fix(ai-chat): hydrate crashed mid-run steps for model replay + cover write/read seams (#492)

F1: the server model-replay loaded history via findAllByChat().map(rowToUiMessage)
WITHOUT hydrating parts from ai_chat_run_steps. A HARD crash mid-run (SIGKILL/OOM)
fires no terminal callback, so the assistant row stays parts:[] and its partial
tool-calls/results/text (durable in the steps table) dropped out of the model's
next-turn context. Hydrate needy assistant rows (role==='assistant' &&
!rowHasInlineParts) via findByMessageIds + hydrateAssistantParts before the replay
map — mirroring the controller's withReconstructedParts exactly — guarded on the
optional repo. Fix the now-false interrupt-resume comment.

F2: add a service int-spec that drives the REAL onStep append-persist WRITE branch
through AiChatService.stream with a real AiChatRunStepRepo injected, asserting the
per-step rows' stepIndex + parts slice and the step-marker metadata match a
single-row flush (catches an stepsPersisted-1 off-by-one).

F3: add a controller int-spec that drives withReconstructedParts through getMessages
WITH the repo present (a mid-run marker-only row + its step rows), asserting the
reconstructed metadata.parts and workspace-scoping.

F4: remove the dead countByMessage (zero prod callers; reconstructRunParts derives
stepsPersisted inline) + its now-unused sql import and the redundant test assertion.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

AGENTS.md

@@ -463,7 +463,6 @@ Vite SPA. Code is organized by feature under `apps/client/src/features/*` (mirro
 - The editor is Tiptap; shared node/mark extensions live in `packages/editor-ext` and are imported by **both the client and the server** (collaboration, schema, `canonicalizeFootnotes`) — editor schema changes often need to be made in `editor-ext`, not just the client. Server-side markdown import/export no longer lives in `editor-ext`: it goes through the canonical converter (#345, see below). The ProseMirror↔Markdown converter and its Docmost schema mirror now live in a SINGLE package, `@docmost/prosemirror-markdown` (#293), consumed by `mcp`, `git-sync`, `apps/server` (#345), and `apps/client` (#347) — do NOT reintroduce a per-package copy. The client uses the package's `browser` entry (`@docmost/prosemirror-markdown/browser`): markdown paste (`markdown-clipboard.ts`), copy-as-markdown, and AI-chat rendering now all go through the canonical converter, so the hand-written `marked`/`turndown` markdown layer that used to live in `editor-ext` was deleted (#347). The browser entry runs the HTML→DOM stage on the native `DOMParser`, so jsdom stays out of the client bundle. `editor-ext` is the upstream source of the Tiptap schema; the package's `docmost-schema.ts` mirrors it and a serializer-contract test (`packages/prosemirror-markdown/test/serializer-contract.test.ts`) guards the boundary (every schema node must have a converter case), so a drift surfaces as a failing test rather than silent divergence. For the converter's property-testing and counterexample→fixture process (P1–P4 invariants, the `PROPERTY_SEED`/`PROPERTY_NUM_RUNS` knobs, and the nightly fuzz workflow), see `packages/prosemirror-markdown/README.md`.
 - API access goes through `apps/client/src/lib/api-client.ts` (axios). The `@` alias maps to `apps/client/src`.
 - Runtime config is injected at build time by `vite.config.ts` via `define` (`APP_URL`, `COLLAB_URL`, `APP_VERSION`, …) — these come from the root `.env`, not from `import.meta.env`.
-- The build also emits `client/dist/version.json` (`{"version": …}`) from a small `vite.config.ts` plugin using the **same** `appVersion` that feeds `define.APP_VERSION`, so the file and the baked-in bundle version are identical by construction. The server reads it at startup (`ws.gateway.ts` via `readClientBuildVersion`/`resolveClientDistPath`) and announces it to each socket on connect (`app-version` event) so a tab left open across a redeploy can guard-reload before hitting a stale chunk (version-coherence). No runtime env / Dockerfile change — the file already ships in `client/dist`; missing/empty file ⇒ feature inert.
 
 ## Conventions
 

CHANGELOG.md

@@ -142,18 +142,6 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   snapshots switched from a fixed interval to a trailing idle-flush with a
   max-wait ceiling, and a boundary snapshot is pinned whenever the editing source
   changes (e.g. a person's edits followed by the AI agent). (#370)
-- **Open tabs pick up a new deploy on their own.** After the server is
-  redeployed while a tab is left open for hours, the tab now learns the new
-  build version over the existing WebSocket (announced per-connect, so a natural
-  reconnect delivers it) and shows a "A new version is available" banner with an
-  Update button. To avoid dropping a half-written comment or form, the tab is
-  not reloaded when you merely switch away from it; instead it auto-reloads at
-  the next safe point — the next in-app navigation (or immediately if you click
-  Update) — before it can hit a stale lazy-loaded chunk. At most one automatic
-  reload happens per 5-minute window, shared with the existing chunk-load
-  recovery, so a permanent version skew degrades to the banner rather than a
-  reload loop while a second deploy in the same tab still recovers. When the
-  build carries no version info the feature stays inert. (#481)
 
 - **Place several images side by side in a row.** A new "Inline (side by
   side)" alignment mode in the image bubble menu renders consecutive inline

apps/client/public/locales/en-US/translation.json

@@ -1,5 +1,4 @@
 {
-  "A new version is available": "A new version is available",
   "Account": "Account",
   "Active": "Active",
   "Add": "Add",

apps/client/public/locales/ru-RU/translation.json

@@ -1,5 +1,4 @@
 {
-  "A new version is available": "Доступна новая версия",
   "Account": "Аккаунт",
   "Active": "Активный",
   "Add": "Добавить",

apps/client/src/components/chunk-load-error-boundary.test.ts

@@ -1,5 +1,5 @@
 import { describe, it, expect } from "vitest";
-import { isChunkLoadError } from "./chunk-load-error-boundary";
+import { isChunkLoadError, shouldAutoReload } from "./chunk-load-error-boundary";
 
 // The detector decides whether a caught render error is a stale-deploy chunk-404
 // (→ auto-reload to fetch the new manifest) vs a genuine app error (→ generic
@@ -35,3 +35,31 @@ describe("isChunkLoadError", () => {
     expect(isChunkLoadError(err)).toBe(false);
   });
 });
+
+// The window gate replaces the old one-shot flag: it must permit recovery across
+// several deploys in one tab (each > window apart) while still stopping an infinite
+// reload loop when a lazy chunk is permanently broken (a second failure < window).
+describe("shouldAutoReload", () => {
+  const WINDOW = 5 * 60 * 1000;
+  const NOW = 1_000_000_000_000;
+
+  it("allows a reload when we have never auto-reloaded", () => {
+    expect(shouldAutoReload(NOW, null, WINDOW)).toBe(true);
+  });
+
+  it("allows a reload when the last one was 6 minutes ago (outside the window)", () => {
+    expect(shouldAutoReload(NOW, NOW - 6 * 60 * 1000, WINDOW)).toBe(true);
+  });
+
+  it("blocks a reload when the last one was 1 minute ago (inside the window)", () => {
+    expect(shouldAutoReload(NOW, NOW - 1 * 60 * 1000, WINDOW)).toBe(false);
+  });
+
+  it("blocks a reload exactly at the window boundary (not strictly older)", () => {
+    expect(shouldAutoReload(NOW, NOW - WINDOW, WINDOW)).toBe(false);
+  });
+
+  it("allows a reload when the stored timestamp is unparseable (NaN)", () => {
+    expect(shouldAutoReload(NOW, NaN, WINDOW)).toBe(true);
+  });
+});

apps/client/src/components/chunk-load-error-boundary.tsx

@@ -1,11 +1,26 @@
 import { ReactNode } from "react";
 import { ErrorBoundary } from "react-error-boundary";
 import { Button, Center, Stack, Text } from "@mantine/core";
-import {
+
-  hasAutoReloaded,
+// sessionStorage key holding the epoch-ms timestamp of the last automatic reload.
-  markAutoReloaded,
+const RELOAD_AT_KEY = "chunk-reload-at";
-  recordReloadBreadcrumb,
+// Allow at most one automatic reload per this window. A stale-deploy 404 is cured
-} from "@/lib/reload-guard";
+// by a single reload, so anything inside the window is treated as a reload loop
+// (permanently-broken chunk) and falls through to the manual UI. A window (rather
+// than a one-shot flag) lets a SECOND deploy in the same tab's lifetime recover too.
+const RELOAD_WINDOW_MS = 5 * 60 * 1000;
+
+// Pure window decision, unit-tested in isolation: auto-reload only if we have never
+// auto-reloaded (lastReloadAt null/NaN) or the last one was strictly older than the
+// window. Anything inside the window is suppressed to break an infinite reload loop.
+export function shouldAutoReload(
+  now: number,
+  lastReloadAt: number | null,
+  windowMs: number,
+): boolean {
+  if (lastReloadAt === null || Number.isNaN(lastReloadAt)) return true;
+  return now - lastReloadAt > windowMs;
+}
 
 // Heuristic detection of a failed dynamic import. Since the code-splitting work,
 // every route (plus Aside / AiChatWindow) is React.lazy: when a new deploy
@@ -24,26 +39,24 @@ export function isChunkLoadError(error: unknown): boolean {
   );
 }
 
-// Exported for tests: the reactive chunk-load reload decision, so the shared
+function handleError(error: unknown) {
-// window budget (invariant: ≤1 auto-reload per window across this path AND the
-// proactive version-coherence path) can be exercised against the real guard.
-export function handleError(error: unknown) {
   if (!isChunkLoadError(error)) return;
   // A stale-chunk 404 is cured by a full reload that re-fetches index.html and
-  // the new chunk manifest. Auto-reload at most once per window via the SHARED
+  // the new chunk manifest. Auto-reload at most once per RELOAD_WINDOW_MS: this
-  // window-based reload guard (see @/lib/reload-guard — the same budget the
+  // recovers across multiple deploys in a single tab's lifetime, yet a
-  // proactive version-coherence path consumes, so a mismatch that arrives on
+  // permanently-broken lazy chunk (which would loop) is stopped after the first
-  // both paths reloads at most once per window across BOTH). This recovers
+  // reload and falls through to the manual recovery UI below.
-  // across multiple deploys in a single tab's lifetime, yet a permanently-broken
+  try {
-  // lazy chunk (which would loop) is stopped after the first reload and falls
+    const raw = sessionStorage.getItem(RELOAD_AT_KEY);
-  // through to the manual recovery UI below. If the shared budget is already
+    const lastReloadAt = raw === null ? null : Number.parseInt(raw, 10);
-  // spent this window, or the stamp write fails (storage unavailable), we return
+    const now = Date.now();
-  // without reloading rather than risk a loop.
+    if (!shouldAutoReload(now, lastReloadAt, RELOAD_WINDOW_MS)) return;
-  if (hasAutoReloaded()) return;
+    sessionStorage.setItem(RELOAD_AT_KEY, String(now));
-  if (!markAutoReloaded()) return;
+  } catch {
-  // Trace before the reload clears the console (same diagnostic breadcrumb the
+    // sessionStorage unavailable (private mode / disabled): skip the automatic
-  // proactive version-coherence path writes, tagged with this path).
+    // reload rather than risk an unguarded loop; the fallback UI still recovers.
-  recordReloadBreadcrumb({ path: "chunk-boundary" });
+    return;
+  }
   window.location.reload();
 }
 

apps/client/src/features/ai-chat/components/message-item-memo.test.tsx

@@ -27,6 +27,7 @@ vi.mock("@/features/ai-chat/utils/markdown.ts", async () => {
 
 import MessageItem from "./message-item";
 import { messageSignature } from "@/features/ai-chat/utils/message-signature.ts";
+import { splitPlainChunks } from "./streaming-plain-text";
 
 // matchMedia (read by MantineProvider) is stubbed globally in vitest.setup.ts.
 
@@ -114,3 +115,89 @@ describe("MessageItem markdown memoization", () => {
     expect(queryByText("streamed answer")).not.toBeNull();
   });
 });
+
+// PERF SMOKE (#492): the whole point of the incremental streaming render is that
+// the ANSWER path costs O(number of markdown blocks), NOT O(number of throttled
+// ~20Hz ticks). Pre-#492 the finalized MarkdownPart re-parsed the WHOLE growing
+// answer on every delta — a synthetic ~100 KB stream measured 394 renderChatMarkdown
+// calls (one per tick). With the incremental render each STABILIZED block is parsed
+// exactly once (memoized in MarkdownChunk) and the live tail is cheap plain text, so
+// the call count collapses to ~= the block count regardless of tick granularity.
+describe("MessageItem streaming answer render is O(blocks), not O(ticks)", () => {
+  // ~100 KB answer. Each section is a heading + a paragraph — TWO blank-line
+  // delimited markdown blocks — so the safe-cut block count is ~2× the section
+  // count. The perf claim is about the BLOCK count (the memoization granularity),
+  // measured directly with splitPlainChunks below, not the section count.
+  const buildAnswer = () => {
+    const SECTIONS = 100;
+    const paragraphs: string[] = [];
+    for (let i = 0; i < SECTIONS; i++) {
+      paragraphs.push(`## Section ${i}\n\n` + "lorem ipsum dolor ".repeat(55));
+    }
+    const full = paragraphs.join("\n\n");
+    // The number of memoized markdown blocks the incremental render splits into
+    // (all but the live tail are parsed once each).
+    return { full, blocks: splitPlainChunks(full).length };
+  };
+
+  const streamMsg = (text: string, state: "streaming" | "done"): UIMessage =>
+    ({
+      id: "m1",
+      role: "assistant",
+      parts: [{ type: "text", text, state }],
+    }) as UIMessage;
+
+  it("parses each block ~once over a 100KB stream (≈blocks, ≪ ticks)", () => {
+    renderChatMarkdownSpy.mockClear();
+    const { full, blocks } = buildAnswer();
+    const CHUNK = 128; // a realistic ~20Hz throttled delta size
+    const ticks = Math.ceil(full.length / CHUNK);
+
+    let msg = streamMsg(full.slice(0, CHUNK), "streaming");
+    const { rerender } = render(
+      <MantineProvider>
+        <MessageItem
+          message={msg}
+          signature={messageSignature(msg)}
+          turnStreaming
+        />
+      </MantineProvider>,
+    );
+    for (let end = 2 * CHUNK; end < full.length; end += CHUNK) {
+      msg = streamMsg(full.slice(0, end), "streaming");
+      rerender(
+        <MantineProvider>
+          <MessageItem
+            message={msg}
+            signature={messageSignature(msg)}
+            turnStreaming
+          />
+        </MantineProvider>,
+      );
+    }
+    // Finalize: the streaming→done flip renders the whole answer through ONE
+    // canonical pass (visual parity), so the finished DOM matches the pre-#492
+    // output. This is the single extra parse on top of the per-block ones.
+    const done = streamMsg(full, "done");
+    rerender(
+      <MantineProvider>
+        <MessageItem message={done} signature={messageSignature(done)} />
+      </MantineProvider>,
+    );
+
+    const calls = renderChatMarkdownSpy.mock.calls.length;
+    // Sanity: the stream really had far more ticks than blocks (else the test is
+    // vacuous — the point is that calls scale with blocks, not ticks).
+    expect(ticks).toBeGreaterThan(blocks * 3);
+    // O(blocks): each stabilized block parsed once + the single final whole-text
+    // parse. A small constant absorbs the finalize render and the live-tail block;
+    // the load-bearing claim is the bound below.
+    expect(calls).toBeLessThanOrEqual(blocks + 2);
+    // ≪ ticks — and, non-vacuously, the blocks WERE parsed (not skipped entirely).
+    expect(calls).toBeLessThan(ticks / 3);
+    expect(calls).toBeGreaterThan(blocks / 2);
+    // MUTATION-VERIFY (documented, not run here): dropping the `memo()` wrapper on
+    // MarkdownChunk (so every stable block re-parses each tick) drives `calls`
+    // toward `ticks` (~394), reddening both upper-bound assertions above.
+  });
+});

apps/client/src/features/ai-chat/components/message-item.render.test.tsx

@@ -0,0 +1,112 @@
+import { describe, expect, it, vi } from "vitest";
+import { render } from "@testing-library/react";
+import { MantineProvider } from "@mantine/core";
+import type { UIMessage } from "@ai-sdk/react";
+
+// Stub react-i18next (the component reads `useTranslation`). Mirrors the other
+// message-item specs.
+vi.mock("react-i18next", () => ({
+  useTranslation: () => ({ t: (key: string) => key }),
+}));
+
+import MessageItem from "./message-item";
+import { messageSignature } from "@/features/ai-chat/utils/message-signature.ts";
+// The REAL canonical renderer (NOT the spy the memo test installs): this file
+// exercises the actual markdown output so the visual-regression assertions below
+// compare against genuine HTML (incl. the schema's `<li><p>` wrappers).
+import { renderChatMarkdown } from "@/features/ai-chat/utils/markdown.ts";
+import classes from "./ai-chat.module.css";
+
+const msg = (
+  parts: UIMessage["parts"],
+  extra?: Partial<UIMessage>,
+): UIMessage =>
+  ({ id: "m1", role: "assistant", parts, ...extra }) as UIMessage;
+
+const renderRow = (message: UIMessage, turnStreaming = false) =>
+  render(
+    <MantineProvider>
+      <MessageItem
+        message={message}
+        signature={messageSignature(message)}
+        turnStreaming={turnStreaming}
+      />
+    </MantineProvider>,
+  );
+
+// A rich multi-block answer that exercises headings, a list (the `<li><p>` case
+// the scoped CSS tightens), inline emphasis, and multiple paragraphs.
+const ANSWER = [
+  "# Заголовок",
+  "",
+  "Первый абзац с **жирным** и `кодом`.",
+  "",
+  "- пункт один",
+  "- пункт два",
+  "",
+  "Второй абзац.",
+].join("\n");
+
+describe("MessageItem final render — visual parity with the canonical pipeline", () => {
+  it("a finalized text part renders exactly renderChatMarkdown(text)", () => {
+    const { container } = renderRow(
+      msg([{ type: "text", text: ANSWER, state: "done" }]),
+    );
+    const block = container.querySelector(`.${classes.markdown}`);
+    expect(block).not.toBeNull();
+    // Byte-for-byte the canonical output (the SAME whole-text pass the pre-#492
+    // MarkdownPart produced), including `<li><p>…</p></li>` wrappers.
+    expect(block!.innerHTML).toBe(renderChatMarkdown(ANSWER, {}));
+    // The list wrapper is really present (guards against a vacuous empty render).
+    expect(container.querySelectorAll("li p").length).toBe(2);
+  });
+
+  it("the streaming incremental view CONVERGES to the canonical render on finish", () => {
+    // Mount mid-stream (live tail) — the DOM here is the incremental view.
+    const { container, rerender } = render(
+      <MantineProvider>
+        <MessageItem
+          message={msg([{ type: "text", text: ANSWER, state: "streaming" }])}
+          signature={messageSignature(
+            msg([{ type: "text", text: ANSWER, state: "streaming" }]),
+          )}
+          turnStreaming
+        />
+      </MantineProvider>,
+    );
+    // Finish the turn: state flips to done AND the turn is no longer streaming.
+    const done = msg([{ type: "text", text: ANSWER, state: "done" }]);
+    rerender(
+      <MantineProvider>
+        <MessageItem message={done} signature={messageSignature(done)} />
+      </MantineProvider>,
+    );
+    // After finish there is exactly ONE canonical markdown container whose HTML is
+    // the whole-text render — identical to the non-streaming path above.
+    const blocks = container.querySelectorAll(`.${classes.markdown}`);
+    expect(blocks.length).toBe(1);
+    expect(blocks[0].innerHTML).toBe(renderChatMarkdown(ANSWER, {}));
+  });
+
+  it("neutralizeInternalLinks is honored on the finalized render", () => {
+    const linkAnswer = "См. [страницу](/p/abc).";
+    const { container } = render(
+      <MantineProvider>
+        <MessageItem
+          message={msg([{ type: "text", text: linkAnswer, state: "done" }])}
+          signature={messageSignature(
+            msg([{ type: "text", text: linkAnswer, state: "done" }]),
+          )}
+          neutralizeInternalLinks
+        />
+      </MantineProvider>,
+    );
+    const block = container.querySelector(`.${classes.markdown}`);
+    expect(block!.innerHTML).toBe(
+      renderChatMarkdown(linkAnswer, { neutralizeInternalLinks: true }),
+    );
+    // The internal link was made inert (no href) by the neutralization flag.
+    const a = container.querySelector("a");
+    expect(a?.hasAttribute("href")).toBe(false);
+  });
+});

apps/client/src/features/ai-chat/components/message-item.tsx

@@ -4,6 +4,7 @@ import { useTranslation } from "react-i18next";
 import type { UIMessage } from "@ai-sdk/react";
 import ToolCallCard from "@/features/ai-chat/components/tool-call-card.tsx";
 import ReasoningBlock from "@/features/ai-chat/components/reasoning-block.tsx";
+import { StreamingMarkdownText } from "@/features/ai-chat/components/streaming-markdown-text.tsx";
 import ChatErrorAlert from "@/features/ai-chat/components/chat-error-alert.tsx";
 import ChatStoppedNotice from "@/features/ai-chat/components/chat-stopped-notice.tsx";
 import { ToolUiPart, isToolPart } from "@/features/ai-chat/utils/tool-parts.tsx";
@@ -86,17 +87,39 @@ interface MessageItemProps {
  * One assistant text part rendered as sanitized markdown. Memoized on its inputs
  * so a finalized text part is NOT re-parsed on every streamed delta: during a
  * turn only the actively-growing tail part changes its `text`, so every earlier
- * part hits the memo and skips the expensive marked + DOMPurify pass. Props are
+ * part hits the memo and skips the expensive canonical parse + DOMPurify pass.
- * primitives, so React.memo's default shallow compare is exactly right (the
+ * Props are primitives, so React.memo's default shallow compare is exactly right
- * `text` string is compared by value).
+ * (the `text` string is compared by value).
+ *
+ * Streaming gate (#492) — mirrors ReasoningBlock:
+ *  - `streaming` (this is the live, actively-growing tail part of an in-flight
+ *    turn): render incrementally via StreamingMarkdownText — the stabilized blocks
+ *    go through the canonical pipeline (each parsed ONCE, memoized) and only the
+ *    live tail is cheap plain text. This makes the per-tick cost O(new blocks),
+ *    not the pre-#492 O(ticks) whole-answer re-parse on every ~20Hz delta.
+ *  - finalized (the common case, and the turn-end flip): render the WHOLE text
+ *    through ONE canonical pass — byte-identical to the pre-#492 output (visual
+ *    parity). The row re-renders on the streaming→done flip because
+ *    `messageSignature` tracks each part's `state` (and `turnStreaming` flips at
+ *    turn end), so the incremental view always converges to this single render.
  */
 const MarkdownPart = memo(function MarkdownPart({
   text,
   neutralizeInternalLinks,
+  streaming,
 }: {
   text: string;
   neutralizeInternalLinks: boolean;
+  streaming: boolean;
 }) {
+  if (streaming) {
+    return (
+      <StreamingMarkdownText
+        text={text}
+        neutralizeInternalLinks={neutralizeInternalLinks}
+      />
+    );
+  }
   const html = renderChatMarkdown(text, { neutralizeInternalLinks });
   if (html) {
     return (
@@ -179,47 +202,10 @@ function MessageItem({
         {resolveAssistantName(assistantName) ?? t("AI agent")}
       </Text>
       {message.parts.map((part, index) => {
-        if (part.type === "reasoning") {
+        // Tool parts (`tool-*` / `dynamic-tool`) are template-literal kinds, so
-          // Reasoning ("thinking") -> a collapsible block with its own token
+        // they cannot be a `switch` case; the runtime guard handles them, and the
-          // count. Empty/whitespace reasoning with no authoritative count carries
+        // switch below covers every CLOSED (literal-typed) part kind with a
-          // nothing to show, so skip it (avoids an empty 0-token block).
+        // compile-time exhaustiveness check in its default.
-          const text = (part as { text?: string }).text ?? "";
-          if (!text.trim() && !(reasoningTokens && reasoningTokens > 0))
-            return null;
-          // Absent state (persisted rows) and "done" both mean finalized.
-          // `messageSignature` already includes each part's `state`, so the
-          // streaming→done flip changes the row signature and re-renders this
-          // row — which is what lets ReasoningBlock switch from chunked plain
-          // text to its one-time markdown parse (see reasoning-block.tsx).
-          // ALSO require the turn to be live: a part stranded at
-          // `state:"streaming"` after the turn ended (no `reasoning-end` — see
-          // the `turnStreaming` prop doc) must still finalize and parse.
-          const streaming =
-            turnStreaming && (part as { state?: string }).state === "streaming";
-          return (
-            <ReasoningBlock
-              key={index}
-              text={text}
-              tokens={reasoningTokens}
-              streaming={streaming}
-            />
-          );
-        }
-
-        if (part.type === "text") {
-          // Skip empty/whitespace-only text parts (a streaming message often
-          // starts with an empty text part before the first token arrives); the
-          // typing indicator covers that gap until real content streams in.
-          if (!part.text.trim()) return null;
-          return (
-            <MarkdownPart
-              key={index}
-              text={part.text}
-              neutralizeInternalLinks={neutralizeInternalLinks}
-            />
-          );
-        }
-
         if (isToolPart(part.type)) {
           return (
             <ToolCallCard
@@ -232,7 +218,76 @@ function MessageItem({
           );
         }
 
-        return null;
+        switch (part.type) {
+          case "reasoning": {
+            // Reasoning ("thinking") -> a collapsible block with its own token
+            // count. Empty/whitespace reasoning with no authoritative count
+            // carries nothing to show, so skip it (avoids an empty 0-token block).
+            const text = part.text ?? "";
+            if (!text.trim() && !(reasoningTokens && reasoningTokens > 0))
+              return null;
+            // Absent state (persisted rows) and "done" both mean finalized.
+            // `messageSignature` already includes each part's `state`, so the
+            // streaming→done flip changes the row signature and re-renders this
+            // row — which is what lets ReasoningBlock switch from chunked plain
+            // text to its one-time markdown parse (see reasoning-block.tsx).
+            // ALSO require the turn to be live: a part stranded at
+            // `state:"streaming"` after the turn ended (no `reasoning-end` — see
+            // the `turnStreaming` prop doc) must still finalize and parse.
+            const streaming = turnStreaming && part.state === "streaming";
+            return (
+              <ReasoningBlock
+                key={index}
+                text={text}
+                tokens={reasoningTokens}
+                streaming={streaming}
+              />
+            );
+          }
+
+          case "text": {
+            // Skip empty/whitespace-only text parts (a streaming message often
+            // starts with an empty text part before the first token arrives); the
+            // typing indicator covers that gap until real content streams in.
+            if (!part.text.trim()) return null;
+            // The live, actively-growing tail part of the in-flight turn renders
+            // incrementally (see MarkdownPart); a finalized part (persisted, or
+            // the turn-end flip) renders the whole text through one canonical
+            // pass. Same liveness rule as the reasoning branch above.
+            const streaming = turnStreaming && part.state === "streaming";
+            return (
+              <MarkdownPart
+                key={index}
+                text={part.text}
+                neutralizeInternalLinks={neutralizeInternalLinks}
+                streaming={streaming}
+              />
+            );
+          }
+
+          case "source-url":
+          case "source-document":
+          case "file":
+          case "step-start":
+            // Not surfaced in the chat bubble (v1) — same as the pre-#492 default.
+            return null;
+
+          default: {
+            // Compile-time exhaustiveness over the CLOSED union members: every
+            // literal-typed part kind is handled above, so the only kinds that
+            // can reach here are the OPEN template-literal ones (`tool-*` — caught
+            // by the guard at runtime — and `data-*`) plus `dynamic-tool`. Adding
+            // a NEW closed part kind to UIMessagePart makes this assignment fail
+            // to compile, forcing it to be handled instead of silently ignored
+            // (this replaces the pre-#492 fall-through `return null` + WARNING).
+            const _exhaustive:
+              | `tool-${string}`
+              | "dynamic-tool"
+              | `data-${string}` = part.type;
+            void _exhaustive;
+            return null;
+          }
+        }
       })}
       {/* A persisted turn error (server stored it in metadata.error). Rendered
           here so it survives a thread remount and shows in reopened history. */}

apps/client/src/features/ai-chat/components/streaming-markdown-text.tsx

@@ -0,0 +1,96 @@
+import { memo, useMemo } from "react";
+import { splitPlainChunks } from "@/features/ai-chat/components/streaming-plain-text.tsx";
+import { renderChatMarkdown } from "@/features/ai-chat/utils/markdown.ts";
+import classes from "@/features/ai-chat/components/ai-chat.module.css";
+
+/**
+ * One STABILIZED markdown block, rendered through the canonical pipeline and
+ * memoized on its string prop. During streaming only the TAIL chunk grows (the
+ * `splitPlainChunks` append-only invariant guarantees every earlier chunk is
+ * byte-identical across deltas), so React skips every stable block and each one
+ * is parsed by `renderChatMarkdown` EXACTLY ONCE — turning the pre-#492
+ * "re-parse the whole accumulated answer on every ~20Hz tick" (O(ticks)) into
+ * O(number of blocks). The markup is DOMPurify-sanitized inside renderChatMarkdown
+ * before it reaches `dangerouslySetInnerHTML`.
+ *
+ * NOTE (transient streaming-only artifact): a safe cut is a blank-line boundary,
+ * so a construct that legitimately contains a blank line (e.g. a fenced code block
+ * with an empty line) can be split across chunks and render oddly WHILE it is still
+ * streaming. This is cosmetic and self-heals: the moment the part finalizes,
+ * MarkdownPart renders the WHOLE text through one canonical pass (visual parity
+ * with the pre-#492 output). The reasoning path makes the same trade (plain text
+ * while streaming, one markdown parse at the end).
+ */
+const MarkdownChunk = memo(function MarkdownChunk({
+  text,
+  neutralizeInternalLinks,
+}: {
+  text: string;
+  neutralizeInternalLinks: boolean;
+}) {
+  const html = renderChatMarkdown(text, { neutralizeInternalLinks });
+  if (html) {
+    return (
+      <div
+        className={classes.markdown}
+        // Sanitized by renderChatMarkdown (DOMPurify) before insertion.
+        dangerouslySetInnerHTML={{ __html: html }}
+      />
+    );
+  }
+  // Malformed/unsupported markdown could not render synchronously: raw text.
+  return (
+    <div className={classes.markdown} style={{ whiteSpace: "pre-wrap" }}>
+      {text}
+    </div>
+  );
+});
+
+/**
+ * The cheap streaming-time stand-in for the finalized answer's one-time markdown
+ * parse (see MarkdownPart in message-item.tsx). Mirrors StreamingPlainText's
+ * chunked-memo pattern but renders the STABILIZED prefix as real markdown (each
+ * block parsed once, memoized) and only the LIVE tail as flat plain text — so the
+ * user sees formatted output for everything up to the last safe cut, and the not-
+ * yet-stable tail (which markdown-parsing every tick would make O(ticks)) stays a
+ * single cheap escaped text node until it stabilizes into a new block.
+ *
+ * `splitPlainChunks` yields chunks where, under append-only growth, every chunk
+ * except the LAST is immutable; the last chunk is the live tail. Index keys are
+ * therefore stable (a given index never changes to a different chunk's content).
+ */
+export function StreamingMarkdownText({
+  text,
+  neutralizeInternalLinks,
+}: {
+  text: string;
+  neutralizeInternalLinks: boolean;
+}) {
+  const chunks = useMemo(() => splitPlainChunks(text), [text]);
+  return (
+    <>
+      {chunks.map((chunk, index) =>
+        index < chunks.length - 1 ? (
+          <MarkdownChunk
+            key={index}
+            text={chunk}
+            neutralizeInternalLinks={neutralizeInternalLinks}
+          />
+        ) : (
+          // The live tail: flat, React-escaped plain text (no markdown parse, no
+          // sanitizer, no innerHTML). `pre-wrap` preserves its newlines; trailing
+          // separator newlines are dropped at display time so the block gap comes
+          // from the markdown margins, not a doubled empty line (mirrors
+          // PlainChunk in streaming-plain-text.tsx).
+          <div
+            key={index}
+            className={classes.markdown}
+            style={{ whiteSpace: "pre-wrap" }}
+          >
+            {chunk.replace(/\n+$/, "")}
+          </div>
+        ),
+      )}
+    </>
+  );
+}

apps/client/src/features/user/guarded-reload.test.tsx

@@ -1,195 +0,0 @@
-import { describe, it, expect, beforeEach, afterEach, vi } from "vitest";
-import { render, act, cleanup } from "@testing-library/react";
-import { MemoryRouter, useNavigate } from "react-router-dom";
-
-// Mocks for the dirty shell's side-effecting collaborators.
-vi.mock("@mantine/notifications", () => ({
-  notifications: { show: vi.fn() },
-}));
-vi.mock("@/i18n.ts", () => ({ default: { t: (k: string) => k } }));
-vi.mock("@/lib/reload-guard", () => ({
-  hasAutoReloaded: vi.fn(() => false),
-  markAutoReloaded: vi.fn(() => true),
-  recordReloadBreadcrumb: vi.fn(),
-  takeReloadBreadcrumb: vi.fn(() => null),
-}));
-
-import { notifications } from "@mantine/notifications";
-import { hasAutoReloaded, markAutoReloaded } from "@/lib/reload-guard";
-import {
-  triggerGuardedReload,
-  useVersionReloadOnNavigation,
-  __resetGuardedReloadForTests,
-} from "./guarded-reload";
-
-const show = notifications.show as unknown as ReturnType<typeof vi.fn>;
-const mockHasAutoReloaded = hasAutoReloaded as unknown as ReturnType<
-  typeof vi.fn
->;
-const mockMarkAutoReloaded = markAutoReloaded as unknown as ReturnType<
-  typeof vi.fn
->;
-
-let reload: ReturnType<typeof vi.fn>;
-let visibility: DocumentVisibilityState;
-
-// Test harness mounted inside a router: it installs the navigation hook and
-// exposes `navigate` so a test can drive an in-app router navigation.
-let doNavigate: (to: string) => void;
-function Harness() {
-  useVersionReloadOnNavigation();
-  const navigate = useNavigate();
-  doNavigate = navigate;
-  return null;
-}
-
-function mountHarness() {
-  render(
-    <MemoryRouter initialEntries={["/start"]}>
-      <Harness />
-    </MemoryRouter>,
-  );
-}
-
-function navigateTo(path: string) {
-  act(() => {
-    doNavigate(path);
-  });
-}
-
-beforeEach(() => {
-  __resetGuardedReloadForTests();
-  vi.clearAllMocks();
-  mockHasAutoReloaded.mockReturnValue(false);
-  mockMarkAutoReloaded.mockReturnValue(true);
-
-  vi.stubGlobal("APP_VERSION", "test-A");
-
-  reload = vi.fn();
-  Object.defineProperty(window, "location", {
-    configurable: true,
-    value: { reload },
-  });
-
-  visibility = "visible";
-  Object.defineProperty(document, "visibilityState", {
-    configurable: true,
-    get: () => visibility,
-  });
-});
-
-afterEach(() => {
-  cleanup();
-  vi.unstubAllGlobals();
-});
-
-describe("triggerGuardedReload (variant C)", () => {
-  it("noop when versions match: no banner, no reload", () => {
-    triggerGuardedReload("test-A");
-    expect(show).not.toHaveBeenCalled();
-    expect(reload).not.toHaveBeenCalled();
-  });
-
-  it("noop when the server version is empty (fail-safe)", () => {
-    triggerGuardedReload("");
-    triggerGuardedReload(undefined);
-    expect(show).not.toHaveBeenCalled();
-    expect(reload).not.toHaveBeenCalled();
-  });
-
-  it("real mismatch shows the banner but does NOT reload immediately", () => {
-    mountHarness();
-    triggerGuardedReload("test-B");
-    expect(show).toHaveBeenCalledTimes(1);
-    expect(show.mock.calls[0][0]).toMatchObject({
-      id: "app-version-reload",
-      autoClose: false,
-      withCloseButton: true,
-    });
-    expect(reload).not.toHaveBeenCalled();
-  });
-
-  it("reloads EXACTLY ONCE on the first in-app navigation after a mismatch", () => {
-    mountHarness();
-    triggerGuardedReload("test-B");
-    expect(reload).not.toHaveBeenCalled();
-
-    navigateTo("/next");
-    expect(reload).toHaveBeenCalledTimes(1);
-
-    // A second navigation must NOT reload again (one-shot was consumed).
-    navigateTo("/again");
-    expect(reload).toHaveBeenCalledTimes(1);
-  });
-
-  it("does NOT reload on a 2nd mismatch after the first was armed/consumed (one-shot)", () => {
-    mountHarness();
-    triggerGuardedReload("test-B");
-    navigateTo("/next");
-    expect(reload).toHaveBeenCalledTimes(1);
-
-    // Another app-version mismatch arrives (reconnect): must not re-arm.
-    triggerGuardedReload("test-C");
-    navigateTo("/again");
-    expect(reload).toHaveBeenCalledTimes(1);
-  });
-
-  it("does NOT reload merely from the tab going to the background", () => {
-    mountHarness();
-    triggerGuardedReload("test-B");
-    expect(reload).not.toHaveBeenCalled();
-
-    visibility = "hidden";
-    act(() => {
-      document.dispatchEvent(new Event("visibilitychange"));
-    });
-    expect(reload).not.toHaveBeenCalled();
-  });
-
-  it("a hidden-at-receipt tab is also NOT reloaded immediately (variant C uniform); reloads on next navigation", () => {
-    visibility = "hidden";
-    mountHarness();
-    triggerGuardedReload("test-B");
-    expect(reload).not.toHaveBeenCalled();
-    expect(show).toHaveBeenCalledTimes(1);
-
-    navigateTo("/next");
-    expect(reload).toHaveBeenCalledTimes(1);
-  });
-
-  it("the banner's Update button reloads immediately", () => {
-    triggerGuardedReload("test-B");
-    const message = show.mock.calls[0][0].message as {
-      props: { onClick: () => void };
-    };
-    message.props.onClick();
-    expect(reload).toHaveBeenCalledTimes(1);
-  });
-
-  it("banner-only (auto-reload already spent): banner, never auto-reload on navigation", () => {
-    mockHasAutoReloaded.mockReturnValue(true);
-    mountHarness();
-    triggerGuardedReload("test-B");
-    expect(show).toHaveBeenCalledTimes(1);
-
-    navigateTo("/next");
-    expect(reload).not.toHaveBeenCalled();
-  });
-
-  it("does NOT reload when the flag write fails; falls back to the banner", () => {
-    mockMarkAutoReloaded.mockReturnValue(false);
-    mountHarness();
-    triggerGuardedReload("test-B");
-    navigateTo("/next");
-    expect(reload).not.toHaveBeenCalled();
-    // performAutoReload falls back to showing the banner (initial + fallback).
-    expect(show).toHaveBeenCalled();
-  });
-
-  it("is idempotent within a tab-load: repeated emits do not stack banners", () => {
-    triggerGuardedReload("test-B");
-    triggerGuardedReload("test-B");
-    triggerGuardedReload("test-C");
-    expect(show).toHaveBeenCalledTimes(1);
-  });
-});

apps/client/src/features/user/guarded-reload.tsx

@@ -1,188 +0,0 @@
-import { useEffect, useRef } from "react";
-import { useLocation } from "react-router-dom";
-import { Button } from "@mantine/core";
-import { notifications } from "@mantine/notifications";
-import i18n from "@/i18n.ts";
-import {
-  hasAutoReloaded,
-  markAutoReloaded,
-  recordReloadBreadcrumb,
-  takeReloadBreadcrumb,
-} from "@/lib/reload-guard";
-import { decideVersionAction } from "@/features/user/version-coherence";
-
-// Dirty shell around the pure `decideVersionAction`: it reads globals
-// (APP_VERSION), touches sessionStorage via the shared reload-guard, drives the
-// Mantine notification, and arms the router-navigation reload hook. Kept
-// separate from the pure module so the decision stays unit-testable without a
-// DOM.
-
-// One fixed id so repeated app-version signals (e.g. every reconnect) update a
-// single banner instead of stacking a new one each time.
-const BANNER_ID = "app-version-reload";
-
-// Module-level idempotency for the current tab-load: once a mismatch has been
-// handled we don't re-arm the navigation reload or re-show the banner on
-// subsequent app-version emits.
-let handled = false;
-
-// Variant C: on a real mismatch we do NOT reload the tab when it merely goes to
-// the background (that would silently drop a half-written comment/form). Instead
-// we arm a one-shot reload for the NEXT in-app router navigation — a point where
-// the user is already leaving the current page, so an in-app navigation would
-// discard that unsaved component-state anyway and the reload adds no extra loss.
-let pendingNavReload = false;
-
-// Remembered from the last detected mismatch for the pre-reload breadcrumb and
-// the (already-visible) banner.
-let lastServerVersion = "";
-let lastClientVersion = "";
-
-// Read the build version baked into THIS bundle. The `typeof` guard avoids a
-// ReferenceError where the `APP_VERSION` global is absent (e.g. under vitest,
-// where Vite's `define` did not run) — an unknown client version makes the
-// pure decision no-op (fail-safe).
-function readClientVersion(): string {
-  return (typeof APP_VERSION !== "undefined" ? APP_VERSION : "").trim();
-}
-
-// Perform the actual reload — but only after the shared one-shot flag is
-// persisted. If the write fails (storage unavailable) we must NOT reload
-// (mirrors the reactive chunk-load boundary's `catch → return`), and fall back
-// to the manual banner so the user can still recover.
-function performAutoReload(): void {
-  if (!markAutoReloaded()) {
-    showReloadBanner();
-    return;
-  }
-  // Trace right before the reload (which clears the console): a persistent
-  // breadcrumb + a log line so the auto-reload is observable in a field report.
-  recordReloadBreadcrumb({
-    path: "proactive",
-    serverVersion: lastServerVersion,
-    clientVersion: lastClientVersion,
-  });
-  console.warn(
-    `[version-coherence] auto-reloading: client=${lastClientVersion} -> server=${lastServerVersion}`,
-  );
-  window.location.reload();
-}
-
-function showReloadBanner(): void {
-  notifications.show({
-    id: BANNER_ID,
-    title: i18n.t("A new version is available"),
-    message: (
-      <Button size="xs" mt="xs" onClick={() => performAutoReload()}>
-        {i18n.t("Update")}
-      </Button>
-    ),
-    autoClose: false,
-    withCloseButton: true,
-  });
-}
-
-/**
- * Handle a server `app-version` announcement: compare it to this bundle's
- * version and, on a real mismatch, show the banner and arm a guarded reload for
- * the next in-app navigation (variant C).
- *
- * - real mismatch (window budget available) → banner + arm navigation reload.
- *   The banner's "Update" button reloads immediately (same shared window guard).
- *   The tab is NOT reloaded on visibility change.
- * - auto-reload already used this window / storage error → banner only (no arm),
- *   so there is at most one automatic reload per RELOAD_WINDOW_MS window (loop
- *   safety).
- * - in sync / unknown version → noop (fail-safe).
- */
-export function triggerGuardedReload(
-  rawServerVersion: string | undefined | null,
-): void {
-  const serverVersion = (rawServerVersion ?? "").trim();
-  const clientVersion = readClientVersion();
-
-  // A storage read error surfaces as autoReloadUsed=true → fail toward NOT
-  // reloading (banner only).
-  const autoReloadUsed = hasAutoReloaded();
-
-  const action = decideVersionAction({
-    serverVersion,
-    clientVersion,
-    autoReloadUsed,
-  });
-  if (action === "noop") return;
-
-  // Idempotent per tab-load: don't re-arm or re-stack the banner across repeated
-  // emits (reconnects) once we've already acted.
-  if (handled) return;
-  handled = true;
-
-  lastServerVersion = serverVersion;
-  lastClientVersion = clientVersion;
-
-  if (action === "banner") {
-    // Entered banner-only (permanent skew, node oscillation, or the window's
-    // auto-reload budget already spent). Log for diagnosability; show the banner.
-    console.warn(
-      `[version-coherence] server=${serverVersion} client=${clientVersion}: ` +
-        "auto-reload budget already spent this window — showing manual banner",
-    );
-    showReloadBanner();
-    return;
-  }
-
-  // action === "reload" (variant C): show the banner and defer the auto-reload
-  // to the next in-app navigation instead of reloading now / on visibility.
-  showReloadBanner();
-  pendingNavReload = true;
-}
-
-/**
- * Consume the armed one-shot navigation reload, if any. Called by
- * `useVersionReloadOnNavigation` on each in-app router navigation.
- */
-export function consumeNavigationReload(): void {
-  if (!pendingNavReload) return;
-  pendingNavReload = false;
-  performAutoReload();
-}
-
-/**
- * Hook (mounted inside the Router) that fires the armed one-shot reload on the
- * NEXT in-app router navigation after a version mismatch. Skips the initial
- * render so it only reacts to real navigations, not the first location.
- */
-export function useVersionReloadOnNavigation(): void {
-  const location = useLocation();
-  const firstRender = useRef(true);
-  useEffect(() => {
-    if (firstRender.current) {
-      firstRender.current = false;
-      return;
-    }
-    consumeNavigationReload();
-  }, [location.key]);
-}
-
-/**
- * Surface (log once) the breadcrumb left by an auto-reload in the previous page
- * load — the reload cleared the console, so this makes a "tab reloaded itself"
- * report diagnosable. Call once on app startup.
- */
-export function surfacePreviousReloadBreadcrumb(): void {
-  const crumb = takeReloadBreadcrumb();
-  if (!crumb) return;
-  console.info(
-    `[version-coherence] previous auto-reload: path=${crumb.path} ` +
-      `client=${crumb.clientVersion ?? ""} -> server=${crumb.serverVersion ?? ""} ` +
-      `at=${new Date(crumb.at).toISOString()}`,
-  );
-}
-
-// Test-only: reset module-level latches between cases.
-export function __resetGuardedReloadForTests(): void {
-  handled = false;
-  pendingNavReload = false;
-  lastServerVersion = "";
-  lastClientVersion = "";
-}

apps/client/src/features/user/reload-budget.integration.test.tsx

@@ -1,171 +0,0 @@
-import { describe, it, expect, beforeEach, afterEach, vi } from "vitest";
-import { render, act, cleanup } from "@testing-library/react";
-import { MemoryRouter, useNavigate } from "react-router-dom";
-
-// Integration test for the SHARED, window-based auto-reload budget (invariant a):
-// the reactive chunk-load boundary and the proactive version-coherence path both
-// route through the REAL @/lib/reload-guard, so at most one automatic reload
-// happens per RELOAD_WINDOW_MS across BOTH paths combined. Only the two paths'
-// side-effecting collaborators are mocked — the reload guard is intentionally
-// REAL so this exercises the actual shared sessionStorage budget.
-vi.mock("@mantine/notifications", () => ({
-  notifications: { show: vi.fn() },
-}));
-vi.mock("@/i18n.ts", () => ({ default: { t: (k: string) => k } }));
-
-import { handleError } from "@/components/chunk-load-error-boundary";
-import {
-  triggerGuardedReload,
-  useVersionReloadOnNavigation,
-  __resetGuardedReloadForTests,
-} from "./guarded-reload";
-import { RELOAD_WINDOW_MS } from "@/lib/reload-guard";
-
-const CHUNK_ERROR = { name: "ChunkLoadError", message: "boom" };
-const T0 = 1_000_000_000_000;
-
-let reload: ReturnType<typeof vi.fn>;
-let nowMock: ReturnType<typeof vi.spyOn>;
-
-// Harness mounted inside a router: installs the navigation hook and exposes
-// `navigate` so a test can drive an in-app router navigation (the point where the
-// proactive path fires its armed reload).
-let doNavigate: (to: string) => void;
-function Harness() {
-  useVersionReloadOnNavigation();
-  doNavigate = useNavigate();
-  return null;
-}
-function mountHarness() {
-  render(
-    <MemoryRouter initialEntries={["/start"]}>
-      <Harness />
-    </MemoryRouter>,
-  );
-}
-function navigateTo(path: string) {
-  act(() => {
-    doNavigate(path);
-  });
-}
-function setNow(t: number) {
-  nowMock.mockReturnValue(t);
-}
-
-beforeEach(() => {
-  sessionStorage.clear();
-  __resetGuardedReloadForTests();
-  vi.clearAllMocks();
-  nowMock = vi.spyOn(Date, "now").mockReturnValue(T0);
-  vi.stubGlobal("APP_VERSION", "test-A");
-  reload = vi.fn();
-  Object.defineProperty(window, "location", {
-    configurable: true,
-    value: { reload },
-  });
-});
-
-afterEach(() => {
-  cleanup();
-  vi.unstubAllGlobals();
-  vi.restoreAllMocks();
-  sessionStorage.clear();
-});
-
-describe("shared window-based reload budget (invariant a)", () => {
-  it("a chunk-load reload spends the budget: a version-coherence mismatch within the window shows the banner but does NOT reload", () => {
-    // Path 1 (reactive): a stale-chunk 404 auto-reloads once and stamps the window.
-    handleError(CHUNK_ERROR);
-    expect(reload).toHaveBeenCalledTimes(1);
-    reload.mockClear();
-
-    // Path 2 (proactive), 1 min later — still inside the window. The shared budget
-    // is spent, so the version mismatch degrades to the banner and never reloads.
-    setNow(T0 + 60_000);
-    mountHarness();
-    triggerGuardedReload("test-B");
-    navigateTo("/next");
-    expect(reload).not.toHaveBeenCalled();
-  });
-
-  it("a version-coherence reload spends the SAME budget: a chunk-load error within the window does NOT reload", () => {
-    // Path 2 (proactive) first: real mismatch → arm → fire on navigation.
-    mountHarness();
-    triggerGuardedReload("test-B");
-    navigateTo("/next");
-    expect(reload).toHaveBeenCalledTimes(1);
-    reload.mockClear();
-
-    // Path 1 (reactive), 2 min later — inside the window. Budget already spent by
-    // the proactive path, so the stale-chunk error must NOT trigger a second reload.
-    setNow(T0 + 2 * 60_000);
-    handleError(CHUNK_ERROR);
-    expect(reload).not.toHaveBeenCalled();
-  });
-
-  it("recovers after the window: a reload strictly older than the window is allowed again (second deploy)", () => {
-    // First auto-reload (reactive) stamps the window.
-    handleError(CHUNK_ERROR);
-    expect(reload).toHaveBeenCalledTimes(1);
-    reload.mockClear();
-
-    // A second deploy arrives after the window has fully elapsed → the proactive
-    // path is allowed to reload again (window, not a permanent one-shot).
-    __resetGuardedReloadForTests();
-    setNow(T0 + RELOAD_WINDOW_MS + 1);
-    mountHarness();
-    triggerGuardedReload("test-B");
-    navigateTo("/next");
-    expect(reload).toHaveBeenCalledTimes(1);
-  });
-
-  it("reactive path fails closed when storage READS but cannot WRITE (quota / Safari private): no unguarded reload", () => {
-    // getItem→null makes hasAutoReloaded() report the budget as available, so
-    // handleError passes the first guard and reaches `if (!markAutoReloaded())
-    // return;`. setItem throws → the stamp cannot stick, so markAutoReloaded()
-    // returns false and that guard MUST bail — otherwise the reactive path would
-    // reload on every stale-chunk error with no persisted budget (an unguarded
-    // loop). This is the asymmetric gap: the proactive path's equivalent is
-    // covered by guarded-reload.test.tsx "does NOT reload when the flag write
-    // fails".
-    vi.stubGlobal("sessionStorage", {
-      getItem: () => null,
-      setItem: () => {
-        throw new Error("quota exceeded");
-      },
-      removeItem: () => {},
-      clear: () => {},
-    });
-    try {
-      handleError(CHUNK_ERROR);
-      expect(reload).not.toHaveBeenCalled();
-    } finally {
-      vi.unstubAllGlobals();
-    }
-  });
-
-  it("sessionStorage unavailable: neither path performs an unguarded reload", () => {
-    // The real guard fails toward NOT reloading when storage throws.
-    vi.stubGlobal("sessionStorage", {
-      getItem: () => {
-        throw new Error("storage disabled");
-      },
-      setItem: () => {
-        throw new Error("storage disabled");
-      },
-      removeItem: () => {},
-      clear: () => {},
-    });
-    try {
-      handleError(CHUNK_ERROR);
-      expect(reload).not.toHaveBeenCalled();
-
-      mountHarness();
-      triggerGuardedReload("test-B");
-      navigateTo("/next");
-      expect(reload).not.toHaveBeenCalled();
-    } finally {
-      vi.unstubAllGlobals();
-    }
-  });
-});

apps/client/src/features/user/user-provider.tsx

@@ -13,12 +13,6 @@ import { useCollabToken } from "@/features/auth/queries/auth-query.tsx";
 import { Error404 } from "@/components/ui/error-404.tsx";
 import { queryClient } from "@/main.tsx";
 import { makeConnectHandler } from "@/features/user/connect-resync.ts";
-import {
-  triggerGuardedReload,
-  useVersionReloadOnNavigation,
-  surfacePreviousReloadBreadcrumb,
-} from "@/features/user/guarded-reload.tsx";
-import type { AppVersionSocketPayload } from "@/features/user/version-coherence.ts";
 
 export function UserProvider({ children }: React.PropsWithChildren) {
   const [, setCurrentUser] = useAtom(currentUserAtom);
@@ -28,16 +22,6 @@ export function UserProvider({ children }: React.PropsWithChildren) {
   // fetch collab token on load
   const { data: collab } = useCollabToken();
 
-  // version-coherence: fire the armed one-shot reload on the next in-app
-  // navigation (variant C — a safe point, not on tab backgrounding).
-  useVersionReloadOnNavigation();
-
-  // Surface any breadcrumb left by an auto-reload in the previous page load
-  // (the reload cleared the console) so a field report stays diagnosable.
-  useEffect(() => {
-    surfacePreviousReloadBreadcrumb();
-  }, []);
-
   useEffect(() => {
     if (isLoading || isError) {
       return;
@@ -63,16 +47,6 @@ export function UserProvider({ children }: React.PropsWithChildren) {
       handleConnect();
     });
 
-    // Register the version-coherence listener SYNCHRONOUSLY, before the socket
-    // connects: the server emits `app-version` immediately in handleConnection,
-    // so a listener attached after connect would miss it on a fast localhost
-    // connect. On a version mismatch the client shows a banner and defers the
-    // auto-reload to the next in-app navigation (variant C — avoids reloading a
-    // backgrounded tab that may hold unsaved input) before it hits a stale chunk.
-    newSocket.on("app-version", (payload?: AppVersionSocketPayload) => {
-      triggerGuardedReload(payload?.version);
-    });
-
     return () => {
       console.log("ws disconnected");
       newSocket.disconnect();

apps/client/src/features/user/version-coherence.test.ts

@@ -1,64 +0,0 @@
-import { describe, it, expect } from "vitest";
-import { decideVersionAction } from "./version-coherence";
-
-describe("decideVersionAction", () => {
-  it("noop when the server version is empty (fail-safe)", () => {
-    expect(
-      decideVersionAction({
-        serverVersion: "",
-        clientVersion: "v1",
-        autoReloadUsed: false,
-      }),
-    ).toBe("noop");
-  });
-
-  it("noop when the client version is empty (fail-safe)", () => {
-    expect(
-      decideVersionAction({
-        serverVersion: "v1",
-        clientVersion: "",
-        autoReloadUsed: false,
-      }),
-    ).toBe("noop");
-  });
-
-  it("noop when versions are equal (in sync)", () => {
-    expect(
-      decideVersionAction({
-        serverVersion: "v1",
-        clientVersion: "v1",
-        autoReloadUsed: false,
-      }),
-    ).toBe("noop");
-  });
-
-  it("reload on a real mismatch the first time this session", () => {
-    expect(
-      decideVersionAction({
-        serverVersion: "test-B",
-        clientVersion: "test-A",
-        autoReloadUsed: false,
-      }),
-    ).toBe("reload");
-  });
-
-  it("banner on a mismatch once the session auto-reload is spent", () => {
-    expect(
-      decideVersionAction({
-        serverVersion: "test-B",
-        clientVersion: "test-A",
-        autoReloadUsed: true,
-      }),
-    ).toBe("banner");
-  });
-
-  it("equal versions stay noop even if auto-reload was already used", () => {
-    expect(
-      decideVersionAction({
-        serverVersion: "v1",
-        clientVersion: "v1",
-        autoReloadUsed: true,
-      }),
-    ).toBe("noop");
-  });
-});

apps/client/src/features/user/version-coherence.ts

@@ -1,32 +0,0 @@
-// Payload of the per-connect `app-version` socket.io event announced by the
-// server (ws.gateway.ts) after a successful auth. A dedicated event — NOT a
-// member of the room-scoped `WebSocketEvent` union (which is discriminated by
-// `operation`), so it never touches use-query-subscription.
-export type AppVersionSocketPayload = { version: string };
-
-/**
- * Pure decision for the version-coherence guard.
- *
- * All inputs are injected (no globals, no side effects) so it is unit-testable
- * without a DOM or the build-time `APP_VERSION` global (undefined under vitest).
- *
- * - `autoReloadUsed` = an automatic reload has already happened within the
- *   current ~5-min window, so we must not auto-reload again (loop safety,
- *   shared window budget with the reactive chunk-load boundary).
- *
- * Returns:
- * - "noop"   — do nothing (unknown version on either side, or already in sync).
- * - "banner" — show the manual "update available" banner only (no auto-reload).
- * - "reload" — real first-time mismatch: eligible for a guarded auto-reload.
- */
-export function decideVersionAction(args: {
-  serverVersion: string;
-  clientVersion: string;
-  autoReloadUsed: boolean;
-}): "reload" | "banner" | "noop" {
-  const { serverVersion, clientVersion, autoReloadUsed } = args;
-  if (!serverVersion || !clientVersion) return "noop"; // fail-safe: unknown version → never act
-  if (serverVersion === clientVersion) return "noop"; // in sync
-  if (autoReloadUsed) return "banner"; // one auto-reload per RELOAD_WINDOW_MS window already spent
-  return "reload"; // real mismatch, window budget available
-}

apps/client/src/lib/reload-guard.test.ts

@@ -1,145 +0,0 @@
-import { describe, it, expect, beforeEach, afterEach, vi } from "vitest";
-import {
-  hasAutoReloaded,
-  markAutoReloaded,
-  shouldAutoReload,
-  recordReloadBreadcrumb,
-  takeReloadBreadcrumb,
-  RELOAD_WINDOW_MS,
-} from "./reload-guard";
-
-// The shared budget is a single sessionStorage timestamp keyed here; both the
-// reactive chunk-load boundary and the proactive version-coherence path read and
-// stamp it, so at most one auto-reload happens per RELOAD_WINDOW_MS across BOTH.
-const RELOAD_AT_KEY = "chunk-reload-at";
-const NOW = 1_000_000_000_000;
-
-describe("reload-guard", () => {
-  beforeEach(() => {
-    sessionStorage.clear();
-    vi.restoreAllMocks();
-  });
-
-  afterEach(() => {
-    sessionStorage.clear();
-    vi.restoreAllMocks();
-  });
-
-  it("hasAutoReloaded is false before any reload, true within the window after mark", () => {
-    expect(hasAutoReloaded(NOW)).toBe(false);
-    expect(markAutoReloaded(NOW)).toBe(true);
-    // Same key both paths share; stores the reload timestamp, not a flag.
-    expect(sessionStorage.getItem(RELOAD_AT_KEY)).toBe(String(NOW));
-    // Inside the window → budget spent → true (fall through to manual UI).
-    expect(hasAutoReloaded(NOW)).toBe(true);
-    expect(hasAutoReloaded(NOW + RELOAD_WINDOW_MS)).toBe(true);
-  });
-
-  it("hasAutoReloaded is false again once the window has elapsed (a later deploy recovers)", () => {
-    markAutoReloaded(NOW);
-    // Strictly older than the window → a new deploy's mismatch may reload again.
-    expect(hasAutoReloaded(NOW + RELOAD_WINDOW_MS + 1)).toBe(false);
-  });
-
-  it("hasAutoReloaded returns true when reading storage throws (fail toward not reloading)", () => {
-    vi.stubGlobal("sessionStorage", {
-      getItem: () => {
-        throw new Error("storage disabled");
-      },
-      setItem: () => {
-        throw new Error("storage disabled");
-      },
-    });
-    try {
-      expect(hasAutoReloaded()).toBe(true);
-    } finally {
-      vi.unstubAllGlobals();
-    }
-  });
-
-  it("hasAutoReloaded treats an unparseable stored timestamp as never-reloaded", () => {
-    sessionStorage.setItem(RELOAD_AT_KEY, "not-a-number");
-    expect(hasAutoReloaded(NOW)).toBe(false);
-  });
-
-  it("markAutoReloaded returns false when writing storage throws", () => {
-    vi.stubGlobal("sessionStorage", {
-      getItem: () => null,
-      setItem: () => {
-        throw new Error("storage disabled");
-      },
-    });
-    try {
-      expect(markAutoReloaded()).toBe(false);
-    } finally {
-      vi.unstubAllGlobals();
-    }
-  });
-
-  it("records and then takes a breadcrumb once (cleared on read)", () => {
-    recordReloadBreadcrumb({
-      path: "proactive",
-      serverVersion: "test-B",
-      clientVersion: "test-A",
-    });
-    const crumb = takeReloadBreadcrumb();
-    expect(crumb).toMatchObject({
-      path: "proactive",
-      serverVersion: "test-B",
-      clientVersion: "test-A",
-    });
-    expect(typeof crumb?.at).toBe("number");
-    // Cleared on read → a second take returns null.
-    expect(takeReloadBreadcrumb()).toBeNull();
-  });
-
-  it("takeReloadBreadcrumb returns null when nothing was recorded", () => {
-    expect(takeReloadBreadcrumb()).toBeNull();
-  });
-
-  it("recordReloadBreadcrumb swallows a storage-write error (diagnostics only)", () => {
-    vi.stubGlobal("sessionStorage", {
-      getItem: () => null,
-      setItem: () => {
-        throw new Error("storage disabled");
-      },
-      removeItem: () => {},
-    });
-    try {
-      expect(() =>
-        recordReloadBreadcrumb({ path: "chunk-boundary" }),
-      ).not.toThrow();
-    } finally {
-      vi.unstubAllGlobals();
-    }
-  });
-});
-
-// The pure window gate replaces the old one-shot flag: it must permit recovery
-// across several deploys in one tab (each > window apart) while still stopping an
-// infinite reload loop when a lazy chunk is permanently broken (a second failure
-// < window). Moved here from the chunk-load boundary now that it is the shared
-// guard both paths route through.
-describe("shouldAutoReload", () => {
-  const WINDOW = RELOAD_WINDOW_MS;
-
-  it("allows a reload when we have never auto-reloaded", () => {
-    expect(shouldAutoReload(NOW, null, WINDOW)).toBe(true);
-  });
-
-  it("allows a reload when the last one was 6 minutes ago (outside the window)", () => {
-    expect(shouldAutoReload(NOW, NOW - 6 * 60 * 1000, WINDOW)).toBe(true);
-  });
-
-  it("blocks a reload when the last one was 1 minute ago (inside the window)", () => {
-    expect(shouldAutoReload(NOW, NOW - 1 * 60 * 1000, WINDOW)).toBe(false);
-  });
-
-  it("blocks a reload exactly at the window boundary (not strictly older)", () => {
-    expect(shouldAutoReload(NOW, NOW - WINDOW, WINDOW)).toBe(false);
-  });
-
-  it("allows a reload when the stored timestamp is unparseable (NaN)", () => {
-    expect(shouldAutoReload(NOW, NaN, WINDOW)).toBe(true);
-  });
-});

apps/client/src/lib/reload-guard.ts

@@ -1,121 +0,0 @@
-// Shared, window-based auto-reload budget.
-//
-// Both auto-reload paths — the reactive chunk-load-error-boundary (recovers
-// AFTER a stale lazy chunk 404s) and the proactive version-coherence feature
-// (reloads BEFORE the tab hits a stale chunk) — go through these functions so
-// they share ONE window-scoped reload budget: at most a single automatic
-// reload per RELOAD_WINDOW_MS across BOTH paths. A window (rather than a
-// permanent one-shot flag) lets a SECOND deploy in the same tab's lifetime
-// recover too, while a permanent skew, node oscillation, or a genuinely-missing
-// chunk still degrades to a manual banner/UI after the first reload instead of
-// looping. When sessionStorage is unavailable every mismatch degrades to the
-// manual UI — no unguarded reload.
-
-// sessionStorage key holding the epoch-ms timestamp of the last automatic reload
-// (shared by both paths).
-const RELOAD_AT_KEY = "chunk-reload-at";
-
-// Allow at most one automatic reload per this window. A stale-deploy 404 is cured
-// by a single reload, so anything inside the window is treated as a reload loop
-// (permanently-broken chunk / permanent skew) and falls through to the manual UI.
-export const RELOAD_WINDOW_MS = 5 * 60 * 1000;
-
-/**
- * Pure window decision, unit-tested in isolation: auto-reload only if we have
- * never auto-reloaded (lastReloadAt null/NaN) or the last one was strictly older
- * than the window. Anything inside the window is suppressed to break an infinite
- * reload loop.
- */
-export function shouldAutoReload(
-  now: number,
-  lastReloadAt: number | null,
-  windowMs: number,
-): boolean {
-  if (lastReloadAt === null || Number.isNaN(lastReloadAt)) return true;
-  return now - lastReloadAt > windowMs;
-}
-
-/**
- * Has an automatic reload already happened within the current window (so the
- * shared budget is spent right now)? Both paths check this before reloading; a
- * `true` return means fall through to the manual banner/UI instead of reloading.
- *
- * A storage read error (private mode / disabled) is reported as `true` so the
- * caller fails toward NOT reloading — an unguarded loop is worse than a stale
- * tab the user can reload manually. Note a window (not a permanent flag): once
- * the window elapses a later deploy's mismatch is allowed to reload again.
- */
-export function hasAutoReloaded(now: number = Date.now()): boolean {
-  try {
-    const raw = sessionStorage.getItem(RELOAD_AT_KEY);
-    const lastReloadAt = raw === null ? null : Number.parseInt(raw, 10);
-    return !shouldAutoReload(now, lastReloadAt, RELOAD_WINDOW_MS);
-  } catch {
-    return true;
-  }
-}
-
-/**
- * Stamp the shared window as consumed now — record that an automatic reload is
- * being performed within the current RELOAD_WINDOW_MS window.
- *
- * Returns whether the write succeeded. A `false` return (storage unavailable)
- * means the caller MUST NOT reload — otherwise the stamp would never stick and
- * the reload could loop.
- */
-export function markAutoReloaded(now: number = Date.now()): boolean {
-  try {
-    sessionStorage.setItem(RELOAD_AT_KEY, String(now));
-    return true;
-  } catch {
-    return false;
-  }
-}
-
-// Diagnostic breadcrumb for an automatic reload. Written right before
-// window.location.reload() (which clears the console) and read back on the next
-// page load, so a "the tab reloaded itself / it's looping" field report is
-// diagnosable: which path fired (proactive version-coherence vs the reactive
-// chunk-load boundary) and which version pair triggered it. sessionStorage
-// survives a same-tab reload, unlike the console.
-const RELOAD_BREADCRUMB_KEY = "reload-breadcrumb";
-
-export type ReloadBreadcrumb = {
-  path: "proactive" | "chunk-boundary";
-  serverVersion?: string;
-  clientVersion?: string;
-  at: number;
-};
-
-/**
- * Persist a best-effort breadcrumb just before an automatic reload. Failures
- * (storage unavailable) are swallowed — this is diagnostics only and must never
- * block or alter the reload decision.
- */
-export function recordReloadBreadcrumb(
-  entry: Omit<ReloadBreadcrumb, "at">,
-): void {
-  try {
-    sessionStorage.setItem(
-      RELOAD_BREADCRUMB_KEY,
-      JSON.stringify({ ...entry, at: Date.now() }),
-    );
-  } catch {
-    // best-effort diagnostics only
-  }
-}
-
-/**
- * Read and clear the breadcrumb left by an auto-reload in the previous page
- * load. Cleared on read so it surfaces exactly once per reload.
- */
-export function takeReloadBreadcrumb(): ReloadBreadcrumb | null {
-  try {
-    const raw = sessionStorage.getItem(RELOAD_BREADCRUMB_KEY);
-    if (!raw) return null;
-    sessionStorage.removeItem(RELOAD_BREADCRUMB_KEY);
-    return JSON.parse(raw) as ReloadBreadcrumb;
-  } catch {
-    return null;
-  }
-}

apps/client/vite.config.ts

@@ -1,8 +1,7 @@
-import { defineConfig, loadEnv, type Plugin } from "vite";
+import { defineConfig, loadEnv } from "vite";
 import react from "@vitejs/plugin-react";
 import { compression } from "vite-plugin-compression2";
 import * as path from "path";
-import * as fs from "node:fs";
 import { execSync } from "node:child_process";
 
 const envPath = path.resolve(process.cwd(), "..", "..");
@@ -25,32 +24,7 @@ function resolveAppVersion(cwd: string): string {
   }
 }
 
-// Emit <outDir>/version.json = { "version": appVersion } so the server can read
-// the exact same build id the bundle was compiled with. The value is the SAME
-// `appVersion` fed into `define.APP_VERSION`, so version.json and the baked-in
-// global are identical by construction — the single source of truth (no
-// runtime-env second copy that could drift and cause a false version mismatch).
-function versionJsonPlugin(version: string): Plugin {
-  let outDir = "dist";
-  return {
-    name: "emit-version-json",
-    apply: "build",
-    configResolved(config) {
-      outDir = config.build.outDir;
-    },
-    writeBundle() {
-      const root = path.resolve(process.cwd(), outDir);
-      fs.mkdirSync(root, { recursive: true });
-      fs.writeFileSync(
-        path.join(root, "version.json"),
-        JSON.stringify({ version }),
-      );
-    },
-  };
-}
-
 export default defineConfig(({ mode }) => {
-  const appVersion = resolveAppVersion(envPath);
   const {
     APP_URL,
     FILE_UPLOAD_SIZE_LIMIT,
@@ -78,11 +52,10 @@ export default defineConfig(({ mode }) => {
         POSTHOG_HOST,
         POSTHOG_KEY,
       },
-      APP_VERSION: JSON.stringify(appVersion),
+      APP_VERSION: JSON.stringify(resolveAppVersion(envPath)),
     },
     plugins: [
       react(),
-      versionJsonPlugin(appVersion),
       // Emit .br and .gz next to every built asset so the server can serve the
       // precompressed copy (see @fastify/static preCompressed in static.module.ts).
       compression({

apps/server/src/common/helpers/client-version.spec.ts

@@ -1,52 +0,0 @@
-import { join } from 'path';
-import * as fs from 'node:fs';
-import * as os from 'node:os';
-import { readClientBuildVersion } from './client-version';
-
-describe('readClientBuildVersion', () => {
-  let dir: string;
-
-  beforeEach(() => {
-    dir = fs.mkdtempSync(join(os.tmpdir(), 'client-version-'));
-  });
-
-  afterEach(() => {
-    fs.rmSync(dir, { recursive: true, force: true });
-  });
-
-  const writeVersionJson = (content: string) =>
-    fs.writeFileSync(join(dir, 'version.json'), content);
-
-  it('returns the version from a valid version.json', () => {
-    writeVersionJson(JSON.stringify({ version: 'test-A' }));
-    expect(readClientBuildVersion(dir)).toBe('test-A');
-  });
-
-  it('trims surrounding whitespace in the version', () => {
-    writeVersionJson(JSON.stringify({ version: '  v1.2.3  ' }));
-    expect(readClientBuildVersion(dir)).toBe('v1.2.3');
-  });
-
-  it('returns "" when version.json is missing', () => {
-    expect(readClientBuildVersion(dir)).toBe('');
-  });
-
-  it('returns "" on malformed JSON', () => {
-    writeVersionJson('{ not json');
-    expect(readClientBuildVersion(dir)).toBe('');
-  });
-
-  it('returns "" when the version field is absent', () => {
-    writeVersionJson(JSON.stringify({ notVersion: 'x' }));
-    expect(readClientBuildVersion(dir)).toBe('');
-  });
-
-  it('returns "" when the version field is not a string', () => {
-    writeVersionJson(JSON.stringify({ version: 123 }));
-    expect(readClientBuildVersion(dir)).toBe('');
-  });
-
-  it('returns "" when the path does not exist at all', () => {
-    expect(readClientBuildVersion(join(dir, 'nope'))).toBe('');
-  });
-});

apps/server/src/common/helpers/client-version.ts

@@ -1,36 +0,0 @@
-import { join } from 'path';
-import * as fs from 'node:fs';
-
-/**
- * Resolve the absolute path to the built client bundle directory
- * (`apps/client/dist`) shipped into the runtime image.
- *
- * The `../` depth is anchored on THIS module's compiled location
- * (`dist/common/helpers`). `integrations/static` sits at the same depth under
- * the compiled root, so both callers (StaticModule and readClientBuildVersion)
- * MUST share this single helper rather than duplicating the depth — a copy in a
- * module at a different depth would silently resolve to the wrong directory.
- */
-export function resolveClientDistPath(): string {
-  return join(__dirname, '..', '..', '..', '..', 'client/dist');
-}
-
-/**
- * Read the build version the client bundle was compiled with, from
- * `<clientDistPath>/version.json` (written by the Vite build — the single
- * source of truth shared by the baked-in `APP_VERSION` global and this file).
- *
- * Fail-safe: any error (missing file, unreadable, bad JSON, non-string
- * version) yields `''`. The caller treats an empty version as "unknown" and
- * the whole version-coherence feature stays silently inert — existing deploys
- * without the file keep working unchanged.
- */
-export function readClientBuildVersion(clientDistPath: string): string {
-  try {
-    const raw = fs.readFileSync(join(clientDistPath, 'version.json'), 'utf8');
-    const version = (JSON.parse(raw) as { version?: unknown }).version;
-    return typeof version === 'string' ? version.trim() : '';
-  } catch {
-    return '';
-  }
-}

apps/server/src/common/helpers/index.ts

@@ -3,4 +3,3 @@ export * from './nanoid.utils';
 export * from './file.helper';
 export * from './constants';
 export * from './security-headers';
-export * from './client-version';

apps/server/src/core/ai-chat/ai-chat.controller.ts

@@ -35,6 +35,7 @@ import {
 import { PaginationOptions } from '@docmost/db/pagination/pagination-options';
 import { AiChatRepo } from '@docmost/db/repos/ai-chat/ai-chat.repo';
 import { AiChatMessageRepo } from '@docmost/db/repos/ai-chat/ai-chat-message.repo';
+import { AiChatRunStepRepo } from '@docmost/db/repos/ai-chat/ai-chat-run-step.repo';
 import { PageRepo } from '@docmost/db/repos/page/page.repo';
 import { UserThrottlerGuard } from '../../integrations/throttle/user-throttler.guard';
 import { AI_CHAT_THROTTLER } from '../../integrations/throttle/throttler-names';
@@ -43,6 +44,8 @@ import {
   AiChatRunHooks,
   AiChatService,
   AiChatStreamBody,
+  rowHasInlineParts,
+  hydrateAssistantParts,
 } from './ai-chat.service';
 import { AiChatRunService } from './ai-chat-run.service';
 import { AiTranscriptionService } from './ai-transcription.service';
@@ -129,8 +132,39 @@ export class AiChatController {
     // production. Only touched on the resumable-stream (flag-on) path.
     private readonly streamRegistry?: AiChatStreamRegistryService,
     private readonly environment?: EnvironmentService,
+    // #492: reconstruct a #492 mid-run record's parts from the steps table before
+    // returning rows to the client / export. OPTIONAL so positional controller
+    // specs compile unchanged; when absent, hydration is skipped (old-era rows
+    // already carry inline parts, so nothing to reconstruct).
+    private readonly aiChatRunStepRepo?: AiChatRunStepRepo,
   ) {}
 
+  /**
+   * Reconstruct parts for any assistant rows that don't carry them INLINE — a
+   * #492 mid-run record whose per-step parts live in `ai_chat_run_steps` (the
+   * append-persist backend). Every FINISHED row (old-era + #492) and every old-era
+   * streaming snapshot already has inline `metadata.parts`, so the common path
+   * fetches NOTHING and returns the rows untouched; only an actively-streaming
+   * new-style row triggers the batch step fetch. Consumers (seed/poll/export) read
+   * `metadata.parts` off the returned rows exactly as before — the era switch is
+   * invisible to them (reconstructRunParts contract).
+   */
+  private async withReconstructedParts(
+    rows: AiChatMessage[],
+    workspaceId: string,
+  ): Promise<AiChatMessage[]> {
+    if (!this.aiChatRunStepRepo) return rows;
+    const needy = rows.filter(
+      (r) => r.role === 'assistant' && !rowHasInlineParts(r),
+    );
+    if (needy.length === 0) return rows;
+    const stepsByMessage = await this.aiChatRunStepRepo.findByMessageIds(
+      needy.map((r) => r.id),
+      workspaceId,
+    );
+    return hydrateAssistantParts(rows, stepsByMessage);
+  }
+
   /** List the requesting user's chats in this workspace (paginated). */
   @HttpCode(HttpStatus.OK)
   @Post('chats')
@@ -184,11 +218,17 @@ export class AiChatController {
     @AuthWorkspace() workspace: Workspace,
   ) {
     await this.assertOwnedChat(dto.chatId, user, workspace);
-    return this.aiChatMessageRepo.findByChat(
+    const page = await this.aiChatMessageRepo.findByChat(
       dto.chatId,
       workspace.id,
       pagination,
     );
+    // #492: reconstruct parts for any active new-style row so the client seed sees
+    // `metadata.parts` unchanged (a no-op for the finished rows that fill a page).
+    return {
+      ...page,
+      items: await this.withReconstructedParts(page.items, workspace.id),
+    };
   }
 
   /**
@@ -225,7 +265,10 @@ export class AiChatController {
       workspace.id,
     );
     return {
-      rows,
+      // #492: the delta of an actively-streaming new-style row carries its parts
+      // reconstructed from the steps table, so the degraded poll shows persisted
+      // progress exactly as the pre-#492 full-row snapshot did.
+      rows: await this.withReconstructedParts(rows, workspace.id),
       cursor,
       run: run ? { id: run.id, status: run.status } : null,
     };
@@ -247,8 +290,10 @@ export class AiChatController {
     @AuthWorkspace() workspace: Workspace,
   ): Promise<{ markdown: string }> {
     const chat = await this.assertOwnedChat(dto.chatId, user, workspace);
-    const rows = await this.aiChatMessageRepo.findAllByChat(
+    const rows = await this.withReconstructedParts(
-      dto.chatId,
+      await this.aiChatMessageRepo.findAllByChat(dto.chatId, workspace.id),
+      // #492: an interrupted-but-still-active turn exports its persisted steps
+      // (reconstructed from the steps table) just like the pre-#492 full row did.
       workspace.id,
     );
     const markdown = buildChatMarkdown({
@@ -288,7 +333,13 @@ export class AiChatController {
           workspace.id,
         )
       : undefined;
-    return { run, message: message ?? null };
+    // #492: reconnect to an IN-FLIGHT run reconstructs the projection row's parts
+    // from the steps table (the row itself carries only the step marker mid-run);
+    // a finished run's row already has inline parts, so this is a no-op.
+    const [hydrated] = message
+      ? await this.withReconstructedParts([message], workspace.id)
+      : [undefined];
+    return { run, message: hydrated ?? null };
   }
 
   /**

apps/server/src/core/ai-chat/ai-chat.service.ts

@@ -22,6 +22,7 @@ import { AiSettingsService } from '../../integrations/ai/ai-settings.service';
 import { describeProviderError } from '../../integrations/ai/ai-error.util';
 import { AiChatRepo } from '@docmost/db/repos/ai-chat/ai-chat.repo';
 import { AiChatMessageRepo } from '@docmost/db/repos/ai-chat/ai-chat-message.repo';
+import { AiChatRunStepRepo } from '@docmost/db/repos/ai-chat/ai-chat-run-step.repo';
 import { AiChatPageSnapshotRepo } from '@docmost/db/repos/ai-chat/ai-chat-page-snapshot.repo';
 import { AiAgentRoleRepo } from '@docmost/db/repos/ai-agent-roles/ai-agent-roles.repo';
 import { PageRepo } from '@docmost/db/repos/page/page.repo';
@@ -518,6 +519,12 @@ export class AiChatService implements OnModuleInit, OnModuleDestroy {
     // constructions compile unchanged; Nest always injects the real singleton, so
     // reconcile sees the SAME in-memory active/zombie maps the runner mutates.
     private readonly aiChatRunService?: AiChatRunService,
+    // #492 append-persist: per-step INSERT into the lightweight steps table (the
+    // O(Σ steps) replacement for the O(n²) full-row `metadata.parts` rewrite).
+    // OPTIONAL so existing positional constructions (int-specs) compile unchanged;
+    // Nest injects the real singleton. When ABSENT the per-step path falls back to
+    // the pre-#492 full-row flush (no regression, only no WAL win).
+    private readonly aiChatRunStepRepo?: AiChatRunStepRepo,
   ) {}
 
   // #487: periodic reconcile timer (single-process phase 1). Started in
@@ -1114,8 +1121,34 @@ export class AiChatService implements OnModuleInit, OnModuleDestroy {
         chatId,
         workspace.id,
       );
+      // #492: HYDRATE needy assistant rows from the steps table BEFORE the replay
+      // map. A #492 mid-run assistant row carries only a step marker
+      // (metadata.parts:[]); its real per-step parts live in `ai_chat_run_steps`.
+      // The graceful terminal callbacks (onFinish/onError/onAbort -> flushAssistant)
+      // assemble the full inline parts, so a normally-ended turn already has them.
+      // But a HARD crash mid-run (SIGKILL/OOM) fires NO terminal callback, so the
+      // row stays parts:[]; without this, rowToUiMessage falls back to an empty
+      // text part and the partial tool-calls/results/text — durable in the steps
+      // table — would DROP OUT of the model's replay context (regressing #183
+      // step-granular durability for the model consumer). Mirrors the controller's
+      // withReconstructedParts EXACTLY (same needy predicate + hydration helper).
+      // Guarded on the optional repo: absent (positional test builds) degrades to
+      // the current behavior rather than crashing.
+      let replayHistory = oldHistory;
+      if (this.aiChatRunStepRepo) {
+        const needy = oldHistory.filter(
+          (r) => r.role === 'assistant' && !rowHasInlineParts(r),
+        );
+        if (needy.length > 0) {
+          const stepsByMessage = await this.aiChatRunStepRepo.findByMessageIds(
+            needy.map((r) => r.id),
+            workspace.id,
+          );
+          replayHistory = hydrateAssistantParts(oldHistory, stepsByMessage);
+        }
+      }
       const uiMessages: Array<Omit<UIMessage, 'id'> & { id: string }> = [
-        ...oldHistory.map(rowToUiMessage),
+        ...replayHistory.map(rowToUiMessage),
         {
           id: 'pending-user',
           role: 'user',
@@ -1154,7 +1187,9 @@ export class AiChatService implements OnModuleInit, OnModuleDestroy {
       // hint — confirm it against the persisted history (the preceding assistant
       // turn must really be aborted/streaming) so a spoofed flag cannot inject the
       // interrupt note onto an ordinary turn. The partial output the model needs is
-      // already in `messages` (the aborted assistant row replays via findRecent).
+      // already in `messages`: a #492 mid-run row's per-step parts live only in the
+      // `ai_chat_run_steps` table and were hydrated into the replay history above,
+      // so the aborted assistant turn replays WITH its partial parts intact.
       // Append the new user turn (shape-only) so index -2 is the prior assistant.
       const interrupted = isInterruptResume(
         [...oldHistory, { role: 'user', status: null, metadata: null }],
@@ -1559,17 +1594,57 @@ export class AiChatService implements OnModuleInit, OnModuleDestroy {
         // connection when finalize runs, so the SQL `WHERE status='streaming'`
         // (not this flag) is what prevents it clobbering the terminal row.
         if (finalized) return null;
-        // Build the flush ONCE so the returned count is EXACTLY the persisted
+        // The count derives from capturedSteps.length at THIS instant, so the
-        // `stepsPersisted` (both derive from capturedSteps.length at this instant).
+        // returned value is EXACTLY the persisted `stepsPersisted` the ring rotates
-        const flushed = flushAssistant(capturedSteps, '', 'streaming', {
+        // on (whether we take the append-persist path or the legacy fallback).
-          pageChanged,
+        const stepsPersisted = capturedSteps.length;
-          partsCache,
-        });
-        const stepsPersisted = flushed.metadata.stepsPersisted as number;
         try {
-          await this.aiChatMessageRepo.update(assistantId, workspace.id, flushed, {
+          if (this.aiChatRunStepRepo) {
-            onlyIfStreaming: true,
+            // #492 APPEND-PERSIST: write only THIS finished step's parts to the
-          });
+            // steps table (O(step) WAL), then bump the row's CHEAP step marker —
+            // NO growing `metadata.parts` blob (that O(n²) full-row rewrite is
+            // exactly what this removes). The full `metadata.parts` is assembled
+            // once at finalize; a mid-run resume seed is reconstructed from the
+            // step rows (reconstructRunParts). The INSERT is idempotent
+            // (ON CONFLICT DO NOTHING), so a re-fired step never doubles the parts.
+            const index = stepsPersisted - 1;
+            if (index >= 0) {
+              const stepParts = assistantParts(
+                [capturedSteps[index]],
+                '',
+                partsCache,
+              );
+              await this.aiChatRunStepRepo.insertStep(
+                assistantId,
+                workspace.id,
+                index,
+                stepParts,
+              );
+            }
+            // Marker UPDATE: advance stepsPersisted + keep the toolTrace era marker
+            // (bumps updatedAt so the delta poll observes the step, and carries the
+            // frontier a resuming client attaches from). Scoped onlyIfStreaming so a
+            // late marker never clobbers the terminal finalize.
+            await this.aiChatMessageRepo.update(
+              assistantId,
+              workspace.id,
+              { metadata: stepMarkerMetadata(stepsPersisted) },
+              { onlyIfStreaming: true },
+            );
+          } else {
+            // Legacy fallback (no steps table wired — positional test builds): the
+            // pre-#492 full-row flush, so parts still land inline on the row.
+            const flushed = flushAssistant(capturedSteps, '', 'streaming', {
+              pageChanged,
+              partsCache,
+            });
+            await this.aiChatMessageRepo.update(
+              assistantId,
+              workspace.id,
+              flushed,
+              { onlyIfStreaming: true },
+            );
+          }
           return stepsPersisted;
         } catch (err) {
           this.logger.warn(
@@ -2749,6 +2824,122 @@ export function rowToUiMessage(row: AiChatMessage): Omit<UIMessage, 'id'> & {
   return { id: row.id, role, parts: parts as UIMessage['parts'] };
 }
 
+/**
+ * Cheap step-marker metadata for the #492 per-step UPDATE. Advances
+ * `stepsPersisted` (the resume attach frontier) and keeps the `toolTraceVersion`
+ * era marker, WITHOUT the growing `parts` blob (those live in the steps table
+ * now; the full `metadata.parts` is assembled once at finalize by flushAssistant).
+ * `parts: []` is kept for shape stability — it reads as an empty inline-parts row,
+ * which is exactly the discriminator that routes reconstruction to the steps table.
+ */
+export function stepMarkerMetadata(
+  stepsPersisted: number,
+): Record<string, unknown> {
+  return { parts: [], toolTraceVersion: 2, stepsPersisted };
+}
+
+/**
+ * Whether an assistant row already carries its full UI parts INLINE on the row
+ * (`metadata.parts`). TRUE for every FINISHED row — old-era rows AND #492 rows,
+ * whose full parts are assembled once at finalize — and for old-era streaming
+ * snapshots (the pre-#492 per-step full-row flush). FALSE for a #492 MID-RUN
+ * record, whose per-step parts live in the `ai_chat_run_steps` table. This is the
+ * era discriminator the reconstruct seam branches on — no schema flag needed.
+ */
+export function rowHasInlineParts(row: { metadata?: unknown }): boolean {
+  const meta = (row.metadata ?? {}) as { parts?: unknown };
+  return Array.isArray(meta.parts) && meta.parts.length > 0;
+}
+
+/**
+ * Concatenate persisted per-step parts (in `stepIndex` order) into the turn's UI
+ * parts (#492). Reproduces EXACTLY what flushAssistant → assistantParts would have
+ * written to `metadata.parts` for those finished steps, since each step row stored
+ * `assistantParts([step])` at persist time.
+ */
+export function assembleStepParts(
+  stepRows: ReadonlyArray<{ stepIndex: number; parts: unknown }>,
+): UIMessage['parts'] {
+  const parts: Array<Record<string, unknown>> = [];
+  for (const step of [...stepRows].sort((a, b) => a.stepIndex - b.stepIndex)) {
+    if (Array.isArray(step.parts)) {
+      parts.push(...(step.parts as Array<Record<string, unknown>>));
+    }
+  }
+  return parts as UIMessage['parts'];
+}
+
+/**
+ * reconstructRunParts (#492) — the single backend-switch seam. Given an assistant
+ * ROW and its persisted step rows, return the turn's UI `parts` + the persisted
+ * step count, reading from the ROW when it already carries inline parts (old-era
+ * records AND every finished record) and from the STEPS TABLE otherwise (a #492
+ * mid-run record). The higher-level consumers (attach seed, delta poll, export)
+ * route their row→parts through this / {@link hydrateAssistantParts}, so old and
+ * new records reconstruct identically WITHOUT the consumers branching on the era.
+ */
+export function reconstructRunParts(
+  row: { metadata?: unknown; content?: string | null },
+  stepRows: ReadonlyArray<{ stepIndex: number; parts: unknown }>,
+): { parts: UIMessage['parts']; stepsPersisted: number } {
+  if (rowHasInlineParts(row)) {
+    const meta = row.metadata as {
+      parts: UIMessage['parts'];
+      stepsPersisted?: number;
+    };
+    return {
+      parts: meta.parts,
+      stepsPersisted:
+        typeof meta.stepsPersisted === 'number'
+          ? meta.stepsPersisted
+          : stepRows.length,
+    };
+  }
+  if (stepRows.length > 0) {
+    return {
+      parts: assembleStepParts(stepRows),
+      stepsPersisted: stepRows.length,
+    };
+  }
+  // No inline parts and no step rows: an old-era seed / empty streaming row. Fall
+  // back to a single text part from `content` (mirrors rowToUiMessage).
+  return {
+    parts: textPart(row.content ?? '') as UIMessage['parts'],
+    stepsPersisted: 0,
+  };
+}
+
+/**
+ * Fill each assistant row's `metadata.parts` from its step rows when the row does
+ * not already carry them inline (a #492 mid-run record), so a consumer that reads
+ * `metadata.parts` off the RAW row (the client seed/poll, the Markdown export)
+ * sees the reconstructed parts with NO change to itself. Rows that already have
+ * inline parts (old-era + finished) and non-assistant rows pass through untouched.
+ * Pure: returns new row objects, never mutates the inputs.
+ */
+export function hydrateAssistantParts<
+  T extends { id: string; role?: string; metadata?: unknown },
+>(
+  rows: ReadonlyArray<T>,
+  stepsByMessage: Map<
+    string,
+    ReadonlyArray<{ stepIndex: number; parts: unknown }>
+  >,
+): T[] {
+  return rows.map((row) => {
+    if (row.role !== 'assistant' || rowHasInlineParts(row)) return row;
+    const steps = stepsByMessage.get(row.id);
+    if (!steps || steps.length === 0) return row;
+    return {
+      ...row,
+      metadata: {
+        ...((row.metadata ?? {}) as Record<string, unknown>),
+        parts: assembleStepParts(steps),
+      },
+    };
+  });
+}
+
 /**
  * The persisted-row patch shape produced by {@link flushAssistant}. It is the
  * SAME shape the assistant repo insert/update consume (content + toolCalls +

apps/server/src/database/database.module.ts

@@ -32,6 +32,7 @@ import { TemplateRepo } from '@docmost/db/repos/template/template.repo';
 import { AiChatRepo } from '@docmost/db/repos/ai-chat/ai-chat.repo';
 import { AiChatMessageRepo } from '@docmost/db/repos/ai-chat/ai-chat-message.repo';
 import { AiChatRunRepo } from '@docmost/db/repos/ai-chat/ai-chat-run.repo';
+import { AiChatRunStepRepo } from '@docmost/db/repos/ai-chat/ai-chat-run-step.repo';
 import { AiChatPageSnapshotRepo } from '@docmost/db/repos/ai-chat/ai-chat-page-snapshot.repo';
 import { AiProviderCredentialsRepo } from '@docmost/db/repos/ai-chat/ai-provider-credentials.repo';
 import { AiMcpServerRepo } from '@docmost/db/repos/ai-chat/ai-mcp-server.repo';
@@ -125,6 +126,7 @@ import { firstSqlToken } from '../integrations/metrics/metrics.constants';
     AiChatRepo,
     AiChatMessageRepo,
     AiChatRunRepo,
+    AiChatRunStepRepo,
     AiChatPageSnapshotRepo,
     AiProviderCredentialsRepo,
     AiMcpServerRepo,
@@ -161,6 +163,7 @@ import { firstSqlToken } from '../integrations/metrics/metrics.constants';
     AiChatRepo,
     AiChatMessageRepo,
     AiChatRunRepo,
+    AiChatRunStepRepo,
     AiChatPageSnapshotRepo,
     AiProviderCredentialsRepo,
     AiMcpServerRepo,

apps/server/src/database/migrations/20260708T120000-ai-chat-run-steps.ts

@@ -0,0 +1,70 @@
+import { type Kysely, sql } from 'kysely';
+
+/**
+ * `ai_chat_run_steps` — append-only per-step persistence for an assistant turn
+ * (#492 wave C). Each finished agent step's UI `parts` (its text part + a part
+ * per tool call, WITH the tool output) is INSERTed as its own lightweight row the
+ * moment the step ends, instead of REWRITING the whole assistant row's growing
+ * `metadata.parts` jsonb on every `onStepFinish`.
+ *
+ * WHY a separate table + INSERT (not a jsonb `||` append on the message row): a
+ * Postgres jsonb UPDATE rewrites the ENTIRE TOASTed row version under MVCC, so
+ * re-persisting a growing `metadata.parts` on every step is O(n²) write volume
+ * (a 50-step run with ~100 KB tool outputs wrote hundreds of MB of WAL / dead
+ * tuples per turn, hammering autovacuum). `||` would only shave the network
+ * payload — the WAL/TOAST rewrite harm remains. An INSERT into a per-step table
+ * writes ONLY that step's bytes, so the per-turn write volume is O(Σ steps).
+ *
+ * The full `metadata.parts` on the message row is assembled ONCE at finalize (the
+ * terminal completed/error/aborted write). Mid-run, a resuming client's seed is
+ * reconstructed by concatenating these step rows in `step_index` order — which
+ * reproduces exactly what the old per-step full-row rewrite persisted. Records
+ * written the OLD way (full `metadata.parts` on the row, no step rows) still
+ * reconstruct from the row unchanged; the two eras are distinguished by whether
+ * the row already carries non-empty `metadata.parts` (see reconstructRunParts /
+ * assembleStepParts in ai-chat.service.ts).
+ *
+ * ON DELETE CASCADE on `message_id`: the step rows are a derived projection of the
+ * assistant message; they must vanish with it (or with its workspace).
+ */
+export async function up(db: Kysely<any>): Promise<void> {
+  await db.schema
+    .createTable('ai_chat_run_steps')
+    .ifNotExists()
+    .addColumn('id', 'uuid', (col) =>
+      col.primaryKey().defaultTo(sql`gen_uuid_v7()`),
+    )
+    // The assistant message row this step belongs to (the #183 projection). The
+    // step rows are a derived, per-step slice of that message, so they cascade.
+    .addColumn('message_id', 'uuid', (col) =>
+      col.references('ai_chat_messages.id').onDelete('cascade').notNull(),
+    )
+    .addColumn('workspace_id', 'uuid', (col) =>
+      col.references('workspaces.id').onDelete('cascade').notNull(),
+    )
+    // 0-based index of the finished step within the turn. Ordering key for
+    // reconstruction; unique per message (idempotent step re-persist).
+    .addColumn('step_index', 'integer', (col) => col.notNull())
+    // The step's UI parts (text part + a `tool-*` part per call, WITH output).
+    // Concatenated in step order to rebuild the turn's `metadata.parts`.
+    .addColumn('parts', 'jsonb', (col) => col.notNull())
+    .addColumn('created_at', 'timestamptz', (col) =>
+      col.notNull().defaultTo(sql`now()`),
+    )
+    .execute();
+
+  // Idempotent per-step persist: a retried INSERT of the same (message, step)
+  // is a no-op (the service uses ON CONFLICT DO NOTHING). This also serves the
+  // reconstruction read (WHERE message_id ORDER BY step_index).
+  await db.schema
+    .createIndex('ai_chat_run_steps_message_step_uidx')
+    .ifNotExists()
+    .on('ai_chat_run_steps')
+    .columns(['message_id', 'step_index'])
+    .unique()
+    .execute();
+}
+
+export async function down(db: Kysely<any>): Promise<void> {
+  await db.schema.dropTable('ai_chat_run_steps').ifExists().execute();
+}

apps/server/src/database/repos/ai-chat/ai-chat-run-step.repo.ts

@@ -0,0 +1,95 @@
+import { Injectable } from '@nestjs/common';
+import { InjectKysely } from 'nestjs-kysely';
+import { KyselyDB, KyselyTransaction } from '../../types/kysely.types';
+import { dbOrTx } from '../../utils';
+import { AiChatRunStep } from '@docmost/db/types/entity.types';
+
+/**
+ * Append-only per-step persistence for an assistant turn (#492). Each finished
+ * agent step's UI `parts` (its text part + a `tool-*` part per call, WITH the
+ * tool output) is INSERTed as its own lightweight row the moment the step ends —
+ * instead of REWRITING the assistant row's growing `metadata.parts` jsonb on every
+ * `onStepFinish` (a Postgres jsonb UPDATE rewrites the whole TOASTed row version
+ * under MVCC, so that was O(n²) WAL/dead-tuple churn per turn).
+ *
+ * The full `metadata.parts` on the message row is assembled ONCE at finalize;
+ * mid-run, a resuming client's seed is rebuilt from these rows in `stepIndex`
+ * order (see `assembleStepParts` / the reconstruct seam in ai-chat.service.ts).
+ * Every method is workspace-scoped as defense-in-depth.
+ */
+@Injectable()
+export class AiChatRunStepRepo {
+  constructor(@InjectKysely() private readonly db: KyselyDB) {}
+
+  /**
+   * Append one finished step's parts. Idempotent: a retried persist of the SAME
+   * (message, stepIndex) is a no-op via ON CONFLICT DO NOTHING — the per-step
+   * writes are fired fire-and-forget + serialized, and a duplicate must never
+   * throw into the stream or double the parts. Returns whether a NEW row landed
+   * (false = the step was already persisted).
+   */
+  async insertStep(
+    messageId: string,
+    workspaceId: string,
+    stepIndex: number,
+    parts: unknown,
+    trx?: KyselyTransaction,
+  ): Promise<boolean> {
+    const db = dbOrTx(this.db, trx);
+    const inserted = await db
+      .insertInto('aiChatRunSteps')
+      .values({
+        messageId,
+        workspaceId,
+        stepIndex,
+        // jsonb column: cast through never (same pattern as the message repo).
+        parts: parts as never,
+      })
+      .onConflict((oc) => oc.columns(['messageId', 'stepIndex']).doNothing())
+      .returning('id')
+      .executeTakeFirst();
+    return inserted !== undefined;
+  }
+
+  /** All persisted steps for ONE assistant message, in step order. */
+  async findByMessage(
+    messageId: string,
+    workspaceId: string,
+  ): Promise<AiChatRunStep[]> {
+    return this.db
+      .selectFrom('aiChatRunSteps')
+      .selectAll('aiChatRunSteps')
+      .where('messageId', '=', messageId)
+      .where('workspaceId', '=', workspaceId)
+      .orderBy('stepIndex', 'asc')
+      .execute();
+  }
+
+  /**
+   * All persisted steps for a SET of assistant messages, grouped by messageId
+   * (each group in step order). One query for the batch — the hydration seam
+   * (getMessages / delta / export) calls this only for the rows that actually
+   * need reconstruction (an active new-style row whose `metadata.parts` is still
+   * empty), which is usually none, so this is skipped on the common path.
+   */
+  async findByMessageIds(
+    messageIds: string[],
+    workspaceId: string,
+  ): Promise<Map<string, AiChatRunStep[]>> {
+    const byMessage = new Map<string, AiChatRunStep[]>();
+    if (messageIds.length === 0) return byMessage;
+    const rows = await this.db
+      .selectFrom('aiChatRunSteps')
+      .selectAll('aiChatRunSteps')
+      .where('messageId', 'in', messageIds)
+      .where('workspaceId', '=', workspaceId)
+      .orderBy('stepIndex', 'asc')
+      .execute();
+    for (const row of rows) {
+      const list = byMessage.get(row.messageId);
+      if (list) list.push(row);
+      else byMessage.set(row.messageId, [row]);
+    }
+    return byMessage;
+  }
+}

apps/server/src/database/types/db.d.ts

@@ -692,6 +692,22 @@ export interface AiChatRuns {
   updatedAt: Generated<Timestamp>;
 }
 
+// Append-only per-step persistence for an assistant turn (#492). Mirrors
+// migration 20260708T120000-ai-chat-run-steps.ts. Each finished agent step's UI
+// `parts` are INSERTed as their own row (instead of rewriting the message row's
+// growing `metadata.parts` jsonb every step — an O(n²) WAL/TOAST churn). The full
+// `metadata.parts` is assembled once at finalize; mid-run a resuming client's seed
+// is rebuilt by concatenating these rows in `stepIndex` order. Cascades with the
+// assistant message row it projects.
+export interface AiChatRunSteps {
+  id: Generated<string>;
+  messageId: string;
+  workspaceId: string;
+  stepIndex: number;
+  parts: Json;
+  createdAt: Generated<Timestamp>;
+}
+
 // Per-(chat,page) snapshot of the open page's Markdown at the END of the agent's
 // previous turn (#274). Mirrors migration 20260702T120000-ai-chat-page-snapshot.ts.
 // The next turn diffs the CURRENT Markdown against `contentMd` to surface edits a
@@ -729,6 +745,7 @@ export interface DB {
   aiChats: AiChats;
   aiChatMessages: AiChatMessages;
   aiChatRuns: AiChatRuns;
+  aiChatRunSteps: AiChatRunSteps;
   aiChatPageSnapshots: AiChatPageSnapshots;
   apiKeys: ApiKeys;
   attachments: Attachments;

apps/server/src/database/types/entity.types.ts

@@ -4,6 +4,7 @@ import {
   AiChats,
   AiChatMessages,
   AiChatRuns,
+  AiChatRunSteps,
   AiChatPageSnapshots,
   Attachments,
   Comments,
@@ -64,6 +65,12 @@ export type InsertableAiChatMessage = Omit<Insertable<AiChatMessages>, 'tsv'>;
 export type AiChatRun = Selectable<AiChatRuns>;
 export type InsertableAiChatRun = Insertable<AiChatRuns>;
 
+// AI Chat Run Step (#492): append-only per-step parts persistence. Each finished
+// agent step's UI parts are stored as their own row; the full turn's parts are
+// assembled from these (in stepIndex order) for a mid-run resume seed.
+export type AiChatRunStep = Selectable<AiChatRunSteps>;
+export type InsertableAiChatRunStep = Insertable<AiChatRunSteps>;
+
 // AI Chat Page Snapshot (#274): per-(chat,page) Markdown snapshot taken at the
 // end of the agent's previous turn, diffed against the current page next turn to
 // detect human edits made between turns.

apps/server/src/integrations/static/static.module.ts

@@ -4,7 +4,6 @@ import { join } from 'path';
 import * as fs from 'node:fs';
 import fastifyStatic from '@fastify/static';
 import { EnvironmentService } from '../environment/environment.service';
-import { resolveClientDistPath } from '../../common/helpers/client-version';
 
 /**
  * Resolve the response headers for a statically served client asset.
@@ -57,7 +56,14 @@ export class StaticModule implements OnModuleInit {
     const httpAdapter = this.httpAdapterHost.httpAdapter;
     const app = httpAdapter.getInstance();
 
-    const clientDistPath = resolveClientDistPath();
+    const clientDistPath = join(
+      __dirname,
+      '..',
+      '..',
+      '..',
+      '..',
+      'client/dist',
+    );
 
     const indexFilePath = join(clientDistPath, 'index.html');
 

apps/server/src/ws/ws.gateway.ts

@@ -9,14 +9,10 @@ import {
 import { Server, Socket } from 'socket.io';
 import { TokenService } from '../core/auth/services/token.service';
 import { JwtPayload, JwtType } from '../core/auth/dto/jwt-payload';
-import { Logger, OnModuleDestroy, OnModuleInit } from '@nestjs/common';
+import { OnModuleDestroy } from '@nestjs/common';
 import { SpaceMemberRepo } from '@docmost/db/repos/space/space-member.repo';
 import { WsService } from './ws.service';
 import { getSpaceRoomName, getUserRoomName } from './ws.utils';
-import {
-  readClientBuildVersion,
-  resolveClientDistPath,
-} from '../common/helpers/client-version';
 import * as cookie from 'cookie';
 
 @WebSocketGateway({
@@ -24,40 +20,17 @@ import * as cookie from 'cookie';
   transports: ['websocket'],
 })
 export class WsGateway
-  implements
+  implements OnGatewayConnection, OnGatewayInit, OnModuleDestroy
-    OnGatewayConnection,
-    OnGatewayInit,
-    OnModuleInit,
-    OnModuleDestroy
 {
   @WebSocketServer()
   server: Server;
 
-  private readonly logger = new Logger(WsGateway.name);
-
-  // The build version of the client bundle shipped in this image, read once at
-  // startup from client/dist/version.json (single source of truth, same value
-  // baked into the client's APP_VERSION). Empty string => version.json missing
-  // or empty => the proactive version-coherence reload feature stays inert.
-  private appVersion = '';
-
   constructor(
     private tokenService: TokenService,
     private spaceMemberRepo: SpaceMemberRepo,
     private wsService: WsService,
   ) {}
 
-  onModuleInit(): void {
-    this.appVersion = readClientBuildVersion(resolveClientDistPath());
-    if (this.appVersion) {
-      this.logger.log(`app-version reload: ACTIVE (v=${this.appVersion})`);
-    } else {
-      this.logger.log(
-        'app-version reload: DISABLED (version.json missing/empty)',
-      );
-    }
-  }
-
   afterInit(server: Server): void {
     this.wsService.setServer(server);
   }
@@ -82,14 +55,6 @@ export class WsGateway
       const spaceRooms = userSpaceIds.map((id) => getSpaceRoomName(id));
 
       client.join([userRoom, workspaceRoom, ...spaceRooms]);
-
-      // Announce this container's client build version to the freshly
-      // authenticated socket. On a redeploy the client reconnects to the new
-      // container and receives the new version here, letting it guard-reload
-      // before it hits a stale lazy chunk. Per-connect only (no broadcast):
-      // natural reconnect covers both single-container and cluster without a
-      // thundering-herd fleet reload.
-      client.emit('app-version', { version: this.appVersion });
     } catch (err) {
       client.emit('Unauthorized');
       client.disconnect();

apps/server/test/integration/ai-chat-append-persist-service.int-spec.ts

@@ -0,0 +1,412 @@
+import * as http from 'node:http';
+import { Kysely } from 'kysely';
+import { tool } from 'ai';
+import { z } from 'zod';
+import { MockLanguageModelV3, convertArrayToReadableStream } from 'ai/test';
+import { AiChatRepo } from '@docmost/db/repos/ai-chat/ai-chat.repo';
+import { AiChatMessageRepo } from '@docmost/db/repos/ai-chat/ai-chat-message.repo';
+import { AiChatRunStepRepo } from '@docmost/db/repos/ai-chat/ai-chat-run-step.repo';
+import {
+  AiChatService,
+  assembleStepParts,
+  assistantParts,
+  rowHasInlineParts,
+  stepMarkerMetadata,
+} from 'src/core/ai-chat/ai-chat.service';
+import {
+  getTestDb,
+  destroyTestDb,
+  createWorkspace,
+  createUser,
+  createChat,
+  createMessage,
+} from './db';
+
+/**
+ * #492 append-persist — the REAL onStep WRITE path (F2) and the model-REPLAY
+ * hydration path (F1), driven through `AiChatService.stream` against a LIVE
+ * Postgres with a REAL `AiChatRunStepRepo` INJECTED. The existing append-persist
+ * int-specs hand-roll the insert+marker cycle via the repos directly and build
+ * the service with `aiChatRunStepRepo: undefined` (only the legacy-fallback branch
+ * is covered), so an off-by-one on `stepsPersisted-1`, a wrong `capturedSteps`
+ * slice, or a broken marker payload would pass all of them. These tests exercise
+ * the actual `updateStreaming` append-persist branch end to end.
+ *
+ * The seam is the injected `model` (a seeded `MockLanguageModelV3` from `ai/test`)
+ * plus a REAL Node `ServerResponse` as the hijacked socket — mirrors
+ * ai-chat-stream.int-spec.ts.
+ */
+
+const sleep = (ms: number) => new Promise((r) => setTimeout(r, ms));
+
+async function waitFor(
+  cond: () => Promise<boolean> | boolean,
+  { timeoutMs = 15_000, stepMs = 25 } = {},
+): Promise<void> {
+  const start = Date.now();
+  while (Date.now() - start < timeoutMs) {
+    if (await cond()) return;
+    await sleep(stepMs);
+  }
+  throw new Error('waitFor: condition not met within timeout');
+}
+
+// A real Node ServerResponse wired to a live socket (identical helper to the
+// stream int-spec) so the SDK's pipe/heartbeat writes behave as in prod.
+function makeRealResponse(): Promise<{
+  res: http.ServerResponse;
+  cleanup: () => Promise<void>;
+}> {
+  return new Promise((resolve) => {
+    const server = http.createServer((_req, res) => {
+      resolve({
+        res,
+        cleanup: () =>
+          new Promise<void>((done) => {
+            try {
+              if (!res.writableEnded) res.end();
+            } catch {
+              /* socket already gone */
+            }
+            server.close(() => done());
+          }),
+      });
+    });
+    server.listen(0, () => {
+      const port = (server.address() as any).port;
+      const creq = http.request({ port, method: 'GET' }, (cres) => {
+        cres.resume();
+      });
+      creq.on('error', () => undefined);
+      creq.end();
+    });
+  });
+}
+
+// Stream parts for a normal, successful single-step turn.
+function successStream() {
+  return convertArrayToReadableStream([
+    { type: 'stream-start', warnings: [] },
+    { type: 'text-start', id: 't1' },
+    { type: 'text-delta', id: 't1', delta: 'Hello' },
+    { type: 'text-delta', id: 't1', delta: ' there' },
+    { type: 'text-end', id: 't1' },
+    {
+      type: 'finish',
+      finishReason: 'stop',
+      usage: { inputTokens: 10, outputTokens: 5, totalTokens: 15 },
+    },
+  ] as any);
+}
+
+// A THREE-step turn: steps 0 and 1 each emit text + an `echo` tool call (the SDK
+// runs the tool and continues); step 2 answers and stops. Three steps is
+// deliberate: the LAST finished step's append-persist write races the terminal
+// finalize (which writes the full inline parts anyway, so a lost last-step row is
+// by design), but the NON-final steps 0 and 1 always drain to the steps table
+// before finalize — so those are what the test asserts on deterministically.
+function threeStepModel(): MockLanguageModelV3 {
+  let step = 0;
+  const toolStep = (i: number) => ({
+    stream: convertArrayToReadableStream([
+      { type: 'stream-start', warnings: [] },
+      { type: 'text-start', id: `s${i}` },
+      { type: 'text-delta', id: `s${i}`, delta: `step ${i} ` },
+      { type: 'text-end', id: `s${i}` },
+      {
+        type: 'tool-call',
+        toolCallId: `c${i}`,
+        toolName: 'echo',
+        input: JSON.stringify({ msg: `m${i}` }),
+      },
+      {
+        type: 'finish',
+        finishReason: 'tool-calls',
+        usage: { inputTokens: 5, outputTokens: 3, totalTokens: 8 },
+      },
+    ] as any),
+  });
+  return new MockLanguageModelV3({
+    doStream: async () => {
+      const n = step++;
+      // Realistic inter-step latency. A real model spends seconds per step, so the
+      // fire-and-forget per-step write chain drains to the steps table BETWEEN
+      // steps; the mock otherwise collapses all steps into microseconds and the
+      // terminal finalize wins the race before any but the first step persists.
+      if (n > 0) await sleep(200);
+      if (n < 2) return toolStep(n);
+      return {
+        stream: convertArrayToReadableStream([
+          { type: 'stream-start', warnings: [] },
+          { type: 'text-start', id: 's2' },
+          { type: 'text-delta', id: 's2', delta: 'final answer' },
+          { type: 'text-end', id: 's2' },
+          {
+            type: 'finish',
+            finishReason: 'stop',
+            usage: { inputTokens: 6, outputTokens: 4, totalTokens: 10 },
+          },
+        ] as any),
+      };
+    },
+  } as any);
+}
+
+describe('#492 append-persist service paths [integration]', () => {
+  let db: Kysely<any>;
+  let aiChatRepo: AiChatRepo;
+  let msgRepo: AiChatMessageRepo;
+  let stepRepo: AiChatRunStepRepo;
+  let workspaceId: string;
+  let userId: string;
+
+  let closeCalls: number;
+  const mcpClients = {
+    toolsFor: async () => ({
+      tools: {},
+      clients: [
+        {
+          close: async () => {
+            closeCalls += 1;
+          },
+        },
+      ],
+      outcomes: [],
+      instructions: [],
+    }),
+  };
+
+  // Build the service WITH a REAL AiChatRunStepRepo injected (the property under
+  // test) — unlike the legacy-fallback harness that passes it as undefined.
+  const echoTool = tool({
+    description: 'echo the message back',
+    inputSchema: z.object({ msg: z.string() }),
+    execute: async ({ msg }) => ({ echoed: msg }),
+  });
+
+  function buildService(): AiChatService {
+    return new AiChatService(
+      { getChatModel: async () => null } as any,
+      aiChatRepo,
+      msgRepo,
+      {} as any, // aiChatPageSnapshotRepo
+      { resolve: async () => null } as any, // aiSettings
+      { forUser: async () => ({ echo: echoTool }) } as any, // tools
+      mcpClients as any,
+      {} as any, // aiAgentRoleRepo
+      {} as any, // pageRepo
+      {} as any, // pageAccess
+      {
+        isAiChatDeferredToolsEnabled: () => false,
+        isAiChatFinalStepLockdownEnabled: () => false,
+      } as any, // environment (deferred OFF -> all tools active every step)
+      undefined, // streamRegistry
+      undefined, // aiChatRunService
+      stepRepo, // #492 aiChatRunStepRepo — the append-persist backend
+    );
+  }
+
+  function userUiMessage(text: string) {
+    return {
+      id: `u-${Math.random()}`,
+      role: 'user',
+      parts: [{ type: 'text', text }],
+    };
+  }
+
+  async function runStream(opts: {
+    model: MockLanguageModelV3;
+    chatId: string;
+    body: any;
+  }): Promise<void> {
+    closeCalls = 0;
+    const service = buildService();
+    const { res, cleanup } = await makeRealResponse();
+    try {
+      await service.stream({
+        user: { id: userId, workspaceId } as any,
+        workspace: { id: workspaceId, name: 'WS' } as any,
+        sessionId: 'sess-1',
+        body: opts.body,
+        res: { raw: res } as any,
+        signal: new AbortController().signal,
+        model: opts.model as any,
+        role: null,
+      } as any);
+      await waitFor(async () => {
+        const rows = await msgRepo.findAllByChat(opts.chatId, workspaceId);
+        return rows.some(
+          (r) =>
+            r.role === 'assistant' &&
+            ['completed', 'error', 'aborted'].includes(r.status as string),
+        );
+      });
+      await waitFor(() => closeCalls > 0, { timeoutMs: 5_000 });
+    } finally {
+      await cleanup();
+    }
+  }
+
+  beforeAll(async () => {
+    db = getTestDb();
+    aiChatRepo = new AiChatRepo(db as any);
+    msgRepo = new AiChatMessageRepo(db as any);
+    stepRepo = new AiChatRunStepRepo(db as any);
+    workspaceId = (await createWorkspace(db)).id;
+    userId = (await createUser(db, workspaceId)).id;
+  });
+
+  afterAll(async () => {
+    await destroyTestDb();
+  });
+
+  // --- F2: the real onStep append-persist WRITE branch -----------------------
+  it('drives steps through the real onStep path: per-step rows + marker match a single-row flush', async () => {
+    const chatId = (await createChat(db, { workspaceId, creatorId: userId })).id;
+    const model = threeStepModel();
+
+    // Capture the mid-run step-marker UPDATEs the append-persist branch writes on
+    // the assistant row (a { parts: [], toolTraceVersion, stepsPersisted } patch).
+    const updateSpy = jest.spyOn(msgRepo, 'update');
+    try {
+      await runStream({
+        model,
+        chatId,
+        body: { chatId, messages: [userUiMessage('call the tool then answer')] },
+      });
+
+      const rows = await msgRepo.findAllByChat(chatId, workspaceId);
+      const assistant = rows.find((r) => r.role === 'assistant')!;
+      expect(assistant).toBeDefined();
+      expect(assistant.status).toBe('completed');
+      // The turn finalizes with the FULL inline parts assembled by a single-row
+      // flush (assistantParts over every step) — the baseline the per-step slices
+      // must reproduce.
+      expect(rowHasInlineParts(assistant)).toBe(true);
+      const finalParts = (assistant.metadata as { parts: any[] }).parts;
+
+      // The two NON-final finished steps each landed their own row, in stepIndex
+      // order. (The fire-and-forget write chain drains before the next step, so
+      // poll until both are on disk; the LAST step's write may lose the finalize
+      // race, which is by design — its parts are already in `finalParts`.)
+      await waitFor(async () => {
+        const s = await stepRepo.findByMessage(assistant.id, workspaceId);
+        return s.length >= 2;
+      });
+      const steps = await stepRepo.findByMessage(assistant.id, workspaceId);
+      expect(steps[0].stepIndex).toBe(0);
+      expect(steps[1].stepIndex).toBe(1);
+
+      // Each per-step row carries a NON-trivial slice: this step's text part + its
+      // paired tool part (guards a mutation that persists empty/whole-turn parts).
+      const s0 = steps[0].parts as any[];
+      expect(s0).toContainEqual({ type: 'text', text: 'step 0 ' });
+      expect(s0.some((p) => p.type === 'tool-echo')).toBe(true);
+
+      // The per-step slices are EXACTLY the corresponding prefix of the single-row
+      // flush: assembleStepParts([step0, step1]) === finalParts[0 .. len0+len1].
+      // This is what an off-by-one on `stepsPersisted-1` (a wrong `capturedSteps`
+      // slice) or a shifted stepIndex breaks — the prefix no longer aligns.
+      const prefixLen =
+        (steps[0].parts as any[]).length + (steps[1].parts as any[]).length;
+      expect(assembleStepParts([steps[0], steps[1]] as any)).toEqual(
+        finalParts.slice(0, prefixLen),
+      );
+
+      // The mid-run step markers advanced 1 -> 2 -> ... (the resume frontier), each
+      // a shape-stable empty-parts marker equal to a single-row flush's marker.
+      const markerCounts = updateSpy.mock.calls
+        .map((c) => (c[2] as any)?.metadata)
+        .filter(
+          (m) =>
+            m &&
+            Array.isArray(m.parts) &&
+            m.parts.length === 0 &&
+            typeof m.stepsPersisted === 'number',
+        )
+        .map((m) => m.stepsPersisted);
+      // Monotonic from 1, covering at least the two non-final steps.
+      expect(markerCounts.slice(0, 2)).toEqual([1, 2]);
+      expect(
+        updateSpy.mock.calls
+          .map((c) => (c[2] as any)?.metadata)
+          .find((m) => m && m.stepsPersisted === 2),
+      ).toEqual(stepMarkerMetadata(2));
+    } finally {
+      updateSpy.mockRestore();
+    }
+  }, 60_000);
+
+  // --- F1: model-REPLAY hydrates a hard-crashed mid-run turn from the steps table
+  it('replays a hard-crashed mid-run turn WITH its partial steps hydrated from the steps table', async () => {
+    const chatId = (await createChat(db, { workspaceId, creatorId: userId })).id;
+
+    // Prior turn: a genuine user question...
+    await createMessage(db, {
+      workspaceId,
+      chatId,
+      userId,
+      role: 'user',
+      content: 'What is in the design doc?',
+      createdAt: new Date(Date.now() - 3000),
+    });
+    // ...and an assistant row that a HARD crash (SIGKILL/OOM) left mid-run: only a
+    // step marker on the row (metadata.parts:[] , content:''), NO terminal
+    // callback ever fired, so its real parts live ONLY in ai_chat_run_steps.
+    const crashed = await createMessage(db, {
+      workspaceId,
+      chatId,
+      role: 'assistant',
+      content: '',
+      status: 'aborted',
+      metadata: stepMarkerMetadata(1),
+      createdAt: new Date(Date.now() - 2000),
+    });
+    // The durable partial step: some reasoning text + a completed getPage tool
+    // call (input + output), exactly what #183 step-granular durability preserves.
+    await stepRepo.insertStep(
+      crashed.id,
+      workspaceId,
+      0,
+      assistantParts(
+        [
+          {
+            text: 'HYDRATED_PARTIAL_STEP the doc says',
+            toolCalls: [
+              { toolCallId: 'g1', toolName: 'getPage', input: { id: 'p1' } },
+            ],
+            toolResults: [
+              {
+                toolCallId: 'g1',
+                toolName: 'getPage',
+                output: { id: 'p1', body: 'PARTIAL_TOOL_OUTPUT budget section' },
+              },
+            ],
+          } as any,
+        ],
+        '',
+      ),
+    );
+
+    // The NEXT turn: the model just answers. The service must REPLAY the crashed
+    // assistant turn with its partial parts hydrated from the steps table.
+    const model = new MockLanguageModelV3({
+      doStream: async () => ({ stream: successStream() }),
+    } as any);
+    await runStream({
+      model,
+      chatId,
+      body: { chatId, messages: [userUiMessage('Continue please')] },
+    });
+
+    expect(model.doStreamCalls.length).toBeGreaterThan(0);
+    const prompt = JSON.stringify(model.doStreamCalls[0].prompt);
+    // The partial step's TEXT reached the model context (it would be an empty text
+    // part without hydration — rowToUiMessage falls back to `content:''`).
+    expect(prompt).toContain('HYDRATED_PARTIAL_STEP');
+    // The partial TOOL RESULT survived too (durable in the steps table, replayed).
+    expect(prompt).toContain('PARTIAL_TOOL_OUTPUT');
+    // The genuine prior user turn is present as well (sanity: real history replay).
+    expect(prompt).toContain('What is in the design doc?');
+  }, 60_000);
+});

apps/server/test/integration/ai-chat-append-persist-wal.int-spec.ts

@@ -0,0 +1,173 @@
+import { randomBytes } from 'crypto';
+import { Kysely, sql } from 'kysely';
+import { AiChatRunStepRepo } from '@docmost/db/repos/ai-chat/ai-chat-run-step.repo';
+import { AiChatMessageRepo } from '@docmost/db/repos/ai-chat/ai-chat-message.repo';
+import {
+  assistantParts,
+  flushAssistant,
+  stepMarkerMetadata,
+} from '../../src/core/ai-chat/ai-chat.service';
+import {
+  getTestDb,
+  destroyTestDb,
+  createWorkspace,
+  createUser,
+  createChat,
+} from './db';
+
+/**
+ * #492 append-persist — WRITE-VOLUME regression on a LIVE Postgres, measured via
+ * the `pg_current_wal_lsn()` delta around a realistic multi-step run driven through
+ * the REAL repos (not a mock — a mock cannot observe MVCC/TOAST rewrite volume, the
+ * whole point). Proves the core claim:
+ *
+ *   NEW (per-step INSERT into ai_chat_run_steps + a CHEAP step-marker UPDATE on the
+ *   message row) writes O(Σ steps) of WAL — each step writes only its own bytes.
+ *
+ *   OLD (the pre-#492 full-row rewrite: re-persist the GROWING metadata.parts on
+ *   every onStepFinish) writes O(n²) — step k rewrites the whole TOASTed jsonb of
+ *   all k prior outputs.
+ *
+ * The OLD path here IS the reverted behavior, so this doubles as the mutation
+ * check: swapping the new path back to `flushAssistant` full-row UPDATEs reddens
+ * the assertion (OLD is many times larger).
+ */
+type Step = {
+  text: string;
+  toolCalls: Array<{ toolCallId: string; toolName: string; input: unknown }>;
+  toolResults: Array<{ toolCallId: string; toolName: string; output: unknown }>;
+};
+
+// ~100 KB INCOMPRESSIBLE output per step (a page read). Random base64 so TOAST
+// cannot compress it away and hide the real write volume.
+function makeStep(i: number, outputBytes = 100_000): Step {
+  const body = randomBytes(Math.ceil(outputBytes * 0.75)).toString('base64');
+  return {
+    text: `step ${i} reasoning`,
+    toolCalls: [
+      { toolCallId: `c${i}`, toolName: 'getPage', input: { id: `p${i}` } },
+    ],
+    toolResults: [
+      {
+        toolCallId: `c${i}`,
+        toolName: 'getPage',
+        output: { id: `p${i}`, title: `Page ${i}`, body },
+      },
+    ],
+  };
+}
+
+async function walDelta(
+  db: Kysely<any>,
+  fn: () => Promise<void>,
+): Promise<number> {
+  const before = (
+    await sql<{ l: string }>`select pg_current_wal_lsn() as l`.execute(db)
+  ).rows[0].l;
+  await fn();
+  // NOTE: no pg_switch_wal() — a segment switch pads the LSN to the next 16 MB
+  // boundary and would swamp the delta. The raw LSN advances by the WAL bytes.
+  const after = (
+    await sql<{ l: string }>`select pg_current_wal_lsn() as l`.execute(db)
+  ).rows[0].l;
+  return Number(
+    (
+      await sql<{
+        d: string;
+      }>`select pg_wal_lsn_diff(${after}::pg_lsn, ${before}::pg_lsn) as d`.execute(
+        db,
+      )
+    ).rows[0].d,
+  );
+}
+
+describe('#492 append-persist write volume (pg_current_wal_lsn delta) [integration]', () => {
+  let db: Kysely<any>;
+  let stepRepo: AiChatRunStepRepo;
+  let msgRepo: AiChatMessageRepo;
+  let workspaceId: string;
+  let userId: string;
+  let chatId: string;
+
+  beforeAll(async () => {
+    db = getTestDb();
+    stepRepo = new AiChatRunStepRepo(db as any);
+    msgRepo = new AiChatMessageRepo(db as any);
+    workspaceId = (await createWorkspace(db)).id;
+    userId = (await createUser(db, workspaceId)).id;
+    chatId = (await createChat(db, { workspaceId, creatorId: userId })).id;
+  });
+
+  afterAll(async () => {
+    await destroyTestDb();
+  });
+
+  const seedRow = () =>
+    msgRepo.insert({
+      chatId,
+      workspaceId,
+      userId,
+      role: 'assistant',
+      content: '',
+      status: 'streaming',
+      metadata: stepMarkerMetadata(0) as never,
+    });
+
+  const STEPS = 40;
+
+  it('NEW per-step INSERT is O(Σ steps); OLD full-row rewrite is O(n²)', async () => {
+    const steps: Step[] = [];
+    for (let i = 0; i < STEPS; i++) steps.push(makeStep(i));
+
+    // NEW: per-step INSERT of THIS step's parts + a cheap marker UPDATE.
+    const newRow = await seedRow();
+    const newWal = await walDelta(db, async () => {
+      for (let i = 0; i < STEPS; i++) {
+        await stepRepo.insertStep(
+          newRow.id,
+          workspaceId,
+          i,
+          assistantParts([steps[i]], ''),
+        );
+        await msgRepo.update(
+          newRow.id,
+          workspaceId,
+          { metadata: stepMarkerMetadata(i + 1) },
+          { onlyIfStreaming: true },
+        );
+      }
+    });
+
+    // OLD (the pre-#492 revert): re-persist the GROWING metadata.parts on the
+    // message row on every step.
+    const oldRow = await seedRow();
+    const oldWal = await walDelta(db, async () => {
+      const acc: Step[] = [];
+      for (let i = 0; i < STEPS; i++) {
+        acc.push(steps[i]);
+        await msgRepo.update(
+          oldRow.id,
+          workspaceId,
+          flushAssistant(acc as never, '', 'streaming'),
+          { onlyIfStreaming: true },
+        );
+      }
+    });
+
+    // eslint-disable-next-line no-console
+    console.log(
+      `[#492 WAL] ${STEPS} steps ×100KB: new=${(newWal / 1e6).toFixed(1)}MB ` +
+        `old=${(oldWal / 1e6).toFixed(1)}MB (${(oldWal / newWal).toFixed(
+          1,
+        )}x smaller)`,
+    );
+
+    // O(Σ steps): ~STEPS × (100KB output + marker) of WAL. 40 × ~100KB parts plus
+    // 40 tiny markers is a few tens of MB at most — bounded, linear in step count.
+    expect(newWal).toBeLessThan(30_000_000);
+    // O(n²): step k rewrites ~k × 100KB. Σ over 40 steps ≈ 80+ MB — far larger.
+    expect(oldWal).toBeGreaterThan(30_000_000);
+    // The load-bearing claim: the new path writes a small FRACTION of the old.
+    expect(newWal).toBeLessThan(oldWal * 0.35);
+  }, 120_000);
+});

apps/server/test/integration/ai-chat-hydrate-controller.int-spec.ts

@@ -0,0 +1,169 @@
+import { Kysely } from 'kysely';
+import { AiChatController } from 'src/core/ai-chat/ai-chat.controller';
+import {
+  assembleStepParts,
+  assistantParts,
+  stepMarkerMetadata,
+} from 'src/core/ai-chat/ai-chat.service';
+import { AiChatRepo } from '@docmost/db/repos/ai-chat/ai-chat.repo';
+import { AiChatMessageRepo } from '@docmost/db/repos/ai-chat/ai-chat-message.repo';
+import { AiChatRunStepRepo } from '@docmost/db/repos/ai-chat/ai-chat-run-step.repo';
+import type { User, Workspace } from '@docmost/db/types/entity.types';
+import {
+  getTestDb,
+  destroyTestDb,
+  createWorkspace,
+  createUser,
+  createChat,
+  createMessage,
+} from './db';
+
+/**
+ * #492 controller hydration (crash-before-finalize RESUME) on a LIVE Postgres.
+ * `AiChatController.withReconstructedParts` is wired into getMessages/delta/export/
+ * run, but `aiChatRunStepRepo` is OPTIONAL and every controller unit spec passes it
+ * as `undefined`, so the hydration branch early-returns and NEVER executes in those
+ * tests. This drives the real read path — a mid-run streaming row (marker only,
+ * empty inline parts) PLUS its `ai_chat_run_steps` rows — through getMessages WITH
+ * the repo present, exercising the `role==='assistant' && !rowHasInlineParts`
+ * needy predicate, the workspace-scoped batch step fetch, and the endpoint binding.
+ */
+describe('#492 controller hydration read path [integration]', () => {
+  let db: Kysely<any>;
+  let aiChatRepo: AiChatRepo;
+  let msgRepo: AiChatMessageRepo;
+  let stepRepo: AiChatRunStepRepo;
+  let workspaceId: string;
+  let otherWorkspaceId: string;
+  let userId: string;
+
+  // Build the controller WITH a real AiChatRunStepRepo injected (position 9), the
+  // seam the unit specs leave undefined. Only the read-path deps are real.
+  function buildController(): AiChatController {
+    return new AiChatController(
+      {} as any, // aiChatService
+      {} as any, // aiChatRunService
+      aiChatRepo,
+      msgRepo,
+      {} as any, // aiTranscription
+      {} as any, // pageRepo
+      undefined, // streamRegistry
+      undefined, // environment
+      stepRepo, // #492 aiChatRunStepRepo
+    );
+  }
+
+  beforeAll(async () => {
+    db = getTestDb();
+    aiChatRepo = new AiChatRepo(db as any);
+    msgRepo = new AiChatMessageRepo(db as any);
+    stepRepo = new AiChatRunStepRepo(db as any);
+    workspaceId = (await createWorkspace(db)).id;
+    otherWorkspaceId = (await createWorkspace(db)).id;
+    userId = (await createUser(db, workspaceId)).id;
+  });
+
+  afterAll(async () => {
+    await destroyTestDb();
+  });
+
+  it('getMessages reconstructs a mid-run row from the steps table (finished rows untouched)', async () => {
+    const chatId = (
+      await createChat(db, { workspaceId, creatorId: userId })
+    ).id;
+    const user = { id: userId } as User;
+    const workspace = { id: workspaceId } as Workspace;
+
+    // A prior FINISHED assistant row that already carries inline parts — the needy
+    // predicate must SKIP it (no step fetch), returned untouched.
+    const finishedParts = assistantParts(
+      [{ text: 'done earlier', toolCalls: [], toolResults: [] } as any],
+      '',
+    );
+    await createMessage(db, {
+      workspaceId,
+      chatId,
+      role: 'assistant',
+      content: 'done earlier',
+      status: 'completed',
+      metadata: { parts: finishedParts, toolTraceVersion: 2, stepsPersisted: 1 },
+      createdAt: new Date(Date.now() - 3000),
+    });
+
+    // The mid-run row a crash-before-finalize left behind: a step marker only
+    // (parts:[] , content:''), status 'streaming'. Its real parts live ONLY in the
+    // steps table.
+    const midRun = await createMessage(db, {
+      workspaceId,
+      chatId,
+      role: 'assistant',
+      content: '',
+      status: 'streaming',
+      metadata: stepMarkerMetadata(2),
+      createdAt: new Date(Date.now() - 1000),
+    });
+
+    const step0 = assistantParts(
+      [
+        {
+          text: 'reasoning about the page',
+          toolCalls: [
+            { toolCallId: 'g1', toolName: 'getPage', input: { id: 'p1' } },
+          ],
+          toolResults: [
+            { toolCallId: 'g1', toolName: 'getPage', output: { id: 'p1', body: 'B' } },
+          ],
+        } as any,
+      ],
+      '',
+    );
+    const step1 = assistantParts(
+      [{ text: 'partial synthesis so far', toolCalls: [], toolResults: [] } as any],
+      '',
+    );
+    await stepRepo.insertStep(midRun.id, workspaceId, 0, step0);
+    await stepRepo.insertStep(midRun.id, workspaceId, 1, step1);
+
+    // Workspace-scoping guard: a step row for the SAME message id under a DIFFERENT
+    // workspace must NEVER leak into this workspace's reconstruction.
+    await stepRepo.insertStep(midRun.id, otherWorkspaceId, 99, [
+      { type: 'text', text: 'FOREIGN_WORKSPACE_LEAK' },
+    ]);
+
+    const res = await buildController().getMessages(
+      { chatId } as any,
+      { limit: 50 } as any,
+      user,
+      workspace,
+    );
+
+    const items = res.items as any[];
+    const finished = items.find((r) => r.status === 'completed');
+    const reconstructed = items.find((r) => r.id === midRun.id);
+
+    // The finished row passed through with its inline parts unchanged.
+    expect(finished.metadata.parts).toEqual(finishedParts);
+
+    // The mid-run row's parts were reconstructed from the two step rows, in order,
+    // exactly as assembleStepParts concatenates them — the client seed sees the
+    // persisted progress with no change to itself.
+    const expected = assembleStepParts([
+      { stepIndex: 0, parts: step0 },
+      { stepIndex: 1, parts: step1 },
+    ] as any);
+    expect(reconstructed.metadata.parts).toEqual(expected);
+    // The foreign-workspace step row did NOT leak in.
+    expect(JSON.stringify(reconstructed.metadata.parts)).not.toContain(
+      'FOREIGN_WORKSPACE_LEAK',
+    );
+    // Sanity: reconstruction produced real content (text + the paired tool part +
+    // the second step's text), not an empty fallback.
+    expect(reconstructed.metadata.parts).toContainEqual({
+      type: 'text',
+      text: 'reasoning about the page',
+    });
+    expect(
+      (reconstructed.metadata.parts as any[]).some((p) => p.type === 'tool-getPage'),
+    ).toBe(true);
+  }, 60_000);
+});

apps/server/test/integration/ai-chat-run-step-repo.int-spec.ts

@@ -0,0 +1,163 @@
+import { randomBytes } from 'crypto';
+import { Kysely } from 'kysely';
+import { AiChatRunStepRepo } from '@docmost/db/repos/ai-chat/ai-chat-run-step.repo';
+import { AiChatMessageRepo } from '@docmost/db/repos/ai-chat/ai-chat-message.repo';
+import {
+  assistantParts,
+  reconstructRunParts,
+  hydrateAssistantParts,
+  stepMarkerMetadata,
+  rowHasInlineParts,
+} from '../../src/core/ai-chat/ai-chat.service';
+import {
+  getTestDb,
+  destroyTestDb,
+  createWorkspace,
+  createUser,
+  createChat,
+} from './db';
+
+/**
+ * #492 append-persist — the reconstruct CONTRACT on a live Postgres. Proves that a
+ * turn persisted the NEW way (per-step rows in `ai_chat_run_steps`, only a step
+ * marker on the message row) reconstructs to the SAME UI parts as a turn persisted
+ * the OLD way (full `metadata.parts` inline on the row, no step rows) — so the
+ * era-switch is invisible to attach / delta-poll / export. Real repos + real jsonb
+ * roundtrip, not a mock (a mock cannot prove the parts survive the jsonb column
+ * byte-identical).
+ */
+type Step = {
+  text: string;
+  toolCalls: Array<{ toolCallId: string; toolName: string; input: unknown }>;
+  toolResults: Array<{ toolCallId: string; toolName: string; output: unknown }>;
+};
+
+// A realistic step: some text + a getPage tool call whose ~100 KB body is
+// INCOMPRESSIBLE random base64 (a 'x'.repeat filler would TOAST away and hide the
+// real bytes). Under MAX_TOOL_OUTPUT_BYTES (200 KB) it is stored uncompacted.
+function makeStep(i: number, outputBytes = 4_000): Step {
+  const body = randomBytes(Math.ceil(outputBytes * 0.75)).toString('base64');
+  return {
+    text: `step ${i} text`,
+    toolCalls: [
+      { toolCallId: `c${i}`, toolName: 'getPage', input: { id: `p${i}` } },
+    ],
+    toolResults: [
+      {
+        toolCallId: `c${i}`,
+        toolName: 'getPage',
+        output: { id: `p${i}`, title: `Page ${i}`, body },
+      },
+    ],
+  };
+}
+
+describe('AiChatRunStepRepo + reconstruct contract [integration]', () => {
+  let db: Kysely<any>;
+  let stepRepo: AiChatRunStepRepo;
+  let msgRepo: AiChatMessageRepo;
+  let workspaceId: string;
+  let userId: string;
+  let chatId: string;
+
+  beforeAll(async () => {
+    db = getTestDb();
+    stepRepo = new AiChatRunStepRepo(db as any);
+    msgRepo = new AiChatMessageRepo(db as any);
+    workspaceId = (await createWorkspace(db)).id;
+    userId = (await createUser(db, workspaceId)).id;
+    chatId = (await createChat(db, { workspaceId, creatorId: userId })).id;
+  });
+
+  afterAll(async () => {
+    await destroyTestDb();
+  });
+
+  const seedRow = (metadata: unknown, status: string) =>
+    msgRepo.insert({
+      chatId,
+      workspaceId,
+      userId,
+      role: 'assistant',
+      content: '',
+      status,
+      metadata: metadata as never,
+    });
+
+  it('insertStep is idempotent per (message, stepIndex) and reads back in order', async () => {
+    const row = await seedRow(stepMarkerMetadata(0), 'streaming');
+    const parts0 = assistantParts([makeStep(0)], '');
+    const parts1 = assistantParts([makeStep(1)], '');
+
+    expect(await stepRepo.insertStep(row.id, workspaceId, 0, parts0)).toBe(true);
+    expect(await stepRepo.insertStep(row.id, workspaceId, 1, parts1)).toBe(true);
+    // A retried persist of the SAME step is a no-op (ON CONFLICT DO NOTHING).
+    expect(await stepRepo.insertStep(row.id, workspaceId, 0, parts0)).toBe(
+      false,
+    );
+
+    const steps = await stepRepo.findByMessage(row.id, workspaceId);
+    expect(steps.map((s) => s.stepIndex)).toEqual([0, 1]);
+
+    // Batch fetch groups by message id in step order.
+    const map = await stepRepo.findByMessageIds([row.id], workspaceId);
+    expect(map.get(row.id)!.map((s) => s.stepIndex)).toEqual([0, 1]);
+  });
+
+  it('a NEW-style (step-table) run reconstructs identically to an OLD-style (inline) run', async () => {
+    const steps = [makeStep(10), makeStep(11)];
+    // The inline parts the OLD full-row flush would have written.
+    const fullParts = assistantParts(steps, '');
+
+    // OLD-style record: full parts inline on the row, NO step rows.
+    const oldRow = await seedRow(
+      { parts: fullParts, toolTraceVersion: 2, stepsPersisted: 2 },
+      'completed',
+    );
+
+    // NEW-style record: only a step marker on the row + per-step rows.
+    const newRow = await seedRow(stepMarkerMetadata(2), 'streaming');
+    for (let i = 0; i < steps.length; i++) {
+      await stepRepo.insertStep(
+        newRow.id,
+        workspaceId,
+        i,
+        assistantParts([steps[i]], ''),
+      );
+    }
+
+    // Re-read both from the DB (proves the jsonb roundtrip).
+    const oldFetched = await msgRepo.findById(oldRow.id, workspaceId);
+    const newFetched = await msgRepo.findById(newRow.id, workspaceId);
+    const oldSteps = await stepRepo.findByMessage(oldRow.id, workspaceId);
+    const newSteps = await stepRepo.findByMessage(newRow.id, workspaceId);
+
+    // The discriminator: the old row carries inline parts, the new one does not.
+    expect(rowHasInlineParts(oldFetched!)).toBe(true);
+    expect(rowHasInlineParts(newFetched!)).toBe(false);
+    expect(oldSteps).toHaveLength(0);
+    expect(newSteps).toHaveLength(2);
+
+    const oldRecon = reconstructRunParts(oldFetched!, oldSteps);
+    const newRecon = reconstructRunParts(newFetched!, newSteps);
+
+    // Both reconstruct to the SAME parts + step count — the era is invisible.
+    expect(newRecon.parts).toEqual(fullParts);
+    expect(oldRecon.parts).toEqual(fullParts);
+    expect(newRecon.parts).toEqual(oldRecon.parts);
+    expect(newRecon.stepsPersisted).toBe(2);
+    expect(oldRecon.stepsPersisted).toBe(2);
+
+    // hydrateAssistantParts fills the new row's metadata.parts to match the old
+    // row's inline parts — so a consumer reading `metadata.parts` off the raw row
+    // (the client seed/poll, export) is unchanged across the era.
+    const map = await stepRepo.findByMessageIds([newRow.id], workspaceId);
+    const [hydrated] = hydrateAssistantParts([newFetched!], map);
+    expect((hydrated.metadata as { parts: unknown }).parts).toEqual(fullParts);
+    // A row that already has inline parts passes through untouched (same ref-shape).
+    const [oldPassThrough] = hydrateAssistantParts([oldFetched!], map);
+    expect((oldPassThrough.metadata as { parts: unknown }).parts).toEqual(
+      fullParts,
+    );
+  });
+});

apps/server/test/integration/ai-chat-run-steps-migration.int-spec.ts

@@ -0,0 +1,63 @@
+import { Kysely, sql } from 'kysely';
+import {
+  up,
+  down,
+} from '../../src/database/migrations/20260708T120000-ai-chat-run-steps';
+import { getTestDb, destroyTestDb } from './db';
+
+/**
+ * #492 migration up/down roundtrip on a LIVE Postgres. global-setup already
+ * migrated docmost_test to latest (so the table exists at start); this drives the
+ * migration's own down()/up() and asserts the table presence toggles, then leaves
+ * it PRESENT (up) so the shared test DB is intact for any spec that runs after.
+ */
+async function tableExists(db: Kysely<any>): Promise<boolean> {
+  const row = (
+    await sql<{ t: string | null }>`select to_regclass('ai_chat_run_steps') as t`.execute(
+      db,
+    )
+  ).rows[0];
+  return row.t !== null;
+}
+
+async function uniqueIndexExists(db: Kysely<any>): Promise<boolean> {
+  const row = (
+    await sql<{
+      t: string | null;
+    }>`select to_regclass('ai_chat_run_steps_message_step_uidx') as t`.execute(db)
+  ).rows[0];
+  return row.t !== null;
+}
+
+describe('20260708 ai_chat_run_steps migration roundtrip [integration]', () => {
+  let db: Kysely<any>;
+
+  beforeAll(() => {
+    db = getTestDb();
+  });
+
+  afterAll(async () => {
+    // Belt-and-suspenders: guarantee the table is present for later specs even if
+    // an assertion threw mid-roundtrip.
+    if (!(await tableExists(db))) await up(db);
+    await destroyTestDb();
+  });
+
+  it('down() drops the table+index and up() recreates them (idempotent)', async () => {
+    // Starts applied (global-setup migrated to latest).
+    expect(await tableExists(db)).toBe(true);
+    expect(await uniqueIndexExists(db)).toBe(true);
+
+    await down(db);
+    expect(await tableExists(db)).toBe(false);
+    expect(await uniqueIndexExists(db)).toBe(false);
+
+    await up(db);
+    expect(await tableExists(db)).toBe(true);
+    expect(await uniqueIndexExists(db)).toBe(true);
+
+    // up() is idempotent (ifNotExists) — a second run is a harmless no-op.
+    await expect(up(db)).resolves.not.toThrow();
+    expect(await tableExists(db)).toBe(true);
+  });
+});