of auto-minting an
-// unlabeled "New Workbench" and dropping the person straight into it —
-// that auto-mint (CL-6138) is exactly the confusing empty-bench landing
-// this hop used to produce. Home as a dashboard does not earn its keep —
-// `/` only exists as this hop onto `/w/:workbenchId` or `/new`. Deep
-// links to other pages are unchanged.
+// the main stage. A brand-new bench with zero workbenches waits for
+// Myra's own definition to exist, then opens her DM the same way
+// "Talk to Myra" does (`openAgentDmChat`) — never `/new`, never
+// `ensureMyraWorkbench`. Home as a dashboard does not earn its keep —
+// `/` only exists as this hop onto `/w/:workbenchId`. Deep links to
+// other pages are unchanged.
 //
 // Right after a provider connect this hop is also the wait (CL-6457's
 // deploys run in the background, so landing here can beat them). CL-6462
 // settled what that wait looks like: one warm loader and nothing else. For
 // a zero-workbench bench the wait is for Myra's own definition to exist at
-// all — the picker's "Create workbench" needs it too, so checking here
-// first means the picker never opens onto a create button that would just
-// throw. The check is simply retried every few seconds, because Myra's
-// readiness IS the test of whether the person can start — she is deployed
-// first (`SETUP_AGENT_ASSET_NAME` leads `DEFAULT_WORKFLOWS`), so the moment
-// she's ready we go, with every other seeded workflow still converging
-// behind us. Readiness is read only to tell a wait from a genuine failure,
-// never to draw a progress number: a seed count is an implementation
-// detail, and "0 of 5" told a waiting person nothing.
+// all — then we open her DM. The check is simply retried every few seconds,
+// because Myra's readiness IS the test of whether the person can start —
+// she is deployed first (`SETUP_AGENT_ASSET_NAME` leads `DEFAULT_WORKFLOWS`),
+// so the moment she's ready we go, with every other seeded workflow still
+// converging behind us. Readiness is read only to tell a wait from a
+// genuine failure, never to draw a progress number: a seed count is an
+// implementation detail, and "0 of 5" told a waiting person nothing.
 //
 // CL-6780: that wait is for the agent, never a workbench that does not
 // exist yet — so the loader says "Preparing your agent", and a skip with
@@ -37,10 +33,13 @@ import { listAllWorkbenches, WorkbenchLoadingState } from "@corbits/chat-ui";
 import { describeApiError } from "@corbits/api-query";
 
 import { fetchAgentReadiness, hasActiveCredential } from "../onboarding";
+import { listAgentDefinitions } from "../agents-api";
+import { openAgentDmChat } from "../agent-dm-launch";
+import { findMyraDefinition } from "../myra-workbench";
 import { useBench } from "../bench-context";
 import { workbenchPath } from "../workbench-path";
 import { useNavigate } from "../navigation";
-import { NEW_WORKBENCH_PATH, ONBOARDING_PATH } from "../routes";
+import { ONBOARDING_PATH } from "../routes";
 
 type LandState =
   /** Working on it: the warm loader, whether we are reading the bench's
@@ -102,18 +101,26 @@ export function HomeRoute({
       retryTimer = setTimeout(() => setAttempt((count) => count + 1), retryMs);
     };
 
-    // Zero workbenches: wait for Myra's own definition to exist, then send
-    // the person to the picker rather than minting anything ourselves —
-    // "she can't start yet" and "here, go create your first workbench"
-    // are different messages, and only the readiness check tells them
-    // apart. Without a credential the drain never starts (CL-6780), so a
-    // not-ready status with no credential is an honest next step, not a
-    // forever spin on "getting ready".
+    // Zero workbenches: wait for Myra's own definition to exist, then
+    // open her DM the same way "Talk to Myra" does — never `/new`.
+    // "She can't start yet" and "here is her chat" are different
+    // messages, and only the readiness check plus a real definition
+    // tell them apart. Without a credential the drain never starts
+    // (CL-6780), so a not-ready status with no credential is an honest
+    // next step, not a forever spin on "getting ready".
     const awaitFirstWorkbench = () => {
       void fetchAgentReadiness().then((readiness) => {
         if (cancelled) return;
         if (readiness.kind === "ready" || readiness.kind === "chat-ready") {
-          navigate(NEW_WORKBENCH_PATH);
+          void listAgentDefinitions(selectedTenantId).then((definitions) => {
+            if (cancelled) return;
+            const myra = findMyraDefinition(definitions);
+            if (myra === undefined) {
+              waitAndRetry();
+              return;
+            }
+            void openAgentDmChat(selectedTenantId, myra.id, navigate);
+          });
           return;
         }
         void hasActiveCredential(selectedTenantId).then((probe) => {