Hermes-agent

apps/desktop/scripts/after-pack.cjs

@@ -0,0 +1,41 @@
+/**
+ * after-pack.cjs — electron-builder afterPack hook.
+ *
+ * Stamps the Hermes icon + identity onto the packed Windows Hermes.exe via
+ * rcedit (delegated to set-exe-identity.cjs). This runs for EVERY packed build
+ * — first install, `hermes desktop`, the installer's --update rebuild, and a
+ * dev's manual `npm run pack` — so the branded exe can never silently revert
+ * to the stock "Electron" icon/name (the bug when the stamp lived only in
+ * install.ps1, which the update path doesn't use).
+ *
+ * Windows-only: rcedit edits PE resources, irrelevant on macOS/Linux where the
+ * app identity comes from the bundle Info.plist / desktop entry. Best-effort:
+ * a stamp failure must never fail an otherwise-good build (worst case is the
+ * stock icon, not a broken app), so we log and resolve rather than throw.
+ *
+ * electron-builder passes a context with:
+ *   - electronPlatformName: 'win32' | 'darwin' | 'linux'
+ *   - appOutDir:            the unpacked app directory for this target
+ *   - packager.appInfo.productFilename: the exe basename (e.g. 'Hermes')
+ */
+
+const path = require('node:path')
+
+const { stampExeIdentity } = require('./set-exe-identity.cjs')
+
+exports.default = async function afterPack(context) {
+  if (context.electronPlatformName !== 'win32') {
+    return
+  }
+
+  const productName = context.packager?.appInfo?.productFilename || 'Hermes'
+  const exe = path.join(context.appOutDir, `${productName}.exe`)
+  const desktopRoot = path.resolve(__dirname, '..')
+
+  try {
+    await stampExeIdentity(exe, desktopRoot)
+  } catch (err) {
+    // Never fail the build over a cosmetic stamp.
+    console.warn(`[after-pack] exe identity stamp failed (${err.message}); Hermes.exe keeps the stock Electron icon`)
+  }
+}

apps/desktop/scripts/assert-dist-built.cjs

@@ -0,0 +1,70 @@
+"use strict"
+
+// Build-time guard: refuse to hand a half-built renderer to electron-builder.
+//
+// `npm run pack` / `npm run dist*` are `npm run build && npm run builder`.
+// If the `build` step (tsc -b && vite build) fails but packaging proceeds
+// anyway — a stale checkout that fails typecheck, an interrupted vite build,
+// or npm not short-circuiting `&&` in some shells — electron-builder happily
+// packages an app with an empty or missing `dist/`. The result launches but
+// blank-pages with `ERR_FILE_NOT_FOUND` for dist/index.html, with no clue why.
+//
+// This runs at the tail of `build`, after vite build, so any packaging path
+// inherits it. It fails loud and early instead of shipping a broken bundle.
+// See issues #39484 (renderer blank page) and #41327 / #39472 (dashboard 404).
+
+const fs = require("fs")
+const path = require("path")
+
+// Pure check — returns { ok: true } or { ok: false, error: "..." }.
+// Kept side-effect-free so it can be unit tested without spawning a process.
+function checkDistBuilt(distDir) {
+  if (!fs.existsSync(distDir) || !fs.statSync(distDir).isDirectory()) {
+    return { ok: false, error: `no dist directory at ${distDir}` }
+  }
+
+  const indexHtml = path.join(distDir, "index.html")
+  if (!fs.existsSync(indexHtml) || !fs.statSync(indexHtml).isFile()) {
+    return { ok: false, error: `dist/index.html is missing at ${indexHtml}` }
+  }
+  if (fs.statSync(indexHtml).size === 0) {
+    return { ok: false, error: `dist/index.html is empty at ${indexHtml}` }
+  }
+
+  // index.html alone isn't enough — vite emits hashed JS into dist/assets.
+  // An index.html with no script bundle still blank-pages.
+  const assetsDir = path.join(distDir, "assets")
+  const hasAssets =
+    fs.existsSync(assetsDir) &&
+    fs.statSync(assetsDir).isDirectory() &&
+    fs.readdirSync(assetsDir).some(name => name.endsWith(".js"))
+  if (!hasAssets) {
+    return { ok: false, error: `dist/assets has no built JS bundle (expected vite output under ${assetsDir})` }
+  }
+
+  return { ok: true }
+}
+
+function main() {
+  const desktopRoot = path.resolve(__dirname, "..")
+  const distDir = path.join(desktopRoot, "dist")
+  const result = checkDistBuilt(distDir)
+
+  if (!result.ok) {
+    console.error(`\n✗ assert-dist-built: ${result.error}`)
+    console.error("  The renderer bundle is missing or incomplete, so packaging")
+    console.error("  would produce an app that launches to a blank page.")
+    console.error("  Re-run the build and check the tsc/vite output above for the")
+    console.error("  real failure, then package again:")
+    console.error(`    cd ${desktopRoot} && npm run build\n`)
+    process.exit(1)
+  }
+
+  console.log("✓ assert-dist-built: dist/index.html + assets present")
+}
+
+if (require.main === module) {
+  main()
+}
+
+module.exports = { checkDistBuilt }

apps/desktop/scripts/assert-dist-built.test.cjs

@@ -0,0 +1,84 @@
+const assert = require('node:assert/strict')
+const fs = require('node:fs')
+const os = require('node:os')
+const path = require('node:path')
+const test = require('node:test')
+
+const { checkDistBuilt } = require('../scripts/assert-dist-built.cjs')
+
+function makeDist(extra) {
+  const tempRoot = fs.mkdtempSync(path.join(os.tmpdir(), 'hermes-assert-dist-'))
+  const distDir = path.join(tempRoot, 'dist')
+  fs.mkdirSync(distDir, { recursive: true })
+  if (extra) extra(distDir)
+  return { tempRoot, distDir }
+}
+
+test('checkDistBuilt passes when index.html + an assets JS bundle exist', () => {
+  const { tempRoot, distDir } = makeDist(d => {
+    fs.writeFileSync(path.join(d, 'index.html'), '<!doctype html><div id=root></div>', 'utf8')
+    fs.mkdirSync(path.join(d, 'assets'))
+    fs.writeFileSync(path.join(d, 'assets', 'index-abc123.js'), 'console.log(1)', 'utf8')
+  })
+  try {
+    assert.deepEqual(checkDistBuilt(distDir), { ok: true })
+  } finally {
+    fs.rmSync(tempRoot, { recursive: true, force: true })
+  }
+})
+
+test('checkDistBuilt fails when the dist directory is absent', () => {
+  const tempRoot = fs.mkdtempSync(path.join(os.tmpdir(), 'hermes-assert-dist-'))
+  try {
+    const result = checkDistBuilt(path.join(tempRoot, 'dist'))
+    assert.equal(result.ok, false)
+    assert.match(result.error, /no dist directory/)
+  } finally {
+    fs.rmSync(tempRoot, { recursive: true, force: true })
+  }
+})
+
+test('checkDistBuilt fails when index.html is missing', () => {
+  const { tempRoot, distDir } = makeDist(d => {
+    fs.mkdirSync(path.join(d, 'assets'))
+    fs.writeFileSync(path.join(d, 'assets', 'index-abc123.js'), 'console.log(1)', 'utf8')
+  })
+  try {
+    const result = checkDistBuilt(distDir)
+    assert.equal(result.ok, false)
+    assert.match(result.error, /index\.html is missing/)
+  } finally {
+    fs.rmSync(tempRoot, { recursive: true, force: true })
+  }
+})
+
+test('checkDistBuilt fails when index.html is empty', () => {
+  const { tempRoot, distDir } = makeDist(d => {
+    fs.writeFileSync(path.join(d, 'index.html'), '', 'utf8')
+    fs.mkdirSync(path.join(d, 'assets'))
+    fs.writeFileSync(path.join(d, 'assets', 'index-abc123.js'), 'console.log(1)', 'utf8')
+  })
+  try {
+    const result = checkDistBuilt(distDir)
+    assert.equal(result.ok, false)
+    assert.match(result.error, /index\.html is empty/)
+  } finally {
+    fs.rmSync(tempRoot, { recursive: true, force: true })
+  }
+})
+
+test('checkDistBuilt fails when assets/ has no JS bundle', () => {
+  const { tempRoot, distDir } = makeDist(d => {
+    fs.writeFileSync(path.join(d, 'index.html'), '<!doctype html>', 'utf8')
+    fs.mkdirSync(path.join(d, 'assets'))
+    // CSS only, no JS — still a blank page at runtime.
+    fs.writeFileSync(path.join(d, 'assets', 'index-abc123.css'), 'body{}', 'utf8')
+  })
+  try {
+    const result = checkDistBuilt(distDir)
+    assert.equal(result.ok, false)
+    assert.match(result.error, /no built JS bundle/)
+  } finally {
+    fs.rmSync(tempRoot, { recursive: true, force: true })
+  }
+})

apps/desktop/scripts/assert-root-install.cjs

@@ -0,0 +1,13 @@
+"use strict"
+
+const fs = require("fs")
+const path = require("path")
+
+const root = path.resolve(__dirname, "..", "..", "..")
+
+try {
+  fs.accessSync(path.join(root, "node_modules", "vite", "package.json"))
+} catch {
+  console.error(`Run from repo root: cd ${root} && npm ci`)
+  process.exit(1)
+}

apps/desktop/scripts/before-build.cjs

@@ -0,0 +1,11 @@
+/**
+ * Desktop bundles ship precompiled renderer assets. Returning false here tells
+ * electron-builder to skip the node_modules collector/install step, which
+ * avoids workspace dependency graph explosions and keeps packaging
+ * deterministic across environments. The Hermes Agent Python payload is no
+ * longer bundled; the Electron app fetches it at first launch via
+ * `install.ps1`'s stage protocol (Windows). See `electron/main.cjs`.
+ */
+module.exports = async function beforeBuild() {
+  return false
+}

apps/desktop/scripts/before-pack.cjs

@@ -0,0 +1,78 @@
+'use strict'
+
+/**
+ * before-pack.cjs — electron-builder beforePack hook.
+ *
+ * Removes any stale unpacked app directory (`appOutDir`) before
+ * electron-builder stages the Electron binaries into it.
+ *
+ * WHY THIS EXISTS
+ * ---------------
+ * electron-builder's final packaging step copies the stock `electron`
+ * binary into `release/<platform>-unpacked/` and then renames it to the
+ * product name (`Hermes`). If a PREVIOUS `npm run pack` was interrupted
+ * (Ctrl-C, OOM kill, crash, full disk) the unpacked directory is left in a
+ * corrupted partial state: it keeps the already-renamed `LICENSE.electron.txt`
+ * and the Chromium payload (.pak/.so/icudtl.dat/chrome-sandbox) but is MISSING
+ * the `electron` binary itself.
+ *
+ * On the next run, electron-builder sees the destination directory already
+ * populated, skips re-copying the binary it thinks is present, then tries to
+ * rename a `electron` file that no longer exists. The build dies with:
+ *
+ *   ENOENT: no such file or directory, rename
+ *   '.../release/linux-unpacked/electron' -> '.../release/linux-unpacked/Hermes'
+ *
+ * This is a hard failure with no obvious cause for the user — `hermes desktop`
+ * just prints "Desktop GUI build failed" and the only fix is to manually
+ * `rm -rf` the release directory, which a normal user has no way to know.
+ *
+ * The packaging step is not idempotent across an interrupted run, so we make
+ * it idempotent ourselves: wipe the target unpacked directory up front so
+ * electron-builder always stages into a clean tree. This is safe — the
+ * directory is a pure build artifact that electron-builder fully recreates
+ * on every pack; nothing else depends on its prior contents.
+ *
+ * Cross-platform: the same partial-state trap exists on macOS
+ * (the mac-unpacked Hermes.app bundle) and Windows (win-unpacked), so we
+ * clean whatever `appOutDir` electron-builder hands us regardless of platform.
+ *
+ * Best-effort: a cleanup failure must never mask the real build. We log and
+ * resolve rather than throw — worst case electron-builder hits the original
+ * ENOENT, which is no worse than not having this hook at all.
+ *
+ * electron-builder passes a context with:
+ *   - appOutDir:            the unpacked app directory about to be staged
+ *   - electronPlatformName: 'win32' | 'darwin' | 'linux'
+ */
+
+const fs = require('node:fs')
+
+function cleanStaleAppOutDir(appOutDir) {
+  if (!appOutDir || typeof appOutDir !== 'string') {
+    return false
+  }
+  if (!fs.existsSync(appOutDir)) {
+    return false
+  }
+  // Recursive + force so a half-written tree (read-only bits, partial files)
+  // can't block the wipe. retry/maxRetries rides out transient EBUSY on
+  // Windows where an AV/indexer may briefly hold a handle.
+  fs.rmSync(appOutDir, { recursive: true, force: true, maxRetries: 5, retryDelay: 100 })
+  return true
+}
+
+exports.cleanStaleAppOutDir = cleanStaleAppOutDir
+
+exports.default = async function beforePack(context) {
+  const appOutDir = context && context.appOutDir
+  try {
+    if (cleanStaleAppOutDir(appOutDir)) {
+      console.log(`[before-pack] removed stale unpacked dir before staging: ${appOutDir}`)
+    }
+  } catch (err) {
+    // Never fail the build over cleanup; surface why so a genuinely stuck
+    // directory (permissions, mount) is still diagnosable.
+    console.warn(`[before-pack] could not clean ${appOutDir} (${err.message}); continuing`)
+  }
+}

apps/desktop/scripts/before-pack.test.cjs

@@ -0,0 +1,53 @@
+const assert = require('node:assert/strict')
+const fs = require('node:fs')
+const os = require('node:os')
+const path = require('node:path')
+const test = require('node:test')
+
+const { cleanStaleAppOutDir } = require('../scripts/before-pack.cjs')
+
+test('cleanStaleAppOutDir removes a populated unpacked directory', () => {
+  const tempRoot = fs.mkdtempSync(path.join(os.tmpdir(), 'hermes-before-pack-'))
+  try {
+    const appOutDir = path.join(tempRoot, 'linux-unpacked')
+    fs.mkdirSync(appOutDir, { recursive: true })
+    // Reproduce the corrupted partial state: license + payload present,
+    // electron binary missing — exactly what trips the ENOENT rename.
+    fs.writeFileSync(path.join(appOutDir, 'LICENSE.electron.txt'), 'x', 'utf8')
+    fs.writeFileSync(path.join(appOutDir, 'resources.pak'), 'x', 'utf8')
+    fs.mkdirSync(path.join(appOutDir, 'resources'), { recursive: true })
+    fs.writeFileSync(path.join(appOutDir, 'resources', 'app.asar'), 'x', 'utf8')
+
+    const removed = cleanStaleAppOutDir(appOutDir)
+
+    assert.equal(removed, true)
+    assert.equal(fs.existsSync(appOutDir), false)
+  } finally {
+    fs.rmSync(tempRoot, { recursive: true, force: true })
+  }
+})
+
+test('cleanStaleAppOutDir is a no-op when the directory is absent', () => {
+  const tempRoot = fs.mkdtempSync(path.join(os.tmpdir(), 'hermes-before-pack-'))
+  try {
+    const missing = path.join(tempRoot, 'does-not-exist')
+    assert.equal(cleanStaleAppOutDir(missing), false)
+  } finally {
+    fs.rmSync(tempRoot, { recursive: true, force: true })
+  }
+})
+
+test('cleanStaleAppOutDir ignores empty or invalid input', () => {
+  assert.equal(cleanStaleAppOutDir(''), false)
+  assert.equal(cleanStaleAppOutDir(undefined), false)
+  assert.equal(cleanStaleAppOutDir(null), false)
+  assert.equal(cleanStaleAppOutDir(42), false)
+})
+
+test('beforePack default export resolves even when cleanup throws', async () => {
+  const { default: beforePack } = require('../scripts/before-pack.cjs')
+  // A directory path that rmSync can't remove is simulated by passing a
+  // context whose appOutDir is a file the hook will try (and be allowed) to
+  // remove; the contract under test is that the hook never rejects.
+  await assert.doesNotReject(beforePack({ appOutDir: '', electronPlatformName: 'linux' }))
+})

apps/desktop/scripts/click-session.mjs

@@ -0,0 +1,51 @@
+// Click on a session by partial title match.
+const list = await (await fetch('http://127.0.0.1:9222/json/list')).json()
+const tgt = list.find(t => t.type === 'page' && t.url.startsWith('http'))
+const ws = new WebSocket(tgt.webSocketDebuggerUrl)
+let id = 0
+const pending = new Map()
+ws.addEventListener('message', ev => {
+  const m = JSON.parse(ev.data)
+  if (m.id != null && pending.has(m.id)) {
+    pending.get(m.id)(m)
+    pending.delete(m.id)
+  }
+})
+await new Promise(r => ws.addEventListener('open', r))
+const send = (method, params = {}) =>
+  new Promise(r => {
+    const i = ++id
+    pending.set(i, r)
+    ws.send(JSON.stringify({ id: i, method, params }))
+  })
+
+const title = process.argv[2] || 'Phaser particle'
+const r = await send('Runtime.evaluate', {
+  expression: `
+    (() => {
+      const titleMatch = ${JSON.stringify(title)}
+      const all = document.querySelectorAll('button, a, div[role="button"]')
+      const found = [...all].find(el => (el.textContent || '').includes(titleMatch))
+      if (!found) return JSON.stringify({ found: false, tried: titleMatch })
+      found.scrollIntoView()
+      found.click()
+      return JSON.stringify({ found: true, tag: found.tagName, text: (found.textContent || '').slice(0, 80) })
+    })()
+  `,
+  returnByValue: true
+})
+console.log('click raw:', JSON.stringify(r, null, 2))
+await new Promise(r => setTimeout(r, 3000))
+
+const status = await send('Runtime.evaluate', {
+  expression: `JSON.stringify({
+    url: location.href,
+    hasComposer: !!document.querySelector('[data-slot="composer-rich-input"]'),
+    threadMessages: document.querySelectorAll('[data-slot="aui_message"]').length,
+    bodyTextSnippet: document.body.innerText.slice(0, 500),
+    title: document.title
+  })`,
+  returnByValue: true
+})
+console.log('after click:', status.result.value)
+ws.close()

apps/desktop/scripts/dev-no-hmr.mjs

@@ -0,0 +1,22 @@
+#!/usr/bin/env node
+// Launch the desktop renderer with HMR disabled so the React Fast Refresh
+// preamble path is skipped. This sidesteps a current Vite 8 / plugin-react 6
+// bug where the preamble script is not injected into index.html → renderer
+// throws "$RefreshReg$ is not defined" on every TSX module → React tree
+// never mounts.
+//
+// We're not trying to use HMR while profiling typing lag anyway. Hermes desktop
+// boots, you type, profiler measures. HMR off is fine.
+//
+// Usage: node apps/desktop/scripts/dev-no-hmr.mjs
+//        (then in another shell, run electron --remote-debugging-port=9222 .)
+
+import { createServer } from 'vite'
+
+const server = await createServer({
+  configFile: new URL('../vite.config.ts', import.meta.url).pathname,
+  root: new URL('../', import.meta.url).pathname,
+  server: { hmr: false, host: '127.0.0.1', port: 5174, strictPort: true }
+})
+await server.listen()
+server.printUrls()

apps/desktop/scripts/diag-jump.mjs

@@ -0,0 +1,115 @@
+// Wrap the thread scroller's properties and observe pin/scroll/RO events
+// in real time during a submit, then print the timeline.
+const list = await (await fetch('http://127.0.0.1:9222/json/list')).json()
+const tgt = list.find(t => t.type === 'page' && t.url.startsWith('http'))
+const ws = new WebSocket(tgt.webSocketDebuggerUrl)
+let id = 0
+const pending = new Map()
+ws.addEventListener('message', ev => {
+  const m = JSON.parse(ev.data)
+  if (m.id != null && pending.has(m.id)) {
+    pending.get(m.id)(m)
+    pending.delete(m.id)
+  }
+})
+await new Promise(r => ws.addEventListener('open', r))
+const send = (m, p = {}) =>
+  new Promise(r => {
+    const i = ++id
+    pending.set(i, r)
+    ws.send(JSON.stringify({ id: i, method: m, params: p }))
+  })
+const evalP = async expr => {
+  const r = await send('Runtime.evaluate', { expression: expr, returnByValue: true })
+  if (r.result?.exceptionDetails) throw new Error(r.result.exceptionDetails.text)
+  return r.result.result.value
+}
+
+await evalP(`(() => {
+  const v = document.querySelector('[data-slot="aui_thread-viewport"]')
+  if (v) v.scrollTop = v.scrollHeight
+})()`)
+await new Promise(r => setTimeout(r, 300))
+
+await evalP(`(() => {
+  const el = document.querySelector('[data-slot="composer-rich-input"]')
+  el.focus()
+  const r = document.createRange(); r.selectNodeContents(el); r.collapse(false)
+  window.getSelection().removeAllRanges(); window.getSelection().addRange(r)
+})()`)
+
+const text = 'short follow-up'
+for (const c of text) {
+  await send('Input.dispatchKeyEvent', { type: 'char', text: c, unmodifiedText: c })
+  await new Promise(r => setTimeout(r, 10))
+}
+await new Promise(r => setTimeout(r, 300))
+
+// Hook into the viewport scrollTop setter + scroll + RO so we see every event
+await evalP(`(() => {
+  const v = document.querySelector('[data-slot="aui_thread-viewport"]')
+  const events = []
+  window.__threadEvents = events
+  const t0 = performance.now()
+  const push = (kind, detail) => events.push({ t: performance.now() - t0, kind, ...detail })
+
+  // intercept scrollTop writes
+  const desc = Object.getOwnPropertyDescriptor(Element.prototype, 'scrollTop')
+  Object.defineProperty(v, 'scrollTop', {
+    get() { return desc.get.call(this) },
+    set(val) {
+      push('scrollTop=', { val, fromScrollHeight: this.scrollHeight, stackTop: (new Error()).stack.split('\\n').slice(2, 5).map(s => s.trim()).join(' | ') })
+      desc.set.call(this, val)
+    },
+    configurable: true
+  })
+
+  // scroll event
+  v.addEventListener('scroll', () => {
+    push('scroll', { scrollTop: v.scrollTop, scrollHeight: v.scrollHeight })
+  }, { passive: true, capture: true })
+
+  // RO on the viewport itself
+  const ro = new ResizeObserver((entries) => {
+    for (const e of entries) {
+      push('RO', { target: e.target.getAttribute('data-slot') || e.target.tagName, h: e.contentRect.height })
+    }
+  })
+  ro.observe(v)
+  if (v.firstElementChild) ro.observe(v.firstElementChild)
+
+  // mutationobserver on the viewport
+  const mo = new MutationObserver((muts) => {
+    push('mut', { count: muts.length, added: muts.reduce((s, m) => s + m.addedNodes.length, 0), removed: muts.reduce((s, m) => s + m.removedNodes.length, 0) })
+  })
+  mo.observe(v, { childList: true, subtree: true, characterData: true })
+
+  window.__teardown = () => { ro.disconnect(); mo.disconnect() }
+  return true
+})()`)
+
+// fire Enter
+await send('Input.dispatchKeyEvent', {
+  type: 'rawKeyDown', windowsVirtualKeyCode: 13, key: 'Enter', code: 'Enter', text: '\r', unmodifiedText: '\r'
+})
+await send('Input.dispatchKeyEvent', { type: 'keyUp', windowsVirtualKeyCode: 13, key: 'Enter', code: 'Enter' })
+
+await new Promise(r => setTimeout(r, 1200))
+
+const events = JSON.parse(await evalP(`JSON.stringify(window.__threadEvents || [])`))
+console.log(`\n${events.length} events:`)
+for (const e of events) {
+  const t = String(e.t.toFixed(0)).padStart(5)
+  const { kind, t: _t, ...rest } = e
+  console.log(`  ${t}ms  ${kind.padEnd(12)} ${JSON.stringify(rest)}`)
+}
+
+await evalP(`window.__teardown?.()`)
+// Cancel running agent
+await evalP(`(() => {
+  for (const b of document.querySelectorAll('button')) {
+    if ((b.getAttribute('aria-label') || '').toLowerCase().includes('stop')) { b.click(); return 'stopped' }
+  }
+})()`)
+
+ws.close()

apps/desktop/scripts/diag-scroll-reset.mjs

@@ -0,0 +1,229 @@
+// Reproduce + diagnose the "scroll wheel resets position while reading" bug.
+//
+// The complaint (Windows, mouse wheel): scrolling UP through a chat to re-read
+// older content randomly yanks the view to a different position, so you have to
+// fight the scrollbar. Mac users on trackpads don't see it.
+//
+// Hypothesis: the thread scroller has the browser default `overflow-anchor:
+// auto`, and the thread renders items in natural document flow (padding
+// spacers, NOT transforms). When an item above the viewport is measured by
+// @tanstack/react-virtual (its real height differs a lot from the 220px
+// estimate) — or when Shiki/images/fonts reflow it — TWO mechanisms both
+// adjust scrollTop for the same delta: TanStack's measurement compensation AND
+// the browser's native scroll anchoring. The double-correction lurches the
+// view. A mouse wheel's coarse, discrete notches mount/measure several
+// under-estimated turns per tick, so the over-correction is large and visible;
+// a trackpad's ~1-3px/frame keeps it sub-perceptual.
+//
+// This script drives synthetic mouse-wheel-UP scrolling on a long thread and
+// measures how much a tracked on-screen turn jumps, first with
+// `overflow-anchor: auto` (reproduce) then `overflow-anchor: none` (the fix).
+// If the fix run shows dramatically fewer/smaller jumps, the hypothesis holds.
+//
+// Prereq: a running desktop app with remote debugging on 9222, on a thread
+// with enough history to scroll (the longer / more code+tool blocks, the
+// better the repro). Then:  node apps/desktop/scripts/diag-scroll-reset.mjs
+
+const NOTCHES = 14 // wheel-up ticks per sweep
+const NOTCH_PX = 120 // Windows wheel notch ≈ 120px
+const NOTCH_GAP_MS = 130 // let each smooth-scroll animation settle
+const REVERSE_JUMP_PX = 6 // tracked turn moving UP while scrolling up = wrong way
+const LURCH_PX = 60 // single-frame on-screen jump that reads as a "reset"
+
+const list = await (await fetch('http://127.0.0.1:9222/json/list')).json()
+const tgt = list.find(t => t.type === 'page' && t.url.startsWith('http'))
+if (!tgt) {
+  console.error('No page target on :9222. Is the desktop app running with --remote-debugging-port=9222?')
+  process.exit(1)
+}
+const ws = new WebSocket(tgt.webSocketDebuggerUrl)
+let id = 0
+const pending = new Map()
+ws.addEventListener('message', ev => {
+  const m = JSON.parse(ev.data)
+  if (m.id != null && pending.has(m.id)) {
+    pending.get(m.id)(m)
+    pending.delete(m.id)
+  }
+})
+await new Promise(r => ws.addEventListener('open', r))
+const send = (m, p = {}) =>
+  new Promise(r => {
+    const i = ++id
+    pending.set(i, r)
+    ws.send(JSON.stringify({ id: i, method: m, params: p }))
+  })
+const evalP = async expr => {
+  const r = await send('Runtime.evaluate', { expression: expr, returnByValue: true })
+  if (r.result?.exceptionDetails) throw new Error(r.result.exceptionDetails.text)
+  return r.result.result.value
+}
+const sleep = ms => new Promise(r => setTimeout(r, ms))
+
+// Install per-sweep instrumentation. `mode` is the overflow-anchor value to
+// force inline so we A/B the exact same thread regardless of any CSS fix.
+// Starts from ~45% down the thread so there's room to scroll up into
+// not-yet-measured turns, tags the turn nearest viewport-center as the anchor,
+// then records (per rAF) scrollTop + that turn's on-screen top, plus every
+// scrollTop *setter* write (TanStack compensation) and ResizeObserver hit.
+async function arm(mode) {
+  await evalP(`(() => {
+    const v = document.querySelector('[data-slot="aui_thread-viewport"]')
+    if (!v) throw new Error('thread viewport not found')
+
+    // Force the overflow-anchor behavior under test (inline beats CSS).
+    v.style.overflowAnchor = ${JSON.stringify(mode)}
+
+    // Park ~45% down so a wheel-up sweep climbs into estimated-but-unmeasured
+    // turns above the fold (where the measurement correction fires).
+    v.scrollTop = Math.round(v.scrollHeight * 0.45)
+
+    // Tag the turn closest to viewport center; we track its on-screen top.
+    const vr = v.getBoundingClientRect()
+    const center = vr.top + v.clientHeight / 2
+    let best = null, bestD = Infinity
+    for (const el of v.querySelectorAll('[data-index]')) {
+      const r = el.getBoundingClientRect()
+      const d = Math.abs((r.top + r.height / 2) - center)
+      if (d < bestD) { bestD = d; best = el }
+    }
+    document.querySelectorAll('[data-se-anchor]').forEach(e => e.removeAttribute('data-se-anchor'))
+    if (best) best.setAttribute('data-se-anchor', '1')
+    const anchorIndex = best ? best.getAttribute('data-index') : null
+
+    const samples = []
+    const writes = []
+    const ros = []
+    const t0 = performance.now()
+
+    // Intercept scrollTop writes → these are JS (TanStack) corrections.
+    // Native browser scroll anchoring does NOT go through this setter, so a
+    // scrollTop change with no write in the same frame is a native adjust.
+    const desc = Object.getOwnPropertyDescriptor(Element.prototype, 'scrollTop')
+    Object.defineProperty(v, 'scrollTop', {
+      configurable: true,
+      get() { return desc.get.call(this) },
+      set(val) {
+        writes.push({ t: performance.now() - t0, val, sh: this.scrollHeight })
+        desc.set.call(this, val)
+      }
+    })
+    window.__restoreScrollTop = () => Object.defineProperty(v, 'scrollTop', desc)
+
+    const ro = new ResizeObserver(entries => {
+      for (const e of entries) {
+        ros.push({ t: performance.now() - t0, slot: e.target.getAttribute?.('data-slot') || e.target.tagName, h: Math.round(e.contentRect.height) })
+      }
+    })
+    ro.observe(v)
+    if (v.firstElementChild) ro.observe(v.firstElementChild)
+
+    let running = true
+    const tick = () => {
+      if (!running) return
+      const a = v.querySelector('[data-se-anchor]')
+      const ar = a ? a.getBoundingClientRect() : null
+      samples.push({
+        t: performance.now() - t0,
+        st: Math.round(v.scrollTop * 100) / 100,
+        sh: v.scrollHeight,
+        ch: v.clientHeight,
+        atop: ar ? Math.round(ar.top * 100) / 100 : null,
+        aconn: !!a
+      })
+      requestAnimationFrame(tick)
+    }
+    requestAnimationFrame(tick)
+
+    window.__se = { samples, writes, ros, anchorIndex, dpr: window.devicePixelRatio, stop() { running = false; ro.disconnect(); window.__restoreScrollTop?.() } }
+    return true
+  })()`)
+}
+
+async function wheelUpSweep() {
+  const { x, y } = await evalP(`(() => {
+    const v = document.querySelector('[data-slot="aui_thread-viewport"]')
+    const r = v.getBoundingClientRect()
+    return { x: Math.round(r.left + r.width / 2), y: Math.round(r.top + r.height / 2) }
+  })()`)
+
+  for (let i = 0; i < NOTCHES; i++) {
+    await send('Input.dispatchMouseEvent', { type: 'mouseWheel', x, y, deltaX: 0, deltaY: -NOTCH_PX })
+    await sleep(NOTCH_GAP_MS)
+  }
+  await sleep(400)
+}
+
+async function collect() {
+  const data = JSON.parse(await evalP(`(() => { window.__se.stop(); return JSON.stringify(window.__se) })()`))
+  return data
+}
+
+function analyze(label, data) {
+  const { samples, writes, ros, anchorIndex, dpr } = data
+  let reverseJumps = 0
+  let reverseSum = 0
+  let lurches = 0
+  let maxJump = 0
+  let nativeMoves = 0
+  let prev = null
+  for (const s of samples) {
+    if (prev && prev.aconn && s.aconn && prev.atop != null && s.atop != null) {
+      const dTop = s.atop - prev.atop // wheel-up should move content DOWN → dTop >= 0
+      const dSt = s.st - prev.st
+      // Native (browser-anchoring) move: scrollTop changed with no setter write in this frame window.
+      const wroteThisFrame = writes.some(w => w.t > prev.t && w.t <= s.t)
+      if (Math.abs(dSt) > 0.5 && !wroteThisFrame) nativeMoves++
+      if (dTop < -REVERSE_JUMP_PX) {
+        reverseJumps++
+        reverseSum += -dTop
+      }
+      if (Math.abs(dTop) > LURCH_PX) lurches++
+      if (Math.abs(dTop) > maxJump) maxJump = Math.abs(dTop)
+    }
+    prev = s
+  }
+  console.log(`\n── ${label} ──`)
+  console.log(`  devicePixelRatio:     ${dpr}${Number.isInteger(dpr) ? '' : '  (fractional — Windows scaling, worsens rounding jitter)'}`)
+  console.log(`  tracked turn index:   ${anchorIndex}`)
+  console.log(`  rAF frames:           ${samples.length}`)
+  console.log(`  scrollTop writes:     ${writes.length}   (TanStack measurement corrections)`)
+  console.log(`  ResizeObserver hits:  ${ros.length}`)
+  console.log(`  native scroll moves:  ${nativeMoves}   (scrollTop moved with NO JS write = browser anchoring)`)
+  console.log(`  reverse jumps:        ${reverseJumps}   (tracked turn yanked UP while scrolling up; total ${reverseSum.toFixed(0)}px)`)
+  console.log(`  big lurches (>${LURCH_PX}px):   ${lurches}`)
+  console.log(`  max single-frame jump: ${maxJump.toFixed(0)}px`)
+  return { reverseJumps, reverseSum, lurches, maxJump, nativeMoves }
+}
+
+console.log(`Wheel-up repro: ${NOTCHES} notches × ${NOTCH_PX}px, anchored mid-thread.\n`)
+
+await arm('auto')
+await sleep(150)
+await wheelUpSweep()
+const a = analyze('overflow-anchor: auto  (current / repro)', await collect())
+
+await sleep(300)
+
+await arm('none')
+await sleep(150)
+await wheelUpSweep()
+const b = analyze('overflow-anchor: none  (proposed fix)', await collect())
+
+// Clean up our tag.
+await evalP(`document.querySelectorAll('[data-se-anchor]').forEach(e => e.removeAttribute('data-se-anchor'))`)
+
+console.log('\n══ verdict ══')
+const drop = (x, y) => (x === 0 ? (y === 0 ? '0' : 'n/a') : `${Math.round((1 - y / x) * 100)}% fewer`)
+console.log(`  reverse jumps:  auto=${a.reverseJumps}  none=${b.reverseJumps}  (${drop(a.reverseJumps, b.reverseJumps)})`)
+console.log(`  big lurches:    auto=${a.lurches}  none=${b.lurches}  (${drop(a.lurches, b.lurches)})`)
+console.log(`  max jump:       auto=${a.maxJump.toFixed(0)}px  none=${b.maxJump.toFixed(0)}px`)
+console.log(`  native moves:   auto=${a.nativeMoves}  none=${b.nativeMoves}  (browser anchoring should ~vanish at none)`)
+if (a.reverseJumps + a.lurches > 0 && b.reverseJumps + b.lurches < a.reverseJumps + a.lurches) {
+  console.log('\n  → Jumps drop sharply with overflow-anchor:none → root cause confirmed.')
+} else if (a.reverseJumps + a.lurches === 0) {
+  console.log('\n  → No jumps captured this run. Use a longer thread (many code/tool blocks),')
+  console.log('    raise NOTCHES, and ensure you start scrolled up from the bottom.')
+}
+
+ws.close()

apps/desktop/scripts/eval.mjs

@@ -0,0 +1,21 @@
+// Simple eval helper — runs an expression and returns the result.value.
+const targets = await (await fetch('http://127.0.0.1:9222/json')).json()
+const t = targets.find((t) => t.url.includes('5174'))
+const ws = new WebSocket(t.webSocketDebuggerUrl)
+let id = 0
+const pending = new Map()
+ws.addEventListener('message', (ev) => {
+  const m = JSON.parse(ev.data)
+  if (pending.has(m.id)) { pending.get(m.id)(m); pending.delete(m.id) }
+})
+await new Promise((r) => ws.addEventListener('open', r))
+const send = (method, params) => new Promise((res) => { const i = ++id; pending.set(i, res); ws.send(JSON.stringify({ id: i, method, params })) })
+
+const expr = process.argv[2] || '1+1'
+const r = await send('Runtime.evaluate', { expression: expr, returnByValue: true, awaitPromise: true })
+if (r.result.exceptionDetails) {
+  console.error('EXCEPTION:', r.result.exceptionDetails.exception?.description)
+} else {
+  console.log(JSON.stringify(r.result.result.value, null, 2))
+}
+ws.close()

apps/desktop/scripts/leak-typing.mjs

@@ -0,0 +1,222 @@
+#!/usr/bin/env node
+// Leak-detection harness — measure detached DOM, listener count, and FiberNode
+// growth as a function of keystrokes typed.
+//
+// Workflow:
+//   1. Open session, focus composer
+//   2. forceGC; capture baseline counts
+//   3. Repeat N rounds: type M chars, forceGC, capture counts, clear composer
+//   4. Print growth-per-round table
+//
+// Usage:
+//   node apps/desktop/scripts/leak-typing.mjs [--rounds=6] [--chars=200] [--cps=40] [--port=9222]
+
+import { writeFileSync } from 'node:fs'
+
+const args = Object.fromEntries(
+  process.argv.slice(2).flatMap(s => {
+    const m = s.match(/^--([^=]+)(?:=(.*))?$/)
+    return m ? [[m[1], m[2] ?? true]] : []
+  })
+)
+const PORT = Number(args.port ?? 9222)
+const ROUNDS = Number(args.rounds ?? 6)
+const CHARS = Number(args.chars ?? 200)
+const CPS = Number(args.cps ?? 40)
+
+const log = (...m) => console.log('[leak]', ...m)
+
+async function pickRenderer() {
+  const list = await (await fetch(`http://127.0.0.1:${PORT}/json/list`)).json()
+  return list.find(t => t.type === 'page' && t.url.startsWith('http'))
+}
+
+function connect(url) {
+  return new Promise((resolve, reject) => {
+    const ws = new WebSocket(url)
+    let id = 0
+    const pending = new Map()
+    const events = new Map()
+    ws.addEventListener('open', () =>
+      resolve({
+        send(method, params = {}) {
+          const myId = ++id
+          ws.send(JSON.stringify({ id: myId, method, params }))
+          return new Promise((res, rej) => pending.set(myId, { res, rej }))
+        },
+        on(method, h) {
+          if (!events.has(method)) events.set(method, [])
+          events.get(method).push(h)
+        },
+        close: () => ws.close()
+      })
+    )
+    ws.addEventListener('error', reject)
+    ws.addEventListener('message', ev => {
+      const m = JSON.parse(typeof ev.data === 'string' ? ev.data : ev.data.toString('utf8'))
+      if (m.id != null) {
+        const p = pending.get(m.id)
+        if (!p) return
+        pending.delete(m.id)
+        m.error ? p.rej(new Error(m.error.message)) : p.res(m.result)
+      } else if (m.method) {
+        ;(events.get(m.method) ?? []).forEach(h => h(m.params))
+      }
+    })
+  })
+}
+
+async function evalInPage(cdp, expr) {
+  const r = await cdp.send('Runtime.evaluate', { expression: expr, returnByValue: true })
+  if (r.exceptionDetails) throw new Error(r.exceptionDetails.text)
+  return r.result.value
+}
+
+async function forceGCAndSettle(cdp) {
+  for (let i = 0; i < 3; i++) {
+    await cdp.send('HeapProfiler.collectGarbage')
+    await new Promise(r => setTimeout(r, 60))
+  }
+}
+
+async function focusComposer(cdp) {
+  return await evalInPage(
+    cdp,
+    `(() => {
+      const el = document.querySelector('[data-slot="composer-rich-input"]')
+      if (!el) return false
+      el.focus()
+      const range = document.createRange()
+      range.selectNodeContents(el)
+      range.collapse(false)
+      const sel = window.getSelection()
+      sel.removeAllRanges()
+      sel.addRange(range)
+      return true
+    })()`
+  )
+}
+
+async function clearComposer(cdp) {
+  await evalInPage(
+    cdp,
+    `(() => {
+      const el = document.querySelector('[data-slot="composer-rich-input"]')
+      if (!el) return false
+      // Clear via the same path as the composer's clear flow:
+      // dispatch a single Backspace until empty would be N round-trips; quicker
+      // to directly assign empty text and fire input.
+      el.innerHTML = ''
+      el.dispatchEvent(new InputEvent('input', { bubbles: true, inputType: 'deleteContentBackward' }))
+      el.focus()
+      return el.innerText.length === 0
+    })()`
+  )
+}
+
+async function snapshotCounts(cdp) {
+  // Counts via Runtime.evaluate using internal V8 counters where possible.
+  // For DOM stats we directly query the document.
+  // Performance metrics include JSHeapUsedSize, Nodes, JSEventListeners, etc.
+  const { metrics } = await cdp.send('Performance.getMetrics')
+  const byName = Object.fromEntries(metrics.map(m => [m.name, m.value]))
+  // Total nodes in document
+  const docNodes = await evalInPage(
+    cdp,
+    `document.getElementsByTagName('*').length + document.querySelectorAll('*').length / 2`
+  )
+  return {
+    heapUsedMB: (byName.JSHeapUsedSize / 1024 / 1024) || 0,
+    heapTotalMB: (byName.JSHeapTotalSize / 1024 / 1024) || 0,
+    nodes: byName.Nodes || 0,
+    jsListeners: byName.JSEventListeners || 0,
+    docNodes,
+    layoutCount: byName.LayoutCount || 0,
+    recalcStyleCount: byName.RecalcStyleCount || 0,
+    fps: byName.FramesPerSecond || 0
+  }
+}
+
+async function typeChars(cdp, text, cps) {
+  const intervalMs = Math.max(1, Math.round(1000 / cps))
+  const start = Date.now()
+  for (let i = 0; i < text.length; i++) {
+    await cdp.send('Input.dispatchKeyEvent', { type: 'char', text: text[i], unmodifiedText: text[i] })
+    const expected = start + (i + 1) * intervalMs
+    const wait = expected - Date.now()
+    if (wait > 0) await new Promise(r => setTimeout(r, wait))
+  }
+}
+
+const lorem =
+  'the quick brown fox jumps over the lazy dog while the agent thinks really hard about why typing into this composer feels like wading through molasses on a hot afternoon '
+function genText(n) {
+  let s = ''
+  while (s.length < n) s += lorem
+  return s.slice(0, n)
+}
+
+async function main() {
+  log(`port ${PORT} · ${ROUNDS} rounds × ${CHARS} chars @ ${CPS} cps`)
+  const tgt = await pickRenderer()
+  log(`target ${tgt.url}`)
+  const cdp = await connect(tgt.webSocketDebuggerUrl)
+  await cdp.send('Runtime.enable')
+  await cdp.send('Performance.enable')
+  await cdp.send('DOM.enable')
+
+  const focused = await focusComposer(cdp)
+  if (!focused) {
+    console.error('composer not focusable')
+    process.exit(2)
+  }
+
+  await forceGCAndSettle(cdp)
+  const baseline = await snapshotCounts(cdp)
+  log('baseline:', JSON.stringify(baseline))
+
+  const text = genText(CHARS)
+  const history = [{ round: 0, ...baseline, charsTyped: 0 }]
+
+  for (let r = 1; r <= ROUNDS; r++) {
+    await typeChars(cdp, text, CPS)
+    await new Promise(res => setTimeout(res, 200))
+    await clearComposer(cdp)
+    await forceGCAndSettle(cdp)
+    const snap = await snapshotCounts(cdp)
+    snap.charsTyped = r * CHARS
+    snap.round = r
+    history.push(snap)
+    log(
+      `round ${r}: heap=${snap.heapUsedMB.toFixed(1)}MB ` +
+        `nodes=${snap.nodes} listeners=${snap.jsListeners} ` +
+        `domNodes=${Math.round(snap.docNodes)} ` +
+        `layoutCount=${snap.layoutCount} ` +
+        `Δheap=+${(snap.heapUsedMB - baseline.heapUsedMB).toFixed(2)}MB ` +
+        `Δnodes=+${snap.nodes - baseline.nodes} ` +
+        `Δlisteners=+${snap.jsListeners - baseline.jsListeners}`
+    )
+  }
+
+  console.log('\n=== GROWTH PER ROUND (averaged over last 5 rounds) ===')
+  const tail = history.slice(-5)
+  const first = tail[0]
+  const last = tail[tail.length - 1]
+  const rounds = last.round - first.round
+  const cells = ['heapUsedMB', 'nodes', 'jsListeners', 'docNodes', 'layoutCount']
+  for (const c of cells) {
+    const delta = last[c] - first[c]
+    const per = delta / Math.max(1, rounds)
+    const perChar = delta / Math.max(1, rounds * CHARS)
+    console.log(`  ${c.padEnd(16)}  Δtotal=${delta.toFixed(2).padStart(10)}  /round=${per.toFixed(2).padStart(8)}  /char=${perChar.toFixed(4).padStart(8)}`)
+  }
+
+  writeFileSync('/tmp/hermes-leak-history.json', JSON.stringify(history, null, 2))
+  log('wrote /tmp/hermes-leak-history.json')
+  cdp.close()
+}
+
+main().catch(e => {
+  console.error('[leak] fatal:', e.stack ?? e.message)
+  process.exit(1)
+})

apps/desktop/scripts/measure-jump.mjs

@@ -0,0 +1,108 @@
+// Measure scroll position before and after Enter on a long thread.
+// The user's complaint: pressing Enter to submit makes the view "jump up".
+//
+// Steps:
+//   1. Scroll to the bottom of the thread
+//   2. Type a short message
+//   3. Record scroll position
+//   4. Hit Enter
+//   5. Record scroll position every 10ms for 1.5s after Enter
+//   6. Report deltas
+//
+// Usage:  node apps/desktop/scripts/measure-jump.mjs
+
+const list = await (await fetch('http://127.0.0.1:9222/json/list')).json()
+const tgt = list.find(t => t.type === 'page' && t.url.startsWith('http'))
+const ws = new WebSocket(tgt.webSocketDebuggerUrl)
+let id = 0
+const pending = new Map()
+ws.addEventListener('message', ev => {
+  const m = JSON.parse(ev.data)
+  if (m.id != null && pending.has(m.id)) {
+    pending.get(m.id)(m)
+    pending.delete(m.id)
+  }
+})
+await new Promise(r => ws.addEventListener('open', r))
+const send = (m, p = {}) =>
+  new Promise(r => {
+    const i = ++id
+    pending.set(i, r)
+    ws.send(JSON.stringify({ id: i, method: m, params: p }))
+  })
+const evalP = async expr => {
+  const r = await send('Runtime.evaluate', { expression: expr, returnByValue: true })
+  if (r.result?.exceptionDetails) throw new Error(r.result.exceptionDetails.text)
+  return r.result.result.value
+}
+
+// Scroll to bottom
+await evalP(`(() => {
+  const v = document.querySelector('[data-slot="aui_thread-viewport"]')
+  if (v) v.scrollTop = v.scrollHeight
+})()`)
+await new Promise(r => setTimeout(r, 300))
+
+// Focus composer and type
+await evalP(`(() => {
+  const el = document.querySelector('[data-slot="composer-rich-input"]')
+  el.focus()
+  const r = document.createRange(); r.selectNodeContents(el); r.collapse(false)
+  window.getSelection().removeAllRanges(); window.getSelection().addRange(r)
+})()`)
+
+const text = 'short follow-up message'
+for (const c of text) {
+  await send('Input.dispatchKeyEvent', { type: 'char', text: c, unmodifiedText: c })
+  await new Promise(r => setTimeout(r, 10))
+}
+await new Promise(r => setTimeout(r, 300))
+
+// Set up sampling — sample scroll position every animation frame
+await evalP(`(() => {
+  const v = document.querySelector('[data-slot="aui_thread-viewport"]')
+  window.__jumpSamples = []
+  window.__jumpStart = performance.now()
+  const tick = () => {
+    if (!v) return
+    window.__jumpSamples.push({
+      t: performance.now() - window.__jumpStart,
+      scrollTop: v.scrollTop,
+      scrollHeight: v.scrollHeight,
+      clientHeight: v.clientHeight,
+      distFromBottom: v.scrollHeight - v.scrollTop - v.clientHeight
+    })
+    if (performance.now() - window.__jumpStart < 2000) {
+      requestAnimationFrame(tick)
+    }
+  }
+  requestAnimationFrame(tick)
+})()`)
+
+// Fire Enter
+await send('Input.dispatchKeyEvent', {
+  type: 'rawKeyDown', windowsVirtualKeyCode: 13, key: 'Enter', code: 'Enter', text: '\r', unmodifiedText: '\r'
+})
+await send('Input.dispatchKeyEvent', { type: 'keyUp', windowsVirtualKeyCode: 13, key: 'Enter', code: 'Enter' })
+
+await new Promise(r => setTimeout(r, 2200))
+
+const samples = JSON.parse(await evalP(`JSON.stringify(window.__jumpSamples || [])`))
+console.log(`\n${samples.length} samples over 2s`)
+console.log(`\n  t(ms)  scrollTop  scrollHeight  clientHeight  distFromBottom`)
+let prev = null
+for (const s of samples) {
+  const marker = prev && Math.abs(s.scrollTop - prev.scrollTop) > 5 ? '  ← jump' : ''
+  console.log(`  ${String(s.t.toFixed(0)).padStart(5)}  ${String(s.scrollTop).padStart(9)}  ${String(s.scrollHeight).padStart(12)}  ${String(s.clientHeight).padStart(12)}  ${String(s.distFromBottom).padStart(14)}${marker}`)
+  prev = s
+}
+
+// Cancel any running agent
+await evalP(`(() => {
+  for (const b of document.querySelectorAll('button')) {
+    if ((b.getAttribute('aria-label') || '').toLowerCase().includes('stop')) { b.click(); return 'stopped' }
+  }
+  return 'no-stop'
+})()`).then(r => console.log('\ncancel:', r))
+
+ws.close()

apps/desktop/scripts/measure-latency.mjs

@@ -0,0 +1,184 @@
+#!/usr/bin/env node
+// Measure end-to-end keystroke→paint latency in the Electron renderer.
+//
+// For each synthetic keystroke we record:
+//   t0 = Input.dispatchKeyEvent send time
+//   t1 = first observed mutation of [data-slot="composer-rich-input"] childList/character data
+//   t2 = first requestAnimationFrame callback after t1 (proxy for next paint)
+//
+// We use Page.startScreencast briefly to also get frame-presentation timestamps;
+// alternatively rely on rAF timing which is close enough for typing UX.
+//
+// Output: per-char latency histogram (min/p50/p95/p99/max) + samples > 16ms.
+//
+// Usage:
+//   node apps/desktop/scripts/measure-latency.mjs [--chars=100] [--cps=15] [--port=9222]
+
+import { writeFileSync } from 'node:fs'
+
+const args = Object.fromEntries(
+  process.argv.slice(2).flatMap(s => {
+    const m = s.match(/^--([^=]+)(?:=(.*))?$/)
+    return m ? [[m[1], m[2] ?? true]] : []
+  })
+)
+const PORT = Number(args.port ?? 9222)
+const CHARS = Number(args.chars ?? 100)
+const CPS = Number(args.cps ?? 15)
+
+const log = (...m) => console.log('[latency]', ...m)
+
+async function pickRenderer() {
+  const list = await (await fetch(`http://127.0.0.1:${PORT}/json/list`)).json()
+  return list.find(t => t.type === 'page' && t.url.startsWith('http'))
+}
+
+function connect(url) {
+  return new Promise((resolve, reject) => {
+    const ws = new WebSocket(url)
+    let id = 0
+    const pending = new Map()
+    const events = new Map()
+    ws.addEventListener('open', () =>
+      resolve({
+        send(method, params = {}) {
+          const myId = ++id
+          ws.send(JSON.stringify({ id: myId, method, params }))
+          return new Promise((res, rej) => pending.set(myId, { res, rej }))
+        },
+        on(method, h) {
+          if (!events.has(method)) events.set(method, [])
+          events.get(method).push(h)
+        },
+        close: () => ws.close()
+      })
+    )
+    ws.addEventListener('error', reject)
+    ws.addEventListener('message', ev => {
+      const m = JSON.parse(typeof ev.data === 'string' ? ev.data : ev.data.toString('utf8'))
+      if (m.id != null) {
+        const p = pending.get(m.id)
+        if (!p) return
+        pending.delete(m.id)
+        m.error ? p.rej(new Error(m.error.message)) : p.res(m.result)
+      } else if (m.method) {
+        ;(events.get(m.method) ?? []).forEach(h => h(m.params))
+      }
+    })
+  })
+}
+
+async function evalInPage(cdp, expr) {
+  const r = await cdp.send('Runtime.evaluate', { expression: expr, returnByValue: true })
+  if (r.exceptionDetails) throw new Error(r.exceptionDetails.text)
+  return r.result.value
+}
+
+async function main() {
+  const tgt = await pickRenderer()
+  log(`target ${tgt.url}`)
+  const cdp = await connect(tgt.webSocketDebuggerUrl)
+  await cdp.send('Runtime.enable')
+
+  await evalInPage(
+    cdp,
+    `(() => {
+      const el = document.querySelector('[data-slot="composer-rich-input"]')
+      if (!el) return false
+      el.focus()
+      const range = document.createRange()
+      range.selectNodeContents(el)
+      range.collapse(false)
+      const sel = window.getSelection()
+      sel.removeAllRanges()
+      sel.addRange(range)
+      window.__keypressTimings = []
+      window.__pendingKey = null
+      // Observe the composer for content/text changes; record the time relative
+      // to the most recent simulated keypress timestamp set on window.__pendingKey.
+      const obs = new MutationObserver(() => {
+        const start = window.__pendingKey
+        if (start === null) return
+        const mutationT = performance.now()
+        window.__pendingKey = null
+        requestAnimationFrame(() => {
+          const paintT = performance.now()
+          window.__keypressTimings.push({
+            start, mutationT, paintT,
+            mutationLatency: mutationT - start,
+            paintLatency: paintT - start
+          })
+        })
+      })
+      obs.observe(el, { childList: true, subtree: true, characterData: true })
+      window.__keystrokeObserver = obs
+      return true
+    })()`
+  )
+
+  const lorem =
+    'the quick brown fox jumps over the lazy dog while typing into this composer feels like wading through molasses on a hot afternoon. '
+  let text = ''
+  while (text.length < CHARS) text += lorem
+  text = text.slice(0, CHARS)
+
+  const intervalMs = Math.max(1, Math.round(1000 / CPS))
+  const start = Date.now()
+  for (let i = 0; i < text.length; i++) {
+    // Mark the keypress time inside the page so it's measured from the same clock.
+    await evalInPage(cdp, `window.__pendingKey = performance.now()`)
+    await cdp.send('Input.dispatchKeyEvent', { type: 'char', text: text[i], unmodifiedText: text[i] })
+    const expected = start + (i + 1) * intervalMs
+    const wait = expected - Date.now()
+    if (wait > 0) await new Promise(r => setTimeout(r, wait))
+  }
+
+  await new Promise(r => setTimeout(r, 500))
+  const samples = await evalInPage(cdp, `window.__keypressTimings`)
+  log(`${samples.length} keystroke samples measured out of ${text.length} typed`)
+
+  // Clear composer for next run
+  await evalInPage(cdp, `
+    (() => {
+      const el = document.querySelector('[data-slot="composer-rich-input"]')
+      if (el) { el.innerHTML = ''; el.dispatchEvent(new InputEvent('input', { bubbles: true, inputType: 'deleteContentBackward' })) }
+      window.__keystrokeObserver?.disconnect()
+    })()
+  `)
+
+  const mutLat = samples.map(s => s.mutationLatency).sort((a, b) => a - b)
+  const paintLat = samples.map(s => s.paintLatency).sort((a, b) => a - b)
+  const stat = arr => ({
+    n: arr.length,
+    min: arr[0]?.toFixed(2),
+    p50: arr[Math.floor(arr.length * 0.5)]?.toFixed(2),
+    p90: arr[Math.floor(arr.length * 0.9)]?.toFixed(2),
+    p95: arr[Math.floor(arr.length * 0.95)]?.toFixed(2),
+    p99: arr[Math.floor(arr.length * 0.99)]?.toFixed(2),
+    max: arr[arr.length - 1]?.toFixed(2),
+    mean: arr.length ? (arr.reduce((s, x) => s + x, 0) / arr.length).toFixed(2) : 0
+  })
+
+  console.log('\n=== keypress → mutation latency (ms) ===')
+  console.log(' ', stat(mutLat))
+  console.log('\n=== keypress → next rAF (≈paint) latency (ms) ===')
+  console.log(' ', stat(paintLat))
+
+  const slow = samples.filter(s => s.paintLatency > 16)
+  console.log(`\n=== ${slow.length}/${samples.length} keystrokes >16ms (one frame) ===`)
+  if (slow.length) {
+    const slowSorted = [...slow].sort((a, b) => b.paintLatency - a.paintLatency).slice(0, 10)
+    for (const s of slowSorted) {
+      console.log(`  paint=${s.paintLatency.toFixed(1)}ms  mut=${s.mutationLatency.toFixed(1)}ms  at t=${s.start.toFixed(0)}`)
+    }
+  }
+
+  writeFileSync('/tmp/hermes-latency-samples.json', JSON.stringify(samples, null, 2))
+
+  cdp.close()
+}
+
+main().catch(e => {
+  console.error('[latency] fatal:', e.stack ?? e.message)
+  process.exit(1)
+})

apps/desktop/scripts/measure-real-stream.mjs

@@ -0,0 +1,252 @@
+// REAL streaming measurement — no React internals.
+//
+// Measures:
+//   1) rAF frame intervals during a verified live stream (long-frame histogram)
+//   2) MutationObserver: how often does the live assistant message mutate, what's the budget per mutation
+//   3) Text length growth rate (chars/sec)
+//   4) PerformanceObserver `longtask` entries (any task > 50ms blocks input)
+//
+// Detects REAL stream by waiting for assistant-message DOM count to grow past baseline.
+// Does NOT cancel — lets the stream run to completion or hits TIMEOUT_MS.
+
+const CDP_HTTP = 'http://127.0.0.1:9222'
+const PROMPT = process.env.PROMPT || 'count from 1 to 80, one number per line'
+const TIMEOUT_MS = Number(process.env.TIMEOUT_MS || 60000)
+
+async function getTarget() {
+  const list = await (await fetch(`${CDP_HTTP}/json`)).json()
+  const t = list.find((t) => t.type === 'page' && /5174/.test(t.url))
+  if (!t) throw new Error('renderer not found')
+  return t
+}
+
+class CDP {
+  constructor(ws) { this.ws = ws; this.id = 0; this.pending = new Map() }
+  static async open(url) {
+    const ws = new WebSocket(url)
+    await new Promise((r, j) => {
+      ws.addEventListener('open', r, { once: true })
+      ws.addEventListener('error', (e) => j(e), { once: true })
+    })
+    const cdp = new CDP(ws)
+    ws.addEventListener('message', (event) => {
+      const m = JSON.parse(event.data.toString())
+      if (m.id != null && cdp.pending.has(m.id)) {
+        const { resolve, reject } = cdp.pending.get(m.id)
+        cdp.pending.delete(m.id)
+        if (m.error) reject(new Error(m.error.message))
+        else resolve(m.result)
+      }
+    })
+    return cdp
+  }
+  send(method, params) {
+    const id = ++this.id
+    return new Promise((res, rej) => {
+      this.pending.set(id, { resolve: res, reject: rej })
+      this.ws.send(JSON.stringify({ id, method, params }))
+    })
+  }
+  async eval(expr) {
+    const r = await this.send('Runtime.evaluate', { expression: expr, returnByValue: true, awaitPromise: true })
+    if (r.exceptionDetails) throw new Error(r.exceptionDetails.exception?.description || 'eval')
+    return r.result.value
+  }
+  close() { this.ws.close() }
+}
+
+async function main() {
+  const target = await getTarget()
+  const cdp = await CDP.open(target.webSocketDebuggerUrl)
+
+  // Install recorders.
+  await cdp.eval(`
+    (() => {
+      // rAF frame intervals
+      window.__FT__ = { times: [], stop: false }
+      let last = performance.now()
+      const tick = () => {
+        if (window.__FT__.stop) return
+        const now = performance.now()
+        window.__FT__.times.push(now - last)
+        last = now
+        requestAnimationFrame(tick)
+      }
+      requestAnimationFrame(tick)
+
+      // longtask observer
+      window.__LT__ = { entries: [], stop: false }
+      try {
+        const po = new PerformanceObserver((list) => {
+          if (window.__LT__.stop) return
+          for (const e of list.getEntries()) {
+            window.__LT__.entries.push({ name: e.name, duration: e.duration, startTime: e.startTime })
+          }
+        })
+        po.observe({ entryTypes: ['longtask'] })
+        window.__LT__.po = po
+      } catch {}
+
+      // mutation observer on streaming message
+      window.__MO__ = { mutations: [], stop: false, currentMsg: null }
+      const tryArm = () => {
+        const all = document.querySelectorAll('[data-slot="aui_assistant-message-root"]')
+        const last = all[all.length - 1]
+        if (!last || last === window.__MO__.currentMsg) return
+        window.__MO__.currentMsg = last
+        if (window.__MO__.obs) window.__MO__.obs.disconnect()
+        const obs = new MutationObserver((muts) => {
+          if (window.__MO__.stop) return
+          const t = performance.now()
+          window.__MO__.mutations.push({ t, count: muts.length, len: last.textContent.length })
+        })
+        obs.observe(last, { childList: true, subtree: true, characterData: true })
+        window.__MO__.obs = obs
+      }
+      window.__MO__.arm = tryArm
+      return 'recorders armed'
+    })()
+  `)
+
+  // Baseline
+  const base = JSON.parse(await cdp.eval(`
+    JSON.stringify({
+      assistantCount: document.querySelectorAll('[data-slot="aui_assistant-message-root"]').length,
+      busy: !!document.querySelector('[data-status="running"], [data-busy="true"]'),
+      hasComposer: !!document.querySelector('[contenteditable="true"]'),
+    })
+  `))
+  console.log('baseline:', base)
+  if (!base.hasComposer) { console.error('no composer'); cdp.close(); return }
+
+  // Type + submit
+  await cdp.eval(`
+    (() => {
+      const ed = document.querySelector('[contenteditable="true"]')
+      ed.focus()
+      document.execCommand('insertText', false, ${JSON.stringify(PROMPT)})
+      return 'typed'
+    })()
+  `)
+  const submitT0 = Date.now()
+  await cdp.eval(`
+    (() => {
+      const ed = document.querySelector('[contenteditable="true"]')
+      ed.dispatchEvent(new KeyboardEvent('keydown', { key: 'Enter', code: 'Enter', bubbles: true, cancelable: true }))
+      return 'submitted'
+    })()
+  `)
+
+  // Poll for REAL stream (assistant count > baseline). 30 seconds — accommodates
+  // slow first-token latencies on big providers.
+  let realStreamT = null
+  for (let i = 0; i < 600; i++) {
+    await new Promise((r) => setTimeout(r, 50))
+    const s = JSON.parse(await cdp.eval(`
+      JSON.stringify({
+        n: document.querySelectorAll('[data-slot="aui_assistant-message-root"]').length,
+        busy: !!document.querySelector('[data-status="running"], [data-busy="true"]'),
+        text: (() => { const a = document.querySelectorAll('[data-slot="aui_assistant-message-root"]'); return a.length ? a[a.length-1].textContent.length : 0 })()
+      })
+    `))
+    if (s.n > base.assistantCount) {
+      realStreamT = Date.now()
+      console.log('REAL stream started after', realStreamT - submitT0, 'ms — busy=', s.busy, 'text=', s.text)
+      // Arm mutation observer on the new message
+      await cdp.eval('window.__MO__.arm()')
+      break
+    }
+  }
+  if (!realStreamT) {
+    console.error('REAL STREAM NEVER STARTED')
+    cdp.close()
+    return
+  }
+
+  // Sample length growth, wait for completion or timeout
+  const samples = []
+  const start = Date.now()
+  while (Date.now() - start < TIMEOUT_MS) {
+    await new Promise((r) => setTimeout(r, 250))
+    const s = JSON.parse(await cdp.eval(`
+      JSON.stringify({
+        t: performance.now(),
+        len: (() => { const a = document.querySelectorAll('[data-slot="aui_assistant-message-root"]'); return a.length ? a[a.length-1].textContent.length : 0 })(),
+        busy: !!document.querySelector('[data-status="running"], [data-busy="true"]')
+      })
+    `))
+    samples.push(s)
+    if (!s.busy && samples.length > 4) {
+      await new Promise((r) => setTimeout(r, 300))
+      break
+    }
+  }
+
+  // Pull recordings
+  const data = JSON.parse(await cdp.eval(`
+    (() => {
+      window.__FT__.stop = true
+      window.__LT__.stop = true
+      window.__MO__.stop = true
+      try { window.__LT__.po && window.__LT__.po.disconnect() } catch {}
+      try { window.__MO__.obs && window.__MO__.obs.disconnect() } catch {}
+      return JSON.stringify({
+        frames: window.__FT__.times,
+        longtasks: window.__LT__.entries,
+        mutations: window.__MO__.mutations,
+      })
+    })()
+  `))
+
+  const { frames, longtasks, mutations } = data
+
+  // Frame histogram (filter to stream window)
+  const buckets = { '<=16.7': 0, '16.7-33': 0, '33-50': 0, '50-100': 0, '100-200': 0, '>200': 0 }
+  let frameTotal = 0
+  let maxFrame = 0
+  for (const f of frames) {
+    frameTotal += f
+    if (f > maxFrame) maxFrame = f
+    if (f <= 16.7) buckets['<=16.7']++
+    else if (f <= 33) buckets['16.7-33']++
+    else if (f <= 50) buckets['33-50']++
+    else if (f <= 100) buckets['50-100']++
+    else if (f <= 200) buckets['100-200']++
+    else buckets['>200']++
+  }
+  const avgFps = frames.length ? (frames.length / (frameTotal / 1000)).toFixed(1) : 'n/a'
+  const slowFrames = frames.filter((f) => f > 33).length
+  const veryslowFrames = frames.filter((f) => f > 100).length
+
+  // Longtask summary
+  const ltMs = longtasks.reduce((a, b) => a + b.duration, 0)
+  const ltMax = longtasks.length ? Math.max(...longtasks.map((e) => e.duration)) : 0
+
+  // Mutation rate
+  let mutTotal = mutations.length
+  let mutDurs = []
+  for (let i = 1; i < mutations.length; i++) {
+    mutDurs.push(mutations[i].t - mutations[i - 1].t)
+  }
+  mutDurs.sort((a, b) => a - b)
+  const mutP50 = mutDurs[Math.floor(mutDurs.length * 0.5)] ?? 0
+  const mutP95 = mutDurs[Math.floor(mutDurs.length * 0.95)] ?? 0
+
+  // Growth rate
+  const firstLen = samples[0]?.len ?? 0
+  const lastLen = samples[samples.length - 1]?.len ?? 0
+  const elapsedS = samples.length ? (samples[samples.length - 1].t - samples[0].t) / 1000 : 0
+  const charsPerSec = elapsedS ? ((lastLen - firstLen) / elapsedS).toFixed(1) : 'n/a'
+
+  console.log('\n=== STREAM RESULTS ===')
+  console.log('window:', (frameTotal / 1000).toFixed(1), 's | frames:', frames.length, '| avgFps:', avgFps, '| maxFrame:', maxFrame.toFixed(1), 'ms')
+  console.log('frame histogram:', buckets)
+  console.log('slow frames (>33ms):', slowFrames, '| very slow (>100ms):', veryslowFrames)
+  console.log('longtasks:', longtasks.length, 'total', ltMs.toFixed(0), 'ms — max', ltMax.toFixed(1), 'ms')
+  console.log('text grew', firstLen, '→', lastLen, 'chars (', charsPerSec, 'char/s )')
+  console.log('mutations on streaming msg:', mutTotal, '| inter-mutation p50:', mutP50.toFixed(1), 'ms', 'p95:', mutP95.toFixed(1), 'ms')
+
+  cdp.close()
+}
+
+main().catch((e) => { console.error(e); process.exit(1) })

apps/desktop/scripts/measure-submit.mjs

@@ -0,0 +1,179 @@
+#!/usr/bin/env node
+// Measure submit (Enter) latency in the composer.
+//
+// For each round:
+//   1. Focus composer, type N chars of stub text
+//   2. Mark a timestamp, fire Enter via Input.dispatchKeyEvent
+//   3. Observe: time until the composer becomes empty (submit accepted),
+//      time until the user message renders in the thread viewport,
+//      time until the optional "running…" indicator appears,
+//      time until the next frame is painted after the message renders.
+//
+// Pre-condition: a session is loaded (load via click-session.mjs first).
+// Note: this DOES talk to the real gateway/agent, so each round triggers
+//       a real prompt submission. Don't run this on a live conversation
+//       you care about — use a throwaway session.
+
+import { writeFileSync } from 'node:fs'
+
+const args = Object.fromEntries(
+  process.argv.slice(2).flatMap(s => {
+    const m = s.match(/^--([^=]+)(?:=(.*))?$/)
+    return m ? [[m[1], m[2] ?? true]] : []
+  })
+)
+const PORT = Number(args.port ?? 9222)
+const ROUNDS = Number(args.rounds ?? 3)
+
+async function pickRenderer() {
+  const list = await (await fetch(`http://127.0.0.1:${PORT}/json/list`)).json()
+  return list.find(t => t.type === 'page' && t.url.startsWith('http'))
+}
+
+function connect(url) {
+  return new Promise((resolve, reject) => {
+    const ws = new WebSocket(url)
+    let id = 0
+    const pending = new Map()
+    ws.addEventListener('open', () =>
+      resolve({
+        send(method, params = {}) {
+          const myId = ++id
+          ws.send(JSON.stringify({ id: myId, method, params }))
+          return new Promise((res, rej) => pending.set(myId, { res, rej }))
+        },
+        close: () => ws.close()
+      })
+    )
+    ws.addEventListener('error', reject)
+    ws.addEventListener('message', ev => {
+      const m = JSON.parse(typeof ev.data === 'string' ? ev.data : ev.data.toString('utf8'))
+      if (m.id != null) {
+        const p = pending.get(m.id)
+        if (!p) return
+        pending.delete(m.id)
+        m.error ? p.rej(new Error(m.error.message)) : p.res(m.result)
+      }
+    })
+  })
+}
+
+async function evalP(cdp, expr) {
+  const r = await cdp.send('Runtime.evaluate', { expression: expr, returnByValue: true, awaitPromise: true })
+  if (r.exceptionDetails) throw new Error(r.exceptionDetails.text)
+  return r.result.value
+}
+
+async function focusAndType(cdp, text) {
+  await evalP(cdp, `
+    (() => {
+      const el = document.querySelector('[data-slot="composer-rich-input"]')
+      if (!el) return
+      el.focus()
+      const range = document.createRange()
+      range.selectNodeContents(el)
+      range.collapse(false)
+      const sel = window.getSelection()
+      sel.removeAllRanges()
+      sel.addRange(range)
+    })()
+  `)
+  for (const c of text) {
+    await cdp.send('Input.dispatchKeyEvent', { type: 'char', text: c, unmodifiedText: c })
+    await new Promise(r => setTimeout(r, 8))
+  }
+}
+
+async function submitAndMeasure(cdp, timeoutMs = 5000) {
+  // Install observers, record submit time as performance.now() inside the page,
+  // and wait for all milestones.
+  return await evalP(cdp, `
+    new Promise((resolve) => {
+      const composer = document.querySelector('[data-slot="composer-rich-input"]')
+      const threadRoot = document.querySelector('[data-slot="aui_thread-content"]') ||
+                         document.querySelector('[data-slot="aui_thread-viewport"]')
+      const startMessageCount = threadRoot ? threadRoot.querySelectorAll('[data-slot="aui_turn-pair"], [data-slot="aui_message"]').length : 0
+      const startComposerText = composer ? composer.innerText : ''
+
+      const milestones = { start: performance.now() }
+      let done = false
+      const finish = (reason) => {
+        if (done) return
+        done = true
+        clearInterval(poll); clearTimeout(timer)
+        composerObs.disconnect()
+        threadObs?.disconnect()
+        milestones.reason = reason
+        milestones.end = performance.now()
+        milestones.totalMs = milestones.end - milestones.start
+        resolve(milestones)
+      }
+
+      const composerObs = new MutationObserver(() => {
+        if (!milestones.composerClearedMs && composer && composer.innerText.length === 0) {
+          milestones.composerClearedMs = performance.now() - milestones.start
+        }
+      })
+      composer && composerObs.observe(composer, { childList: true, subtree: true, characterData: true })
+
+      let threadObs = null
+      if (threadRoot) {
+        threadObs = new MutationObserver(() => {
+          const c = threadRoot.querySelectorAll('[data-slot="aui_turn-pair"], [data-slot="aui_message"]').length
+          if (!milestones.userMessageRenderedMs && c > startMessageCount) {
+            milestones.userMessageRenderedMs = performance.now() - milestones.start
+            requestAnimationFrame(() => {
+              milestones.userMessagePaintMs = performance.now() - milestones.start
+              finish('paint')
+            })
+          }
+        })
+        threadObs.observe(threadRoot, { childList: true, subtree: true })
+      }
+
+      const poll = setInterval(() => {
+        if (milestones.composerClearedMs && !milestones.userMessageRenderedMs &&
+            performance.now() - milestones.start > 2000) {
+          finish('timeout-after-clear')
+        }
+      }, 100)
+      const timer = setTimeout(() => finish('timeout-overall'), ${timeoutMs})
+
+      // Send Enter immediately
+      window.dispatchEvent(new KeyboardEvent('keydown'))  // no-op marker
+      const enterEv = new KeyboardEvent('keydown', { key: 'Enter', code: 'Enter', bubbles: true, cancelable: true })
+      composer?.dispatchEvent(enterEv)
+    })
+  `)
+}
+
+async function main() {
+  const tgt = await pickRenderer()
+  console.log('target', tgt.url)
+  const cdp = await connect(tgt.webSocketDebuggerUrl)
+  await cdp.send('Runtime.enable')
+
+  const samples = []
+  for (let i = 1; i <= ROUNDS; i++) {
+    await focusAndType(cdp, `latency test ${i} ${'x'.repeat(40)}`)
+    await new Promise(r => setTimeout(r, 300))
+    const result = await submitAndMeasure(cdp, 4000)
+    samples.push({ round: i, ...result })
+    console.log(
+      `r${i}: clear=${(result.composerClearedMs ?? -1).toFixed?.(0) ?? '?'}ms ` +
+        `userMsg=${(result.userMessageRenderedMs ?? -1).toFixed?.(0) ?? '?'}ms ` +
+        `paint=${(result.userMessagePaintMs ?? -1).toFixed?.(0) ?? '?'}ms ` +
+        `reason=${result.reason}`
+    )
+    // wait for any agent activity to finish before next round so we're not piling up
+    await new Promise(r => setTimeout(r, 4000))
+  }
+  writeFileSync('/tmp/hermes-submit-latency.json', JSON.stringify(samples, null, 2))
+  console.log('\nwrote /tmp/hermes-submit-latency.json')
+  cdp.close()
+}
+
+main().catch(e => {
+  console.error('fatal:', e.stack ?? e.message)
+  process.exit(1)
+})

apps/desktop/scripts/measure-synthetic-stream.mjs

@@ -0,0 +1,322 @@
+// Measure render cost of a synthetic stream driven through the live $messages atom.
+//
+// Why synthetic: the user's LLM credits are depleted; we can't fire a real stream.
+// The synthetic stream exercises the exact same React pipeline (assistant-ui runtime →
+// repository.addOrUpdateMessage → MessagePrimitive re-render → markdown reflow) as a
+// real stream. The only thing it does NOT exercise is the gateway → SSE → optimistic-
+// merge path, which is orthogonal to the rendering question.
+//
+// What we record:
+//   1) rAF frame intervals (long-frame histogram; >33ms = perceived jank, >100ms = bad)
+//   2) PerformanceObserver `longtask` entries (task >50ms blocks input)
+//   3) MutationObserver: per-message mutation count & inter-mutation latency
+//   4) Optional: typing latency overlay — typing into composer while streaming
+//
+// Output is plain text suitable for terminal + a JSON sidecar for diffing across runs.
+
+import { writeFileSync } from 'node:fs'
+
+const CDP_HTTP = 'http://127.0.0.1:9222'
+const TOKENS = Number(process.env.TOKENS || 300)
+const INTERVAL_MS = Number(process.env.INTERVAL_MS || 16)
+// Upstream flush throttle to apply in the synthetic driver. Mirrors what the
+// real gateway path does in `use-message-stream.scheduleDeltaFlush`. 0
+// disables (worst-case, every token = one React commit).
+const FLUSH_MIN_MS = Number(process.env.FLUSH_MIN_MS || 0)
+const CHUNK = process.env.CHUNK || 'lorem ipsum '
+const TYPE_WHILE_STREAMING = process.env.TYPE_WHILE_STREAMING === '1'
+const LABEL = process.env.LABEL || 'baseline'
+const OUT = process.env.OUT || `frame-times-${LABEL}.json`
+
+async function getTarget() {
+  const list = await (await fetch(`${CDP_HTTP}/json`)).json()
+  const t = list.find((t) => t.type === 'page' && /5174/.test(t.url))
+  if (!t) throw new Error('renderer not found')
+  return t
+}
+
+class CDP {
+  constructor(ws) { this.ws = ws; this.id = 0; this.pending = new Map() }
+  static async open(url) {
+    const ws = new WebSocket(url)
+    await new Promise((r, j) => {
+      ws.addEventListener('open', r, { once: true })
+      ws.addEventListener('error', (e) => j(e), { once: true })
+    })
+    const cdp = new CDP(ws)
+    ws.addEventListener('message', (ev) => {
+      const m = JSON.parse(ev.data.toString())
+      if (m.id != null && cdp.pending.has(m.id)) {
+        const { resolve, reject } = cdp.pending.get(m.id)
+        cdp.pending.delete(m.id)
+        if (m.error) reject(new Error(m.error.message))
+        else resolve(m.result)
+      }
+    })
+    return cdp
+  }
+  send(method, params) {
+    const id = ++this.id
+    return new Promise((res, rej) => {
+      this.pending.set(id, { resolve: res, reject: rej })
+      this.ws.send(JSON.stringify({ id, method, params }))
+    })
+  }
+  async eval(expr) {
+    const r = await this.send('Runtime.evaluate', { expression: expr, returnByValue: true, awaitPromise: true })
+    if (r.exceptionDetails) throw new Error(r.exceptionDetails.exception?.description || 'eval')
+    return r.result.value
+  }
+  close() { this.ws.close() }
+}
+
+function pct(arr, p) {
+  if (!arr.length) return 0
+  const i = Math.min(arr.length - 1, Math.floor(arr.length * p))
+  return arr[i]
+}
+
+async function main() {
+  const target = await getTarget()
+  const cdp = await CDP.open(target.webSocketDebuggerUrl)
+
+  // Sanity check driver is loaded.
+  const probeOk = await cdp.eval('!!window.__PERF_DRIVE__ && !!window.__PERF_DRIVE__.stream')
+  if (!probeOk) {
+    console.error('__PERF_DRIVE__ not on window — did you reload the renderer after editing perf-probe.tsx?')
+    cdp.close()
+    process.exit(2)
+  }
+
+  // Install recorders.
+  await cdp.eval(`
+    (() => {
+      window.__FT__ = { times: [], stop: false }
+      let last = performance.now()
+      const tick = () => {
+        if (window.__FT__.stop) return
+        const now = performance.now()
+        window.__FT__.times.push(now - last)
+        last = now
+        requestAnimationFrame(tick)
+      }
+      requestAnimationFrame(tick)
+
+      window.__LT__ = { entries: [], stop: false }
+      try {
+        const po = new PerformanceObserver((list) => {
+          if (window.__LT__.stop) return
+          for (const e of list.getEntries()) {
+            window.__LT__.entries.push({ name: e.name, duration: e.duration, startTime: e.startTime })
+          }
+        })
+        po.observe({ entryTypes: ['longtask'] })
+        window.__LT__.po = po
+      } catch {}
+
+      window.__MO__ = { mutations: [], stop: false, currentMsg: null }
+      const arm = () => {
+        const all = document.querySelectorAll('[data-slot="aui_assistant-message-root"]')
+        const last = all[all.length - 1]
+        if (!last || last === window.__MO__.currentMsg) return
+        window.__MO__.currentMsg = last
+        if (window.__MO__.obs) window.__MO__.obs.disconnect()
+        const obs = new MutationObserver((muts) => {
+          if (window.__MO__.stop) return
+          const t = performance.now()
+          window.__MO__.mutations.push({ t, count: muts.length, len: last.textContent.length })
+        })
+        obs.observe(last, { childList: true, subtree: true, characterData: true })
+        window.__MO__.obs = obs
+      }
+      window.__MO__.arm = arm
+
+      // Optional: typing observer — fires keystroke timings if asked.
+      window.__TYP__ = { times: [], stop: false, lastKey: 0 }
+      return 'recorders armed'
+    })()
+  `)
+
+  // Baseline state.
+  const base = JSON.parse(await cdp.eval(`
+    JSON.stringify({
+      assistantCount: document.querySelectorAll('[data-slot="aui_assistant-message-root"]').length,
+      atomCount: window.__PERF_DRIVE__.snapshotMsgs()
+    })
+  `))
+  console.log('baseline:', base)
+
+  // Drive a synthetic stream.
+  const streamStart = Date.now()
+  await cdp.eval(`window.__PERF_DRIVE__.stream({ chunk: ${JSON.stringify(CHUNK)}, intervalMs: ${INTERVAL_MS}, totalTokens: ${TOKENS}, flushMinMs: ${FLUSH_MIN_MS} })`)
+
+  // After the first paint, arm MO on the new message.
+  await new Promise((r) => setTimeout(r, 200))
+  await cdp.eval('window.__MO__.arm()')
+
+  // Optional: type while streaming.
+  if (TYPE_WHILE_STREAMING) {
+    await new Promise((r) => setTimeout(r, 400))
+    await cdp.eval(`(() => {
+      const ed = document.querySelector('[contenteditable="true"]')
+      ed.focus()
+      window.__TYP__.startedAt = performance.now()
+      const text = 'the quick brown fox jumps over the lazy dog '
+      let i = 0
+      const tick = () => {
+        if (i >= text.length) return
+        const t0 = performance.now()
+        document.execCommand('insertText', false, text[i])
+        // requestAnimationFrame to wait for next paint
+        requestAnimationFrame(() => {
+          window.__TYP__.times.push(performance.now() - t0)
+        })
+        i++
+        setTimeout(tick, 60)
+      }
+      tick()
+      return 'typing'
+    })()`)
+  }
+
+  // Wait for stream to complete + small grace.
+  const expectedMs = TOKENS * INTERVAL_MS + 1500
+  await new Promise((r) => setTimeout(r, expectedMs))
+
+  // Pull recordings.
+  const data = JSON.parse(await cdp.eval(`
+    (() => {
+      window.__FT__.stop = true
+      window.__LT__.stop = true
+      window.__MO__.stop = true
+      window.__TYP__.stop = true
+      try { window.__LT__.po && window.__LT__.po.disconnect() } catch {}
+      try { window.__MO__.obs && window.__MO__.obs.disconnect() } catch {}
+      return JSON.stringify({
+        frames: window.__FT__.times,
+        longtasks: window.__LT__.entries,
+        mutations: window.__MO__.mutations,
+        typing: window.__TYP__.times,
+        finalText: (() => { const a = document.querySelectorAll('[data-slot="aui_assistant-message-root"]'); return a.length ? a[a.length-1].textContent.length : 0 })()
+      })
+    })()
+  `))
+
+  // Reset DOM back to baseline so we don't accumulate fake messages.
+  await cdp.eval('window.__PERF_DRIVE__.reset()')
+
+  // Analysis (trim warm-up: drop frames before first mutation timestamp).
+  const firstMut = data.mutations[0]?.t
+  const frames = data.frames
+
+  // Sum durations to figure out when each frame happened (relative to recorder start).
+  const frameTimeline = []
+  let acc = 0
+  for (const f of frames) { acc += f; frameTimeline.push(acc) }
+
+  // Mutations are in performance.now() ms; frames started recording when we installed
+  // the recorder (before stream). To align: compute total stream window from frames
+  // after mutation activity began. Simpler heuristic: drop first 500ms of frames as warm-up.
+  const WARMUP_MS = 500
+  let dropIdx = 0
+  for (let i = 0; i < frames.length; i++) {
+    if (frameTimeline[i] >= WARMUP_MS) { dropIdx = i; break }
+  }
+  const streamFrames = frames.slice(dropIdx)
+
+  const buckets = { '<=16.7': 0, '16.7-33': 0, '33-50': 0, '50-100': 0, '100-200': 0, '>200': 0 }
+  let frameTotal = 0
+  let maxFrame = 0
+  for (const f of streamFrames) {
+    frameTotal += f
+    if (f > maxFrame) maxFrame = f
+    if (f <= 16.7) buckets['<=16.7']++
+    else if (f <= 33) buckets['16.7-33']++
+    else if (f <= 50) buckets['33-50']++
+    else if (f <= 100) buckets['50-100']++
+    else if (f <= 200) buckets['100-200']++
+    else buckets['>200']++
+  }
+  const sortedFrames = [...streamFrames].sort((a, b) => a - b)
+  const fAvgFps = streamFrames.length ? (streamFrames.length / (frameTotal / 1000)).toFixed(1) : 'n/a'
+  const fP50 = pct(sortedFrames, 0.5).toFixed(1)
+  const fP95 = pct(sortedFrames, 0.95).toFixed(1)
+  const fP99 = pct(sortedFrames, 0.99).toFixed(1)
+  const slowFrames = streamFrames.filter((f) => f > 33).length
+  const veryslowFrames = streamFrames.filter((f) => f > 100).length
+
+  const ltDur = data.longtasks.map((e) => e.duration).sort((a, b) => a - b)
+  const ltMs = ltDur.reduce((a, b) => a + b, 0)
+  const ltMax = ltDur.length ? ltDur[ltDur.length - 1] : 0
+  const ltP95 = pct(ltDur, 0.95)
+
+  // Mutation cadence.
+  const mutDurs = []
+  for (let i = 1; i < data.mutations.length; i++) mutDurs.push(data.mutations[i].t - data.mutations[i - 1].t)
+  mutDurs.sort((a, b) => a - b)
+  const mutP50 = pct(mutDurs, 0.5)
+  const mutP95 = pct(mutDurs, 0.95)
+  const mutMax = mutDurs.length ? mutDurs[mutDurs.length - 1] : 0
+
+  // Typing latency (optional).
+  let typingSummary = null
+  if (TYPE_WHILE_STREAMING && data.typing.length) {
+    const t = [...data.typing].sort((a, b) => a - b)
+    typingSummary = {
+      n: t.length,
+      p50: pct(t, 0.5).toFixed(1),
+      p95: pct(t, 0.95).toFixed(1),
+      max: t[t.length - 1].toFixed(1)
+    }
+  }
+
+  const result = {
+    label: LABEL,
+    timestamp: new Date().toISOString(),
+    config: { TOKENS, INTERVAL_MS, CHUNK, TYPE_WHILE_STREAMING, FLUSH_MIN_MS },
+    streamWallMs: Date.now() - streamStart,
+    frames: {
+      total: streamFrames.length,
+      avgFps: fAvgFps,
+      windowS: (frameTotal / 1000).toFixed(1),
+      p50: fP50,
+      p95: fP95,
+      p99: fP99,
+      max: maxFrame.toFixed(1),
+      slow33: slowFrames,
+      veryslow100: veryslowFrames,
+      histogram: buckets
+    },
+    longtasks: {
+      n: data.longtasks.length,
+      totalMs: ltMs.toFixed(0),
+      maxMs: ltMax.toFixed(1),
+      p95Ms: ltP95.toFixed(1)
+    },
+    mutations: {
+      n: data.mutations.length,
+      finalTextLen: data.finalText,
+      interMutP50ms: mutP50.toFixed(1),
+      interMutP95ms: mutP95.toFixed(1),
+      interMutMaxMs: mutMax.toFixed(1)
+    },
+    typing: typingSummary
+  }
+
+  writeFileSync(OUT, JSON.stringify(result, null, 2))
+
+  console.log('\n=== SYNTHETIC STREAM RESULTS ===')
+  console.log('label:', LABEL, '| tokens:', TOKENS, '@', INTERVAL_MS, 'ms')
+  console.log('streamWallMs:', result.streamWallMs)
+  console.log('FRAMES: avgFps', fAvgFps, '| p50', fP50, 'ms | p95', fP95, 'ms | p99', fP99, 'ms | max', maxFrame.toFixed(1), 'ms')
+  console.log('FRAMES histogram:', buckets)
+  console.log('FRAMES slow(>33):', slowFrames, '/ veryslow(>100):', veryslowFrames, 'of', streamFrames.length)
+  console.log('LONGTASKS:', data.longtasks.length, '| total', ltMs.toFixed(0), 'ms | max', ltMax.toFixed(1), 'ms | p95', ltP95.toFixed(1), 'ms')
+  console.log('MUTATIONS:', data.mutations.length, '| finalLen', data.finalText, 'chars | inter p50', mutP50.toFixed(1), 'ms | p95', mutP95.toFixed(1), 'ms')
+  if (typingSummary) console.log('TYPING-WHILE-STREAMING latency: p50', typingSummary.p50, 'ms | p95', typingSummary.p95, 'ms | n=', typingSummary.n)
+  console.log('written to', OUT)
+
+  cdp.close()
+}
+
+main().catch((e) => { console.error(e); process.exit(1) })

apps/desktop/scripts/notarize-artifact.cjs

@@ -0,0 +1,77 @@
+const fs = require('node:fs')
+const os = require('node:os')
+const path = require('node:path')
+const { execFile } = require('node:child_process')
+
+function run(command, args) {
+  return new Promise((resolve, reject) => {
+    execFile(command, args, (error, stdout, stderr) => {
+      if (error) {
+        // Intentionally omit args from the rejection message: callers pass
+        // notarization credentials (key id, issuer, key file path) here, and
+        // surfacing them in error output would land in CI logs.
+        reject(new Error(`${command} failed: ${stderr?.trim() || stdout?.trim() || error.message}`))
+        return
+      }
+      resolve()
+    })
+  })
+}
+
+function inlineKeyLooksValid(value) {
+  return value.includes('BEGIN PRIVATE KEY') && value.includes('END PRIVATE KEY')
+}
+
+function resolveApiKeyPath(rawValue) {
+  const value = String(rawValue || '').trim()
+  if (!value) return { keyPath: '', cleanup: () => {} }
+
+  if (fs.existsSync(value)) {
+    return { keyPath: value, cleanup: () => {} }
+  }
+
+  if (!inlineKeyLooksValid(value)) {
+    throw new Error('APPLE_API_KEY must be a file path or inline .p8 key content')
+  }
+
+  const tempPath = path.join(os.tmpdir(), `hermes-notary-${Date.now()}-${process.pid}.p8`)
+  fs.writeFileSync(tempPath, value, 'utf8')
+  return {
+    keyPath: tempPath,
+    cleanup: () => fs.rmSync(tempPath, { force: true })
+  }
+}
+
+async function main() {
+  const artifactPath = process.argv[2]
+  if (!artifactPath || !fs.existsSync(artifactPath)) {
+    throw new Error(`Missing artifact to notarize: ${artifactPath || '(none)'}`)
+  }
+
+  const profile = String(process.env.APPLE_NOTARY_PROFILE || '').trim()
+  if (profile) {
+    await run('xcrun', ['notarytool', 'submit', artifactPath, '--keychain-profile', profile, '--wait'])
+    await run('xcrun', ['stapler', 'staple', '-v', artifactPath])
+    return
+  }
+
+  const keyId = String(process.env.APPLE_API_KEY_ID || '').trim()
+  const issuer = String(process.env.APPLE_API_ISSUER || '').trim()
+  const rawApiKey = process.env.APPLE_API_KEY
+  if (!rawApiKey || !keyId || !issuer) {
+    throw new Error('APPLE_API_KEY, APPLE_API_KEY_ID, and APPLE_API_ISSUER are required')
+  }
+
+  const { keyPath, cleanup } = resolveApiKeyPath(rawApiKey)
+  try {
+    await run('xcrun', ['notarytool', 'submit', artifactPath, '--key', keyPath, '--key-id', keyId, '--issuer', issuer, '--wait'])
+    await run('xcrun', ['stapler', 'staple', '-v', artifactPath])
+  } finally {
+    cleanup()
+  }
+}
+
+main().catch(() => {
+  console.error('Notarization failed. Check configuration and command output in secure CI logs.')
+  process.exit(1)
+})

apps/desktop/scripts/notarize.cjs

@@ -0,0 +1,100 @@
+const fs = require('node:fs')
+const os = require('node:os')
+const path = require('node:path')
+const { execFile } = require('node:child_process')
+
+function run(command, args) {
+  return new Promise((resolve, reject) => {
+    execFile(command, args, (error, stdout, stderr) => {
+      if (error) {
+        reject(
+          new Error(
+            `${command} ${args.join(' ')} failed: ${stderr?.trim() || stdout?.trim() || error.message}`
+          )
+        )
+        return
+      }
+      resolve({ stdout, stderr })
+    })
+  })
+}
+
+function inlineKeyLooksValid(value) {
+  return value.includes('BEGIN PRIVATE KEY') && value.includes('END PRIVATE KEY')
+}
+
+function resolveApiKeyPath(rawValue) {
+  const value = String(rawValue || '').trim()
+  if (!value) return { keyPath: '', cleanup: () => {} }
+
+  if (fs.existsSync(value)) {
+    return { keyPath: value, cleanup: () => {} }
+  }
+
+  if (!inlineKeyLooksValid(value)) {
+    throw new Error('APPLE_API_KEY must be a file path or inline .p8 key content')
+  }
+
+  const tempPath = path.join(os.tmpdir(), `hermes-notary-${Date.now()}-${process.pid}.p8`)
+  fs.writeFileSync(tempPath, value, 'utf8')
+  return {
+    keyPath: tempPath,
+    cleanup: () => {
+      try {
+        fs.rmSync(tempPath, { force: true })
+      } catch {
+        // Best-effort cleanup.
+      }
+    }
+  }
+}
+
+exports.default = async function notarize(context) {
+  const { electronPlatformName, appOutDir, packager } = context
+  if (electronPlatformName !== 'darwin') return
+
+  const appName = packager.appInfo.productFilename
+  const appPath = path.join(appOutDir, `${appName}.app`)
+  if (!fs.existsSync(appPath)) {
+    throw new Error(`Cannot notarize missing app bundle: ${appPath}`)
+  }
+
+  const profile = String(process.env.APPLE_NOTARY_PROFILE || '').trim()
+  if (profile) {
+    const zipPath = path.join(appOutDir, `${appName}.zip`)
+    await run('ditto', ['-c', '-k', '--sequesterRsrc', '--keepParent', appPath, zipPath])
+    await run('xcrun', ['notarytool', 'submit', zipPath, '--keychain-profile', profile, '--wait'])
+    await run('xcrun', ['stapler', 'staple', '-v', appPath])
+    try {
+      fs.rmSync(zipPath, { force: true })
+    } catch {
+      // Best-effort cleanup.
+    }
+    return
+  }
+
+  const keyId = String(process.env.APPLE_API_KEY_ID || '').trim()
+  const issuer = String(process.env.APPLE_API_ISSUER || '').trim()
+  const rawApiKey = process.env.APPLE_API_KEY
+  if (!rawApiKey || !keyId || !issuer) {
+    console.log(
+      'Skipping notarization: APPLE_API_KEY, APPLE_API_KEY_ID, and APPLE_API_ISSUER are not fully configured.'
+    )
+    return
+  }
+
+  const { keyPath, cleanup } = resolveApiKeyPath(rawApiKey)
+  const zipPath = path.join(appOutDir, `${appName}.zip`)
+  try {
+    await run('ditto', ['-c', '-k', '--sequesterRsrc', '--keepParent', appPath, zipPath])
+    await run('xcrun', ['notarytool', 'submit', zipPath, '--key', keyPath, '--key-id', keyId, '--issuer', issuer, '--wait'])
+    await run('xcrun', ['stapler', 'staple', '-v', appPath])
+  } finally {
+    try {
+      fs.rmSync(zipPath, { force: true })
+    } catch {
+      // Best-effort cleanup.
+    }
+    cleanup()
+  }
+}

apps/desktop/scripts/probe-renderer.mjs

@@ -0,0 +1,38 @@
+// quick probe — read state of the renderer
+const list = await (await fetch('http://127.0.0.1:9222/json/list')).json()
+const tgt = list.find(t => t.type === 'page' && t.url.startsWith('http'))
+console.log('target:', tgt?.url)
+if (!tgt) process.exit(1)
+const ws = new WebSocket(tgt.webSocketDebuggerUrl)
+let id = 0
+const pending = new Map()
+ws.addEventListener('message', ev => {
+  const m = JSON.parse(ev.data)
+  if (m.id != null && pending.has(m.id)) {
+    pending.get(m.id)(m)
+    pending.delete(m.id)
+  }
+})
+await new Promise(r => ws.addEventListener('open', r))
+const send = (method, params = {}) =>
+  new Promise(r => {
+    const i = ++id
+    pending.set(i, r)
+    ws.send(JSON.stringify({ id: i, method, params }))
+  })
+
+const r = await send('Runtime.evaluate', {
+  expression: `({
+    url: location.href,
+    title: document.title,
+    rootChildren: document.getElementById('root')?.children.length ?? 0,
+    rootInner: (document.getElementById('root')?.innerHTML ?? '').slice(0, 300),
+    hasComposer: !!document.querySelector('[data-slot="composer-rich-input"]'),
+    bootStage: (document.querySelector('[data-slot*="boot"]')?.getAttribute('data-slot')) ?? null,
+    bodyText: document.body.innerText.slice(0, 300),
+    errorCount: window.__errors?.length ?? 'n/a'
+  })`,
+  returnByValue: true
+})
+console.log('raw:', JSON.stringify(r, null, 2))
+ws.close()

apps/desktop/scripts/probe-thread.mjs

@@ -0,0 +1,40 @@
+// Probe the cloud shadows thread state — count messages, turn pairs,
+// thread height, composer state
+const list = await (await fetch('http://127.0.0.1:9222/json/list')).json()
+const tgt = list.find(t => t.type === 'page' && t.url.startsWith('http'))
+const ws = new WebSocket(tgt.webSocketDebuggerUrl)
+let id = 0
+const pending = new Map()
+ws.addEventListener('message', ev => {
+  const m = JSON.parse(ev.data)
+  if (m.id != null && pending.has(m.id)) {
+    pending.get(m.id)(m)
+    pending.delete(m.id)
+  }
+})
+await new Promise(r => ws.addEventListener('open', r))
+const send = (m, p = {}) =>
+  new Promise(r => {
+    const i = ++id
+    pending.set(i, r)
+    ws.send(JSON.stringify({ id: i, method: m, params: p }))
+  })
+
+const r = await send('Runtime.evaluate', {
+  expression: `JSON.stringify({
+    url: location.href,
+    title: document.title,
+    turnPairs: document.querySelectorAll('[data-slot="aui_turn-pair"]').length,
+    assistantMsgs: document.querySelectorAll('[data-slot="aui_assistant-message-root"]').length,
+    userMsgs: document.querySelectorAll('[data-message-role="user"], [data-slot="aui_user-message-root"]').length,
+    totalDomNodes: document.querySelectorAll('*').length,
+    threadViewportScrollHeight: document.querySelector('[data-slot="aui_thread-viewport"]')?.scrollHeight ?? null,
+    threadViewportClientHeight: document.querySelector('[data-slot="aui_thread-viewport"]')?.clientHeight ?? null,
+    threadViewportScrollTop: document.querySelector('[data-slot="aui_thread-viewport"]')?.scrollTop ?? null,
+    composer: !!document.querySelector('[data-slot="composer-rich-input"]'),
+    busy: !!document.querySelector('[aria-label*="Stop"]')
+  })`,
+  returnByValue: true
+})
+console.log(JSON.parse(r.result.result.value))
+ws.close()

apps/desktop/scripts/profile-long-stream.mjs

@@ -0,0 +1,191 @@
+#!/usr/bin/env node
+// Long-running stream profile + frame-rate timeline. Submits a prompt that
+// asks for ~30 paragraphs of output, then captures both a CPU profile and
+// a per-100ms frame counter so we can see if FPS sags as the message grows.
+
+import { writeFileSync } from 'node:fs'
+
+const args = Object.fromEntries(
+  process.argv.slice(2).flatMap(s => {
+    const m = s.match(/^--([^=]+)(?:=(.*))?$/)
+    return m ? [[m[1], m[2] ?? true]] : []
+  })
+)
+const PORT = Number(args.port ?? 9222)
+const OUT = String(args.out ?? `/tmp/hermes-long-stream-${Date.now()}`)
+const STREAM_SEC = Number(args.seconds ?? 25)
+
+async function pickRenderer() {
+  const list = await (await fetch(`http://127.0.0.1:${PORT}/json/list`)).json()
+  return list.find(t => t.type === 'page' && t.url.startsWith('http'))
+}
+
+function connect(url) {
+  return new Promise((resolve, reject) => {
+    const ws = new WebSocket(url)
+    let id = 0
+    const pending = new Map()
+    ws.addEventListener('open', () =>
+      resolve({
+        send(method, params = {}) {
+          const myId = ++id
+          ws.send(JSON.stringify({ id: myId, method, params }))
+          return new Promise((res, rej) => pending.set(myId, { res, rej }))
+        },
+        close: () => ws.close()
+      })
+    )
+    ws.addEventListener('error', reject)
+    ws.addEventListener('message', ev => {
+      const m = JSON.parse(typeof ev.data === 'string' ? ev.data : ev.data.toString('utf8'))
+      if (m.id != null) {
+        const p = pending.get(m.id)
+        if (!p) return
+        pending.delete(m.id)
+        m.error ? p.rej(new Error(m.error.message)) : p.res(m.result)
+      }
+    })
+  })
+}
+
+async function evalP(cdp, expr) {
+  const r = await cdp.send('Runtime.evaluate', { expression: expr, returnByValue: true })
+  if (r.exceptionDetails) throw new Error(r.exceptionDetails.text)
+  return r.result.value
+}
+
+async function main() {
+  const tgt = await pickRenderer()
+  console.log('target', tgt.url)
+  const cdp = await connect(tgt.webSocketDebuggerUrl)
+  await cdp.send('Runtime.enable')
+  await cdp.send('Profiler.enable')
+  await cdp.send('Performance.enable')
+
+  // Submit a long-form prompt
+  await evalP(
+    cdp,
+    `(() => {
+      const el = document.querySelector('[data-slot="composer-rich-input"]')
+      el.focus()
+      const r = document.createRange(); r.selectNodeContents(el); r.collapse(false)
+      window.getSelection().removeAllRanges(); window.getSelection().addRange(r)
+    })()`
+  )
+  const prompt = 'write 15 paragraphs about gpu memory bandwidth, memory hierarchies, roofline model, and how modern transformer inference benefits from these. include diagrams in ascii where relevant. no code. fully detailed.'
+  for (const c of prompt) {
+    await cdp.send('Input.dispatchKeyEvent', { type: 'char', text: c, unmodifiedText: c })
+    await new Promise(r => setTimeout(r, 5))
+  }
+  await new Promise(r => setTimeout(r, 200))
+  await cdp.send('Input.dispatchKeyEvent', {
+    type: 'rawKeyDown', windowsVirtualKeyCode: 13, key: 'Enter', code: 'Enter', text: '\r', unmodifiedText: '\r'
+  })
+  await cdp.send('Input.dispatchKeyEvent', { type: 'keyUp', windowsVirtualKeyCode: 13, key: 'Enter', code: 'Enter' })
+
+  console.log('waiting for assistant…')
+  let streaming = false
+  for (let i = 0; i < 100; i++) {
+    const c = await evalP(cdp, `document.querySelectorAll('[data-slot="aui_assistant-message-root"]').length`)
+    if (c > 0) { streaming = true; break }
+    await new Promise(r => setTimeout(r, 100))
+  }
+  if (!streaming) {
+    console.error('no assistant message')
+    cdp.close()
+    return
+  }
+
+  // Install a per-rAF frame counter
+  await evalP(
+    cdp,
+    `(() => {
+      window.__fpsSamples = []
+      window.__fpsT0 = performance.now()
+      window.__fpsLast = performance.now()
+      window.__fpsFrameCount = 0
+      window.__fpsHistogram = []  // {t, fps, contentLen}
+      const tick = () => {
+        const now = performance.now()
+        const dt = now - window.__fpsLast
+        window.__fpsLast = now
+        window.__fpsFrameCount++
+        window.__fpsSamples.push({ t: now - window.__fpsT0, dt })
+        if (performance.now() - window.__fpsT0 < ${STREAM_SEC * 1000}) {
+          requestAnimationFrame(tick)
+        }
+      }
+      requestAnimationFrame(tick)
+      // Bucket fps every 500ms
+      window.__fpsBucket = setInterval(() => {
+        const now = performance.now()
+        const recentCount = window.__fpsSamples.filter(s => now - window.__fpsT0 - s.t < 500).length
+        const root = document.querySelector('[data-slot="aui_thread-content"]')
+        const len = root ? root.innerText.length : 0
+        const v = document.querySelector('[data-slot="aui_thread-viewport"]')
+        window.__fpsHistogram.push({
+          t: now - window.__fpsT0,
+          frames500ms: recentCount,
+          fps: recentCount * 2,
+          contentLen: len,
+          scrollTop: v?.scrollTop ?? 0,
+          scrollHeight: v?.scrollHeight ?? 0
+        })
+      }, 500)
+    })()`
+  )
+
+  // Start CPU profile
+  await cdp.send('Profiler.setSamplingInterval', { interval: 1000 })
+  await cdp.send('Profiler.start')
+
+  await new Promise(r => setTimeout(r, STREAM_SEC * 1000))
+
+  const { profile } = await cdp.send('Profiler.stop')
+  await evalP(cdp, `clearInterval(window.__fpsBucket)`)
+
+  writeFileSync(`${OUT}.cpuprofile`, JSON.stringify(profile))
+  console.log(`cpu profile → ${OUT}.cpuprofile`)
+
+  // Pull fps histogram
+  const hist = JSON.parse(await evalP(cdp, `JSON.stringify(window.__fpsHistogram || [])`))
+  writeFileSync(`${OUT}.fps.json`, JSON.stringify(hist, null, 2))
+
+  console.log(`\n=== FPS over time ===`)
+  console.log(`  t(s)   fps   contentLen   scrollTop/scrollHeight`)
+  for (const h of hist) {
+    const bar = '█'.repeat(Math.min(40, Math.max(0, Math.round(h.fps / 2))))
+    console.log(`  ${(h.t / 1000).toFixed(1).padStart(5)}  ${String(h.fps).padStart(3)}  ${String(h.contentLen).padStart(10)}   ${h.scrollTop}/${h.scrollHeight}  ${bar}`)
+  }
+
+  // Top self frames
+  const total = (profile.endTime - profile.startTime) / 1000
+  const intMs = total / Math.max(1, profile.samples?.length ?? 1)
+  const counts = new Map()
+  for (const s of profile.samples ?? []) counts.set(s, (counts.get(s) ?? 0) + 1)
+  const rows = profile.nodes
+    .map(n => ({ id: n.id, fn: n.callFrame.functionName || '(anon)', url: n.callFrame.url || '', line: n.callFrame.lineNumber, self: counts.get(n.id) ?? 0 }))
+    .sort((a, b) => b.self - a.self)
+    .slice(0, 25)
+  console.log(`\n=== ${total.toFixed(0)}ms wall, ${profile.samples?.length ?? 0} samples (${intMs.toFixed(2)}ms each) ===`)
+  for (const r of rows) {
+    if (r.self === 0) break
+    const url = r.url.replace(/^.*\/src\//, 'src/').replace(/\?.*$/, '').slice(0, 70)
+    console.log(`  ${(r.self * intMs).toFixed(1).padStart(7)}ms  (${String(r.self).padStart(4)} samp)  ${r.fn.padEnd(45)} ${url}:${r.line}`)
+  }
+
+  await evalP(cdp, `
+    (() => {
+      for (const b of document.querySelectorAll('button')) {
+        if ((b.getAttribute('aria-label') || '').toLowerCase().includes('stop')) { b.click(); return }
+      }
+    })()
+  `)
+
+  cdp.close()
+}
+
+main().catch(e => {
+  console.error('fatal:', e.stack ?? e.message)
+  process.exit(1)
+})

apps/desktop/scripts/profile-real-stream.mjs

@@ -0,0 +1,137 @@
+// CPU-profile during a real LLM stream — confirms or refutes whether the
+// synthetic stream's hotspots (Streamdown markdown re-parse, FadeText)
+// match real-world content.
+//
+// Run *after* model is set to something fast + cheap (gpt-4o-mini etc.).
+// Sends a prompt likely to produce markdown + a numbered list.
+
+import { writeFileSync } from 'node:fs'
+
+const CDP_HTTP = 'http://127.0.0.1:9222'
+const PROMPT = process.env.PROMPT || 'Give me a numbered list of 8 useful bash one-liners. For each: a brief description, then the command in a code block. No preamble.'
+const OUT = process.env.OUT || `/tmp/real-stream-${Date.now()}.cpuprofile`
+const START_TIMEOUT = Number(process.env.START_TIMEOUT || 45000)
+const STREAM_TIMEOUT = Number(process.env.STREAM_TIMEOUT || 60000)
+
+class CDP {
+  constructor(ws) { this.ws = ws; this.id = 0; this.pending = new Map() }
+  static async open(url) {
+    const ws = new WebSocket(url)
+    await new Promise((r) => ws.addEventListener('open', r, { once: true }))
+    const cdp = new CDP(ws)
+    ws.addEventListener('message', (ev) => {
+      const m = JSON.parse(ev.data.toString())
+      if (m.id != null && cdp.pending.has(m.id)) {
+        const { resolve, reject } = cdp.pending.get(m.id)
+        cdp.pending.delete(m.id)
+        if (m.error) reject(new Error(m.error.message))
+        else resolve(m.result)
+      }
+    })
+    return cdp
+  }
+  send(method, params) {
+    const id = ++this.id
+    return new Promise((res, rej) => {
+      this.pending.set(id, { resolve: res, reject: rej })
+      this.ws.send(JSON.stringify({ id, method, params }))
+    })
+  }
+  async eval(expr) {
+    const r = await this.send('Runtime.evaluate', { expression: expr, returnByValue: true, awaitPromise: true })
+    if (r.exceptionDetails) throw new Error(r.exceptionDetails.exception?.description || 'eval')
+    return r.result.value
+  }
+  close() { this.ws.close() }
+}
+
+async function main() {
+  const list = await (await fetch(`${CDP_HTTP}/json`)).json()
+  const target = list.find((t) => t.type === 'page' && /5174/.test(t.url))
+  const cdp = await CDP.open(target.webSocketDebuggerUrl)
+
+  const baseCount = await cdp.eval('document.querySelectorAll("[data-slot=aui_assistant-message-root]").length')
+
+  // Submit prompt
+  await cdp.eval(`(() => {
+    const ed = document.querySelector('[contenteditable="true"]')
+    ed.focus()
+    document.execCommand('insertText', false, ${JSON.stringify(PROMPT)})
+    ed.dispatchEvent(new KeyboardEvent('keydown', { key: 'Enter', code: 'Enter', which: 13, keyCode: 13, bubbles: true, cancelable: true }))
+    return 'submitted'
+  })()`)
+
+  // Wait for real stream start (assistant count grows).
+  const submitT0 = Date.now()
+  let streamT = null
+  for (let i = 0; i < START_TIMEOUT / 50; i++) {
+    await new Promise((r) => setTimeout(r, 50))
+    const n = await cdp.eval('document.querySelectorAll("[data-slot=aui_assistant-message-root]").length')
+    if (n > baseCount) { streamT = Date.now(); break }
+  }
+  if (!streamT) {
+    console.error('stream never started within', START_TIMEOUT, 'ms')
+    cdp.close()
+    process.exit(2)
+  }
+  console.log('REAL stream started after', streamT - submitT0, 'ms — starting CPU profile NOW')
+
+  // Start CPU profile NOW, only during stream phase.
+  await cdp.send('Profiler.enable')
+  await cdp.send('Profiler.setSamplingInterval', { interval: 100 })
+  await cdp.send('Profiler.start')
+
+  // Wait until busy goes false + grace, or timeout.
+  const cutoff = Date.now() + STREAM_TIMEOUT
+  while (Date.now() < cutoff) {
+    await new Promise((r) => setTimeout(r, 500))
+    const busy = await cdp.eval('!!document.querySelector("[data-status=running], [data-busy=true]")')
+    if (!busy) {
+      await new Promise((r) => setTimeout(r, 500))
+      break
+    }
+  }
+
+  const { profile } = await cdp.send('Profiler.stop')
+  writeFileSync(OUT, JSON.stringify(profile))
+  console.log('wrote', OUT)
+
+  const samples = profile.samples || []
+  const timeDeltas = profile.timeDeltas || []
+  const nodes = new Map(profile.nodes.map((n) => [n.id, n]))
+  const selfTime = new Map()
+  for (let i = 0; i < samples.length; i++) {
+    const id = samples[i]
+    const dt = timeDeltas[i] ?? 0
+    selfTime.set(id, (selfTime.get(id) || 0) + dt)
+  }
+  const ranked = [...selfTime.entries()]
+    .map(([id, us]) => {
+      const n = nodes.get(id)
+      const cf = n?.callFrame || {}
+      return {
+        ms: us / 1000,
+        name: cf.functionName || '(anonymous)',
+        url: (cf.url || '').slice(-60),
+        line: cf.lineNumber
+      }
+    })
+    .filter((x) => !/\(root\)|\(idle\)|\(garbage collector\)|\(program\)/.test(x.name))
+    .sort((a, b) => b.ms - a.ms)
+    .slice(0, 25)
+
+  const finalText = await cdp.eval(`(() => {
+    const all = document.querySelectorAll('[data-slot="aui_assistant-message-root"]')
+    return all.length ? all[all.length-1].textContent.length : 0
+  })()`)
+  console.log('\nfinal assistant message length:', finalText, 'chars')
+
+  console.log('\n=== TOP 25 SELF TIME (ms) DURING REAL STREAM ===')
+  for (const r of ranked) {
+    console.log(`${r.ms.toFixed(1).padStart(7)}  ${r.name.padEnd(40)}  ${r.url}:${r.line}`)
+  }
+
+  cdp.close()
+}
+
+main().catch((e) => { console.error(e); process.exit(1) })

apps/desktop/scripts/profile-synth-stream.mjs

@@ -0,0 +1,103 @@
+// CPU-profile a synthetic stream — outputs a .cpuprofile and a top-self ranking.
+// Open the .cpuprofile in Chrome DevTools Performance panel for a flamegraph.
+
+import { writeFileSync } from 'node:fs'
+
+const CDP_HTTP = 'http://127.0.0.1:9222'
+const TOKENS = Number(process.env.TOKENS || 400)
+const INTERVAL_MS = Number(process.env.INTERVAL_MS || 8)
+const CHUNK = process.env.CHUNK || '**word** in _italic_ with `code` '
+const LABEL = process.env.LABEL || 'profile'
+const OUT = process.env.OUT || `synth-${LABEL}.cpuprofile`
+
+class CDP {
+  constructor(ws) { this.ws = ws; this.id = 0; this.pending = new Map() }
+  static async open(url) {
+    const ws = new WebSocket(url)
+    await new Promise((r) => ws.addEventListener('open', r, { once: true }))
+    const cdp = new CDP(ws)
+    ws.addEventListener('message', (ev) => {
+      const m = JSON.parse(ev.data.toString())
+      if (m.id != null && cdp.pending.has(m.id)) {
+        const { resolve, reject } = cdp.pending.get(m.id)
+        cdp.pending.delete(m.id)
+        if (m.error) reject(new Error(m.error.message))
+        else resolve(m.result)
+      }
+    })
+    return cdp
+  }
+  send(method, params) {
+    const id = ++this.id
+    return new Promise((res, rej) => {
+      this.pending.set(id, { resolve: res, reject: rej })
+      this.ws.send(JSON.stringify({ id, method, params }))
+    })
+  }
+  async eval(expr) {
+    const r = await this.send('Runtime.evaluate', { expression: expr, returnByValue: true, awaitPromise: true })
+    if (r.exceptionDetails) throw new Error(r.exceptionDetails.exception?.description || 'eval')
+    return r.result.value
+  }
+  close() { this.ws.close() }
+}
+
+async function main() {
+  const list = await (await fetch(`${CDP_HTTP}/json`)).json()
+  const target = list.find((t) => t.type === 'page' && /5174/.test(t.url))
+  const cdp = await CDP.open(target.webSocketDebuggerUrl)
+
+  if (!await cdp.eval('!!window.__PERF_DRIVE__')) {
+    console.error('no __PERF_DRIVE__')
+    cdp.close()
+    process.exit(2)
+  }
+
+  await cdp.send('Profiler.enable')
+  // High-resolution sampling: 100us
+  await cdp.send('Profiler.setSamplingInterval', { interval: 100 })
+  await cdp.send('Profiler.start')
+
+  await cdp.eval(`window.__PERF_DRIVE__.stream({ chunk: ${JSON.stringify(CHUNK)}, intervalMs: ${INTERVAL_MS}, totalTokens: ${TOKENS} })`)
+  await new Promise((r) => setTimeout(r, TOKENS * INTERVAL_MS + 1500))
+  await cdp.eval('window.__PERF_DRIVE__.reset()')
+
+  const { profile } = await cdp.send('Profiler.stop')
+  writeFileSync(OUT, JSON.stringify(profile))
+  console.log('wrote', OUT)
+
+  // Compute top self time per function.
+  const samples = profile.samples || []
+  const timeDeltas = profile.timeDeltas || []
+  const nodes = new Map(profile.nodes.map((n) => [n.id, n]))
+  const selfTime = new Map() // id -> microseconds
+  for (let i = 0; i < samples.length; i++) {
+    const id = samples[i]
+    const dt = timeDeltas[i] ?? 0
+    selfTime.set(id, (selfTime.get(id) || 0) + dt)
+  }
+  const ranked = [...selfTime.entries()]
+    .map(([id, us]) => {
+      const n = nodes.get(id)
+      const cf = n?.callFrame || {}
+      return {
+        us,
+        ms: us / 1000,
+        name: cf.functionName || '(anonymous)',
+        url: (cf.url || '').slice(-60),
+        line: cf.lineNumber
+      }
+    })
+    .filter((x) => !/\(root\)|\(idle\)|\(garbage collector\)|\(program\)/.test(x.name))
+    .sort((a, b) => b.us - a.us)
+    .slice(0, 30)
+
+  console.log('\n=== TOP 30 SELF TIME (ms) ===')
+  for (const r of ranked) {
+    console.log(`${r.ms.toFixed(1).padStart(7)}  ${r.name.padEnd(40)}  ${r.url}:${r.line}`)
+  }
+
+  cdp.close()
+}
+
+main().catch((e) => { console.error(e); process.exit(1) })

apps/desktop/scripts/profile-typing-lag.md

@@ -0,0 +1,381 @@
+# Profiling renderer typing lag
+
+Workflow for empirically measuring (and fixing) typing/submit lag in the
+desktop chat composer.
+
+## Quick boot for profiling
+
+Vite 8 + plugin-react 6 has a known issue where the React Fast Refresh
+preamble script isn't injected into `index.html`, so opening Electron at
+`http://127.0.0.1:5174` throws `$RefreshReg$ is not defined` on every TSX
+module and the React tree never mounts. Workaround: run vite with HMR off.
+
+```bash
+# Terminal A — start dev server without HMR
+cd apps/desktop
+node scripts/dev-no-hmr.mjs
+
+# Terminal B — start Electron with CDP exposed
+cd apps/desktop
+XCURSOR_SIZE=24 HERMES_DESKTOP_DEV_SERVER=http://127.0.0.1:5174 \
+  ../../node_modules/.bin/electron --remote-debugging-port=9222 .
+```
+
+Terminal C is yours to run the harnesses.
+
+## Harnesses
+
+All zero-dep — Node 24 built-in `WebSocket` + `fetch`.
+
+### Typing latency — `measure-latency.mjs`
+
+Per-keystroke `keypress → next paint` latency, p50/p90/p99/max.
+Synthesizes keystrokes via `Input.dispatchKeyEvent` so the run is
+reproducible.
+
+```bash
+node apps/desktop/scripts/measure-latency.mjs --chars=120 --cps=20
+```
+
+Anything > 16ms is a dropped frame. On a freshly-loaded session
+(`scripts/click-session.mjs 'Phaser particle'`) we currently see:
+
+| | unpatched | patched |
+|---|---|---|
+| p50 paint | 1.9 ms | 2.0 ms |
+| p90 paint | 3.3 ms | 13.7 ms |
+| p99 paint | 16.7 ms | 15.2 ms |
+| max paint | 20.5 ms | 30.4 ms |
+| >16ms drops | 2/120 | 1/120 |
+
+Roughly even on a quick session — patches don't fix typing latency
+under benign synthetic conditions because the existing baseline is
+already snappy on synthetic input. The real wins are in the leak counters
+(see below). If the user reports typing jank, capture a profile + heap
+diff during their actual usage and compare against the synthetic baseline
+to identify what condition (long thread, popover open, paste, etc.)
+makes the path slow.
+
+### Leak counters — `leak-typing.mjs`
+
+Types N chars per round, clears, force-GCs, captures
+`Performance.getMetrics` deltas. Reveals leaked event listeners, heap
+drift, document node growth, and forced-layout counts.
+
+```bash
+# After clicking into a real session (e.g. via click-session.mjs):
+node apps/desktop/scripts/leak-typing.mjs --rounds=8 --chars=200 --cps=50
+```
+
+**Real-session numbers (Phaser thread, 8 rounds × 200 chars):**
+
+| | unpatched (HEAD~2) | patched (HEAD) |
+|---|---|---|
+| jsListeners growth/round | +0 | +0 |
+| DOM nodes growth/round | +0 | +0 |
+| heap growth/round | ~0 (V8 housekeeping) | ~0 |
+| **forced layouts/char** | **7.02** | **2.35** (3× fewer) |
+
+The forced-layout count is the load-bearing number — typing into a real
+session was triggering ~7 layouts per character on the unpatched build
+(scrollHeight reads + per-px CSS var writes + FadeText scrollWidth reads
+all stacking up). After the patches it's down to ~2.35/char, which is
+Blink's natural cost for a 1px/char-growing contentEditable and can't
+be lowered further without architectural changes.
+
+The initial "+35 listeners/round leak" I called out on the first
+unpatched run turned out to be transient warm-up (popovers initializing,
+etc.); steady-state listener growth was 0 both before and after.
+
+### CPU profile + heap snapshot — `profile-typing.mjs`
+
+Records a CPU profile while typing, plus before/after heap snapshots so
+you can do a comparison diff in Chrome DevTools Memory tab.
+
+```bash
+node apps/desktop/scripts/profile-typing.mjs \
+  --chars=400 --cps=30 --out=/tmp/hermes-typing
+# → /tmp/hermes-typing.cpuprofile  (open in Chrome DevTools Performance)
+# → /tmp/hermes-typing.before.heapsnapshot
+# → /tmp/hermes-typing.after.heapsnapshot
+```
+
+Loading the cpuprofile: Chrome DevTools → Performance tab → drag the file
+in, or VS Code → open the `.cpuprofile` directly.
+
+For heap diff: Chrome DevTools → Memory → Load snapshot → load "before",
+then Comparison view → load "after". Sort by `# Delta`. Stay alert for
+detached DOM, FiberNodes (unmounted), and listener growth.
+
+## Helpers
+
+- `probe-renderer.mjs` — dump page state (URL, composer mounted?, body text)
+- `click-session.mjs <title>` — click a sidebar session by partial title match
+- `reload-renderer.mjs` — force Page.reload via CDP (no HMR available)
+- `dump-state.mjs` — richer state dump (thread message count, sticky session, etc.)
+- `probe-console.mjs` — dump recent console errors / exceptions
+
+## Findings
+
+See commit message for `apps/desktop/src/app/chat/composer/index.tsx`
+edits. Three changes:
+
+1. **Per-keystroke `scrollHeight` read removed.** The expansion useEffect
+   used to read `editorRef.current.scrollHeight` on every draft change
+   (forces synchronous layout). Replaced with a `draft.length > 60`
+   heuristic; the ResizeObserver catches anything the heuristic misses.
+
+2. **Bucketed CSS custom-property writes.** `syncComposerMetrics`
+   used to `setProperty('--composer-measured-height', height + 'px')`
+   on every observed resize, invalidating computed style for the whole
+   tree. Now writes only when the height crosses an 8 px bucket, so
+   typing in a fixed-height row produces no style invalidation at all.
+
+3. **Removed dead `$composerDraft` → `aui.composer().setText` round-trip.**
+   Nothing outside the composer subscribed to `$composerDraft` (verified
+   via grep). The two useEffects that pushed draft → store and store →
+   composer were pure overhead per keystroke. `reconcileComposerTerminalSelections`
+   was also called per keystroke; can be deferred to submit time (it's a
+   stale-pruning step, not a correctness one — `terminalContextBlocksFromDraft`
+   walks the current text directly at submit and ignores stale labels).
+
+4. **`refreshTrigger` fast-bails when no `@`/`/` in draft.** Previously
+   `textBeforeCaret()` did `range.toString()` (O(n)) on every keystroke
+   even when no trigger char was present.
+
+The biggest win is the listener leak in (3) — without it, each round of
+typing leaked ~35 event listeners until a steady state.
+
+## Submit / TTFT stall (open)
+
+User reports a perceived stall *after* Enter, before the assistant starts
+streaming. `scripts/measure-submit.mjs` measures
+`enter → composer-cleared → user-message-rendered → first-paint`. The
+script triggers a real prompt submission, so use it on a throwaway
+session. Not enabled in CI.
+
+## Streaming "5fps" investigation (May 21, 2026)
+
+User complaint: "the streaming must bring fps to like 5? lol" — felt
+hitches during assistant streaming on long threads.
+
+### Tooling added
+
+- **`src/app/chat/perf-probe.tsx`** — dev-only side-effect import (guarded by
+  `import.meta.env.MODE !== 'production'` in `main.tsx`). Attaches two
+  helpers to `window`:
+  - `__PERF_PROBE__` — React `<Profiler>` recorder. Currently inert because
+    Vite is serving the production React build (see "Vite dev-build issue"
+    below); kept for when that's fixed.
+  - `__PERF_DRIVE__` — synthetic stream driver. Pushes tokens through the
+    live `$messages` atom at a fixed cadence, so the assistant-ui runtime,
+    incremental repository, Streamdown markdown renderer, and React commit
+    pipeline all see the same workload they'd see from a real LLM stream —
+    but with no LLM call (and no credit cost).
+- **`scripts/measure-synthetic-stream.mjs`** — drives `__PERF_DRIVE__`,
+  records rAF frame intervals, `PerformanceObserver({entryTypes:['longtask']})`
+  entries, `MutationObserver` cadence on the live message, and optional
+  type-while-streaming keystroke latency.
+- **`scripts/profile-synth-stream.mjs`** — CPU profile during a synthetic
+  stream; writes a `.cpuprofile` (open in Chrome DevTools Performance panel)
+  and a top-30 self-time table.
+- **`scripts/measure-real-stream.mjs`** — same harness as the synthetic but
+  fires a real LLM prompt. Use when you have credits and want to confirm
+  the synthetic predictions hold.
+- **`scripts/profile-real-stream.mjs`** — CPU profile over the duration of
+  a real LLM stream.
+
+Helpers: `scripts/eval.mjs` (one-shot CDP eval), `scripts/reload.mjs`
+(hard reload renderer over CDP).
+
+### Findings
+
+Measured on the Cloud Shadows session (7 turns, ~11k px scrollHeight) and
+the 34 MB session `session_20260514_215353_fe0ac8.json` (110 FadeText
+instances, lots of historical tool calls).
+
+| metric | Cloud Shadows | 34 MB session |
+|---|---|---|
+| avgFps (60 tok/sec, 5s) | 60.0 | 58.6 |
+| frame p50 / p95 / p99 (ms) | 16.7 / 18.0 / 21.1 | 16.6 / 25.6 / 31.4 |
+| max frame (ms) | 31.1 | 97-127 (varies) |
+| longtasks per 5s window | 0 | 1-2, 75-127 ms |
+| type-while-stream p95 latency (ms) | 17 | — |
+
+A single real-LLM stream on Cloud Shadows (gpt-4o-mini, 39s window) saw
+12 longtasks totalling 1.26 s — same cadence the synthetic predicted
+(~1 hitch per 3.25 s, max 123 ms). So the **synthetic stream is a faithful
+proxy for the real one** and is fine for iterating on fixes without paying
+for tokens.
+
+### CPU profile during streaming (synthetic, markdown content)
+
+Top self-time costs (5 s window, 400 tokens at 125 tok/s, markdown chunks):
+
+| ms (self) | function | source |
+|---|---|---|
+| 260 | `bn$1` | `chunk-BO2N…js:20003` (micromark tokenize) |
+| 249 | `m$1` | `chunk-BO2N…js:19949` (micromark) |
+| 128 | `compile` | `chunk-BO2N…js:21884` (mdast → hast compile) |
+| 73 | FadeText body | `components/ui/fade-text.tsx` |
+| 62 | `parser` | `chunk-BO2N…js:22680` |
+| 49 | `fromThreadMessageLike` | `@assistant-ui/internal` |
+
+That `chunk-BO2N2NFS` is the vendored bundle containing `micromark`,
+`mdast-util-from-markdown`, `mdast-util-to-hast`, `rehype-raw`,
+`hast-util-sanitize`, etc. — i.e. **Streamdown's markdown pipeline,
+re-parsing the entire growing assistant message on every token append**.
+Cost scales linearly with message length.
+
+Compare plain-text (no markdown) — the `chunk-BO2N…` entries drop out
+of the top 30 entirely; total work per 5 s window halves.
+
+### Fix landed: `FadeText` memo
+
+`FadeText` is used in `tool-fallback.tsx` (110 instances on a tool-heavy
+thread). Before: each parent re-render during streaming triggered a
+`useEffect([children])` that forced a `scrollWidth` layout read — even
+when the title text was unchanged. The `useResizeObserver` already covers
+the genuine resize case, so the effect was strictly redundant.
+
+After: wrapped in `React.memo` with a custom comparator that compares
+`children` (scalar fast-path), `className`, `fadeWidth`, and `style`
+field-by-field. Verified via temporary render counter:
+**122 renders during a 2 s synthetic stream vs ~11 000 without memo**
+(110 instances × ~100 stream updates). Doesn't move the longtask needle
+on its own — Streamdown dwarfs it — but eliminates a class of forced
+layouts and removes a steady CPU floor.
+
+### Also landed: `MarkdownText` plugins memo + upstream flush floor
+
+Two smaller follow-ups in the same investigation:
+
+1. **`MarkdownText` `plugins` object useMemo'd.** The inline
+   `plugins={{ math: mathPlugin, ...(isStreaming ? {} : { code }) }}`
+   was constructing a new object on every render, which churns
+   `<Streamdown>`'s outer memo and forces its internal `rehypePlugins` /
+   `remarkPlugins` arrays to rebuild. CPU profile after the change shows
+   `parser` self-time dropping out of the top 10, `compile` cut roughly
+   in half, and `bn$1` / `m$1` (micromark internals) dropping off the
+   top entries.
+
+2. **`use-message-stream.scheduleDeltaFlush` got a real minimum floor.**
+   Previously the rAF-only path effectively meant "at most one flush per
+   frame," but at typical LLM token rates of 30-80 tok/sec each token
+   arrives slower than rAF cadence and gets its own React commit. With
+   `STREAM_DELTA_FLUSH_MS = 33` (two frames) and a `lastFlushAt`-tracked
+   floor, slower streams now coalesce ~2 tokens per commit, halving
+   markdown re-parses. React's auto-batching already covers part of this
+   probabilistically; the floor makes the batching deterministic so the
+   max-longtask number tightens up.
+
+A/B on the 34 MB session, 300 tokens at 50 tok/sec, markdown chunks
+(3 trials each):
+
+| | avgFps | p99 frame | LTs/5s | max LT | mutations |
+|---|---|---|---|---|---|
+| no throttle  | 54.0 | 38 ms | 2.0 | 145 ms | varies (2-112) |
+| 33 ms throttle | 54.3 | 41 ms | 1.7 | 110 ms | ~135 |
+
+Modest. `inter-mutation` p50 tightens from 22-28 ms to a clean 33 ms,
+which is what you'd expect from a deterministic floor.
+
+### Also landed: `useDeferredValue` at the streamdown-text boundary
+
+The longtask CPU was unavoidable inside the block-memo pattern — the live
+tail re-parses every commit, scales linearly with current length, and
+nothing about Streamdown's architecture changes that without forking. The
+fix is to stop having that work *block* the main thread.
+
+`<DeferStreamingText>` in `markdown-text.tsx` is a 12-line wrapper that
+reads the message-part state via `useMessagePartText`, runs it through
+`useDeferredValue`, and re-publishes via assistant-ui's
+`<TextMessagePartProvider>`. The inner `StreamdownTextPrimitive` reads the
+deferred value through the normal `useMessagePartText` hook — no fork,
+no internal-path imports, fully on the assistant-ui public API.
+
+What React's concurrent scheduler now does:
+
+- When a new token arrives mid-render, the in-flight deferred render
+  is abandoned and a fresh one starts with the latest text.
+- When the main thread has urgent work (typing, scroll, layout), the
+  Streamdown render gets deprioritized — input stays responsive even
+  while a 100 ms parse is queued.
+
+Streamdown already uses `useTransition` internally for its block-array
+setState; `useDeferredValue` here just lifts the deferral all the way up
+to the consumer text boundary, so the whole pipeline — preprocess,
+block split, repair, parse, render — runs at low priority during streaming.
+This is the industry-standard approach (see
+[Streamdown architecture analysis](https://tigerabrodi.blog/how-to-build-a-performant-ai-markdown-renderer)
+and Chrome's [LLM-response render best practices](https://developer.chrome.google.cn/docs/ai/render-llm-responses)).
+
+A/B on the 34 MB session, 300 tokens at 50 tok/sec, markdown chunks
+(four trials each, prod-throttle (33 ms) on for both):
+
+| | avgFps | p99 frame | LTs / 5 s | max LT | typing p95 |
+|---|---|---|---|---|---|
+| pre-defer | 54.3 | 41 ms | 1.7 | 110 ms | ~17 ms |
+| **post-defer** | **58.5** | **31 ms** | 2.0 | 117 ms | 14-18 ms |
+
+Longtask count and max LT are unchanged — `useDeferredValue` doesn't
+reduce CPU, only its priority. The avgFps lift and p99 frame drop are
+the proof that the existing CPU is no longer blocking 60 fps cadence:
+when React can defer the parse, frames stay clean. One particularly
+clean run logged **MUTATIONS=0** — React skipped every intermediate
+text state and only committed the final one, the textbook
+useDeferredValue behaviour.
+
+### Not fixed: Streamdown markdown re-parse cost (the elephant)
+
+Total CPU spent in micromark/mdast/hast pipeline per 5 s window is still
+the same ~700 ms. With `useDeferredValue` that work no longer blocks
+input, but if you watch a CPU profile you'll see the same hot functions
+(`Tn$1`, `bn$1`, `m$1`, `parser`, `compile`).
+
+The path to actually *reduce* that cost (not just defer it) is to
+replace the parser with a state machine like
+[Flowdown](https://github.com/Atomics-hub/flowdown) — process each
+character exactly once, emit DOM ops directly, no re-parse of the prefix
+on every token. Claimed ~2,000× over `marked`. Trades: not a
+`react-markdown`-compatible API, no rehype security pipeline, would
+require replacing Streamdown wholesale. Worth investigating only if
+even the deferred work shows up in user-perceptible ways (e.g.
+trackpad-scrolling a stream-in-progress stutters).
+
+The synthetic harness now mirrors the real upstream pipeline via the
+`flushMinMs` option in `__PERF_DRIVE__.stream({ flushMinMs: 33 })`, so
+future Streamdown / Flowdown experiments can A/B without LLM credit cost.
+The synthetic numbers tracked the one real-LLM run we caught within
+noise, so it's a reliable proxy.
+
+Possible approaches (none implemented here):
+
+1. **Coalesce/throttle Streamdown updates** — render at most every 32 ms
+   instead of every set-state. Reduces parses but doesn't reduce
+   per-parse cost; trades latency for smoothness.
+2. **Memoize per-prefix** — diff the new text against the prior parsed
+   version; only re-parse the changed suffix.
+3. **Render in stable segments** — close-form historical paragraphs as
+   immutable React nodes; only the live tail goes through markdown each
+   token. Probably the highest-impact change but requires forking or
+   patching `@assistant-ui/react-streamdown`.
+4. **Move parsing to a Web Worker** — main thread no longer blocks on
+   markdown. Largest surgery; requires double-buffered hast.
+
+### Vite dev-build issue (separate)
+
+`http://127.0.0.1:5174/node_modules/.vite/deps/react.js` resolves to
+`react/cjs/react.production.js`, and `react-dom_client.js` →
+`react-dom-client.production.js`. As a result:
+
+- `<React.Profiler>` `onRender` is never called (production build is a
+  no-op).
+- `import.meta.env.DEV` is `false`, `PROD` is `true` even under `vite dev`
+  (hence `MODE !== 'production'` as the workaround in `main.tsx`).
+- All the React 19 dev-only warnings/devtools backend hooks are absent.
+
+Root cause likely sits in `vite.config.ts` aliasing + dedupe + Vite 8's
+new `optimizeDeps` defaults. Worth a separate fix pass — when it's
+resolved, the `<PerfProbe>` blocks in `perf-probe.tsx` become useful
+(per-id commit timings) instead of inert.

apps/desktop/scripts/profile-typing.mjs

@@ -0,0 +1,260 @@
+#!/usr/bin/env node
+// Profile typing lag in the Electron renderer by:
+//  1. Connecting to a running renderer via CDP (--remote-debugging-port=9222)
+//  2. Focusing the composer contentEditable
+//  3. Starting CPU profile + heap snapshot
+//  4. Synthesizing keystrokes via Input.dispatchKeyEvent (so the run is
+//     reproducible, no human-typing variance)
+//  5. Stopping the profile + capturing a second heap snapshot
+//  6. Saving .cpuprofile + .heapsnapshot
+//
+// Usage:
+//   node apps/desktop/scripts/profile-typing.mjs
+//     [--port=9222] [--out=/tmp/hermes-typing]
+//     [--chars=400]              # how many characters to type
+//     [--cps=30]                 # keystrokes per second
+//     [--text="..."]             # override generated text
+//     [--no-heap]                # skip heap snapshots
+//     [--seconds=N]              # idle-record for N seconds instead of typing
+//
+// Zero deps — uses Node 24's global WebSocket + fetch.
+
+import { writeFileSync } from 'node:fs'
+
+const args = Object.fromEntries(
+  process.argv.slice(2).flatMap(s => {
+    const m = s.match(/^--([^=]+)(?:=(.*))?$/)
+    return m ? [[m[1], m[2] ?? true]] : []
+  })
+)
+
+const PORT = Number(args.port ?? 9222)
+const OUT = String(args.out ?? `/tmp/hermes-typing-${Date.now()}`)
+const CHARS = Number(args.chars ?? 400)
+const CPS = Number(args.cps ?? 30)
+const HEAP = args['no-heap'] ? false : true
+const IDLE_SECONDS = args.seconds ? Number(args.seconds) : null
+const CUSTOM_TEXT = args.text === undefined || args.text === true ? null : String(args.text)
+
+const log = (...m) => console.log('[profile]', ...m)
+const banner = m => console.log(`\n========== ${m} ==========`)
+
+async function pickRenderer() {
+  const list = await (await fetch(`http://127.0.0.1:${PORT}/json/list`)).json()
+  const pages = list.filter(t => t.type === 'page' && t.url.startsWith('http'))
+  if (!pages.length) {
+    console.error('No renderer page. Targets:')
+    list.forEach(t => console.error(' ', t.type, t.url))
+    process.exit(2)
+  }
+  return pages[0]
+}
+
+function connect(url) {
+  return new Promise((resolve, reject) => {
+    const ws = new WebSocket(url)
+    let id = 0
+    const pending = new Map()
+    const events = new Map()
+    ws.addEventListener('open', () =>
+      resolve({
+        send(method, params = {}) {
+          const myId = ++id
+          ws.send(JSON.stringify({ id: myId, method, params }))
+          return new Promise((res, rej) => pending.set(myId, { res, rej }))
+        },
+        on(method, h) {
+          if (!events.has(method)) events.set(method, [])
+          events.get(method).push(h)
+        },
+        close: () => ws.close()
+      })
+    )
+    ws.addEventListener('error', reject)
+    ws.addEventListener('message', ev => {
+      const txt = typeof ev.data === 'string' ? ev.data : ev.data.toString('utf8')
+      const m = JSON.parse(txt)
+      if (m.id != null) {
+        const p = pending.get(m.id)
+        if (!p) return
+        pending.delete(m.id)
+        m.error ? p.rej(new Error(m.error.message)) : p.res(m.result)
+      } else if (m.method) {
+        ;(events.get(m.method) ?? []).forEach(h => h(m.params))
+      }
+    })
+  })
+}
+
+async function captureHeap(cdp, path) {
+  log(`heap snapshot → ${path}`)
+  const chunks = []
+  cdp.on('HeapProfiler.addHeapSnapshotChunk', ({ chunk }) => chunks.push(chunk))
+  await cdp.send('HeapProfiler.enable')
+  await cdp.send('HeapProfiler.takeHeapSnapshot', { reportProgress: false, captureNumericValue: true })
+  writeFileSync(path, chunks.join(''))
+  log(`  ${(Buffer.byteLength(chunks.join(''), 'utf8') / 1024 / 1024).toFixed(1)} MB`)
+}
+
+async function focusComposer(cdp) {
+  // Focus the rich-input contentEditable. RICH_INPUT_SLOT is the data-slot
+  // value used by the composer's editable div. If focus fails (no composer
+  // mounted yet — disabled state, etc.) the script logs and continues; the
+  // profile will still show idle behavior.
+  const result = await cdp.send('Runtime.evaluate', {
+    expression: `
+      (() => {
+        const el = document.querySelector('[data-slot="composer-rich-input"]')
+        if (!el) return { ok: false, reason: 'composer-rich-input not found' }
+        el.focus()
+        // place caret at end
+        const range = document.createRange()
+        range.selectNodeContents(el)
+        range.collapse(false)
+        const sel = window.getSelection()
+        sel.removeAllRanges()
+        sel.addRange(range)
+        return { ok: true, text: el.innerText.length }
+      })()
+    `,
+    returnByValue: true
+  })
+  if (!result.result.value?.ok) {
+    log(`focus failed: ${result.result.value?.reason ?? 'unknown'}`)
+    return false
+  }
+  log(`composer focused (existing text length: ${result.result.value.text})`)
+  return true
+}
+
+function genText(n) {
+  const lorem =
+    'the quick brown fox jumps over the lazy dog while the agent thinks really hard about why typing into this composer feels like wading through molasses on a hot afternoon '
+  let s = ''
+  while (s.length < n) s += lorem
+  return s.slice(0, n)
+}
+
+async function dispatchChar(cdp, ch) {
+  // For printable chars, char + keypress is enough — Electron treats it as text input
+  // and the contentEditable input event fires. For Enter / Space we could add
+  // specials; this run is one long line.
+  await cdp.send('Input.dispatchKeyEvent', {
+    type: 'char',
+    text: ch,
+    unmodifiedText: ch
+  })
+}
+
+async function typeText(cdp, text, cps) {
+  const intervalMs = Math.max(1, Math.round(1000 / cps))
+  const start = Date.now()
+  for (let i = 0; i < text.length; i++) {
+    await dispatchChar(cdp, text[i])
+    // Pace evenly; account for dispatch latency so we don't drift much.
+    const expected = start + (i + 1) * intervalMs
+    const wait = expected - Date.now()
+    if (wait > 0) await new Promise(r => setTimeout(r, wait))
+  }
+}
+
+async function main() {
+  log(`CDP port ${PORT}, out ${OUT}`)
+  const target = await pickRenderer()
+  log(`target ${target.url}`)
+  const cdp = await connect(target.webSocketDebuggerUrl)
+  await cdp.send('Runtime.enable')
+  await cdp.send('Page.enable')
+  await cdp.send('Profiler.enable')
+
+  // Pre-GC so the cpu profile + heap delta are clean.
+  try {
+    await cdp.send('HeapProfiler.collectGarbage')
+  } catch (e) {
+    log('GC skipped:', e.message)
+  }
+
+  if (HEAP) await captureHeap(cdp, `${OUT}.before.heapsnapshot`)
+
+  // 1ms sampling — fine enough for per-frame React work.
+  await cdp.send('Profiler.setSamplingInterval', { interval: 1000 })
+
+  let typedText = ''
+  if (!IDLE_SECONDS) {
+    const focused = await focusComposer(cdp)
+    if (!focused) {
+      log('aborting — composer not focusable. Make sure the app is past the boot screen.')
+      cdp.close()
+      process.exit(3)
+    }
+    typedText = CUSTOM_TEXT ?? genText(CHARS)
+  }
+
+  await cdp.send('Profiler.start')
+
+  if (IDLE_SECONDS) {
+    banner(`IDLE recording for ${IDLE_SECONDS}s — DO NOT TOUCH`)
+    await new Promise(r => setTimeout(r, IDLE_SECONDS * 1000))
+  } else {
+    banner(`TYPING ${typedText.length} chars @ ${CPS} cps (≈${(typedText.length / CPS).toFixed(1)}s)`)
+    const t0 = Date.now()
+    await typeText(cdp, typedText, CPS)
+    log(`typing wall time: ${((Date.now() - t0) / 1000).toFixed(2)}s`)
+    // Settle frame for trailing React work.
+    await new Promise(r => setTimeout(r, 500))
+  }
+
+  banner('STOP — saving profile')
+  const { profile } = await cdp.send('Profiler.stop')
+  writeFileSync(`${OUT}.cpuprofile`, JSON.stringify(profile))
+  log(`cpu profile → ${OUT}.cpuprofile (${(JSON.stringify(profile).length / 1024 / 1024).toFixed(1)} MB)`)
+
+  if (HEAP) {
+    try {
+      await cdp.send('HeapProfiler.collectGarbage')
+    } catch {}
+    await captureHeap(cdp, `${OUT}.after.heapsnapshot`)
+  }
+
+  // Quick triage: top-self-time frames from the profile.
+  const top = summarizeProfile(profile)
+  banner('TOP SELF-TIME FRAMES')
+  for (const row of top.slice(0, 20)) {
+    console.log(
+      `  ${row.selfMs.toFixed(1).padStart(7)}ms  ${row.functionName || '(anonymous)'}` +
+        `  ${row.url ? '· ' + row.url.replace(/^.*\/src\//, 'src/').slice(0, 80) : ''}`
+    )
+  }
+  console.log()
+  log(`total samples: ${top.totalSamples}, total time: ${(top.totalMs / 1000).toFixed(2)}s`)
+
+  cdp.close()
+}
+
+function summarizeProfile(profile) {
+  // Cumulative samples = how many sampling ticks landed on each node.
+  // selfMs = own time only, using sampling interval.
+  const intervalMs = (profile.endTime - profile.startTime) / 1000 / Math.max(1, profile.samples?.length ?? 1)
+  const counts = new Map()
+  for (const s of profile.samples ?? []) counts.set(s, (counts.get(s) ?? 0) + 1)
+  const rows = profile.nodes.map(n => {
+    const self = counts.get(n.id) ?? 0
+    return {
+      id: n.id,
+      functionName: n.callFrame.functionName,
+      url: n.callFrame.url,
+      lineNumber: n.callFrame.lineNumber,
+      selfSamples: self,
+      selfMs: self * intervalMs
+    }
+  })
+  rows.sort((a, b) => b.selfSamples - a.selfSamples)
+  rows.totalSamples = (profile.samples ?? []).length
+  rows.totalMs = ((profile.endTime - profile.startTime) / 1000)
+  return rows
+}
+
+main().catch(e => {
+  console.error('[profile] fatal:', e.stack ?? e.message)
+  process.exit(1)
+})

apps/desktop/scripts/reload-renderer.mjs

@@ -0,0 +1,25 @@
+// Reload the renderer via CDP so it picks up the latest from Vite.
+const list = await (await fetch('http://127.0.0.1:9222/json/list')).json()
+const tgt = list.find(t => t.type === 'page' && t.url.startsWith('http'))
+const ws = new WebSocket(tgt.webSocketDebuggerUrl)
+let id = 0
+const pending = new Map()
+ws.addEventListener('message', ev => {
+  const m = JSON.parse(ev.data)
+  if (m.id != null && pending.has(m.id)) {
+    pending.get(m.id)(m)
+    pending.delete(m.id)
+  }
+})
+await new Promise(r => ws.addEventListener('open', r))
+const send = (method, params = {}) =>
+  new Promise(r => {
+    const i = ++id
+    pending.set(i, r)
+    ws.send(JSON.stringify({ id: i, method, params }))
+  })
+await send('Page.enable')
+await send('Page.reload', { ignoreCache: true })
+console.log('reload requested')
+await new Promise(r => setTimeout(r, 200))
+ws.close()

apps/desktop/scripts/reload.mjs

@@ -0,0 +1,36 @@
+// Hard reload the Electron renderer over CDP. Vite-no-HMR mode means edits
+// don't auto-apply — call this after editing source.
+const targets = await (await fetch('http://127.0.0.1:9222/json')).json()
+const t = targets.find((t) => t.url.includes('5174'))
+if (!t) {
+  console.error('renderer not found')
+  process.exit(1)
+}
+const ws = new WebSocket(t.webSocketDebuggerUrl)
+let id = 0
+const pending = new Map()
+ws.addEventListener('message', (ev) => {
+  const m = JSON.parse(ev.data)
+  if (pending.has(m.id)) {
+    pending.get(m.id)(m)
+    pending.delete(m.id)
+  }
+})
+await new Promise((r) => ws.addEventListener('open', r))
+const send = (method, params = {}) =>
+  new Promise((res) => {
+    const i = ++id
+    pending.set(i, res)
+    ws.send(JSON.stringify({ id: i, method, params }))
+  })
+
+await send('Page.reload', { ignoreCache: true })
+console.log('reload sent')
+// Wait for new doc.
+await new Promise((r) => setTimeout(r, 2500))
+const r = await send('Runtime.evaluate', {
+  expression: 'JSON.stringify({ hasProbe: !!window.__PERF_PROBE__, composer: !!document.querySelector("[contenteditable=true]"), url: location.hash })',
+  returnByValue: true,
+})
+console.log(r.result.result.value)
+ws.close()

apps/desktop/scripts/set-exe-identity.cjs

@@ -0,0 +1,94 @@
+#!/usr/bin/env node
+// set-exe-identity.cjs — stamp the Hermes icon + version metadata onto the
+// built Hermes.exe using rcedit, completely decoupled from electron-builder's
+// signing path.
+//
+// WHY THIS EXISTS
+// ---------------
+// apps/desktop/package.json sets build.win.signAndEditExecutable=false. That
+// flag is load-bearing: turning electron-builder's own exe-editing ON also
+// re-enables its signtool step, which fetches winCodeSign-2.6.0.7z, whose
+// macOS symlinks crash 7-Zip on non-admin Windows (no Developer Mode = no
+// SeCreateSymbolicLinkPrivilege). That is an unfixable dead end — we do NOT
+// try to extract winCodeSign.
+//
+// The cost of disabling signAndEditExecutable is that electron-builder also
+// skips rcedit, so the unpacked Hermes.exe keeps the stock Electron icon and
+// "Electron" taskbar name. This script restores the icon + identity by calling
+// rcedit DIRECTLY. rcedit is a pure PE resource editor: no signing, no certs,
+// no winCodeSign, no symlinks.
+//
+// HOW IT RUNS
+// -----------
+// Primarily as an electron-builder `afterPack` hook (scripts/after-pack.cjs),
+// so EVERY packed build — first install, `hermes desktop`, the installer's
+// --update rebuild, or a dev's manual `npm run pack` — gets a branded exe from
+// one place. Previously this stamp lived only in install.ps1, so the update
+// path (which rebuilds via `hermes desktop --build-only`, never install.ps1)
+// shipped a stock "Electron" exe. Keeping it in afterPack closes that gap.
+//
+// Also runnable standalone for ad-hoc re-stamping:
+//   node scripts/set-exe-identity.cjs <path-to-Hermes.exe>
+//
+// Exits 0 on success, non-zero on failure when run as a CLI. As a hook,
+// stampExeIdentity() resolves on success and rejects on failure; the caller
+// (after-pack.cjs) swallows the rejection so a stamp failure never fails an
+// otherwise-good build (worst case: stock icon, not a broken app).
+
+const path = require('node:path')
+const fs = require('node:fs')
+
+// Stamp the Hermes icon + identity onto `exe`. Resolves on success, throws on
+// failure. `desktopRoot` defaults to this script's package root so the icon and
+// the rcedit dependency resolve regardless of cwd.
+async function stampExeIdentity(exe, desktopRoot = path.resolve(__dirname, '..')) {
+  if (!exe || !fs.existsSync(exe)) {
+    throw new Error(`target exe not found: ${exe}`)
+  }
+
+  // Icon lives at apps/desktop/assets/icon.ico
+  const icon = path.join(desktopRoot, 'assets', 'icon.ico')
+  if (!fs.existsSync(icon)) {
+    throw new Error(`icon not found: ${icon}`)
+  }
+
+  // rcedit is a direct devDependency of apps/desktop, so it resolves whether
+  // we're run from the desktop dir or the repo root (workspace hoist).
+  // rcedit@5 exports a NAMED `rcedit` function (CommonJS: { rcedit }), not a
+  // default export.
+  const mod = require('rcedit')
+  const rcedit = typeof mod === 'function' ? mod : mod.rcedit
+  if (typeof rcedit !== 'function') {
+    throw new Error(`unexpected rcedit export shape: ${typeof mod} keys=${Object.keys(mod)}`)
+  }
+
+  console.log(`[set-exe-identity] stamping ${exe}`)
+  console.log(`[set-exe-identity] icon: ${icon}`)
+
+  await rcedit(exe, {
+    icon,
+    'version-string': {
+      ProductName: 'Hermes',
+      FileDescription: 'Hermes',
+      CompanyName: 'Nous Research',
+      LegalCopyright: 'Copyright (c) 2026 Nous Research'
+    }
+  })
+
+  console.log('[set-exe-identity] done — Hermes icon + identity stamped')
+}
+
+module.exports = { stampExeIdentity }
+
+// CLI entry point: `node scripts/set-exe-identity.cjs <exe>`.
+if (require.main === module) {
+  const exe = process.argv[2]
+  if (!exe) {
+    console.error('[set-exe-identity] usage: set-exe-identity.cjs <path-to-exe>')
+    process.exit(2)
+  }
+  stampExeIdentity(exe).catch(err => {
+    console.error(`[set-exe-identity] ${err.message}`)
+    process.exit(1)
+  })
+}

apps/desktop/scripts/stage-native-deps.cjs

@@ -0,0 +1,159 @@
+'use strict'
+
+/**
+ * Stage native node-modules dependencies for electron-builder packaging.
+ *
+ * Workspace dedup hoists `node-pty` into the root `node_modules/`, which
+ * electron-builder's default file collector (when `files:` is explicitly set
+ * in package.json) cannot reach.  The result: packaged builds ship with no
+ * .node binaries and PTY initialization fails at runtime ("PTY support is
+ * unavailable").
+ *
+ * Rather than restructure the workspace dedup (would require nohoist /
+ * package.json shenanigans and risk breaking dev) or balloon the package
+ * with the whole node_modules tree, we copy ONLY the runtime-essential
+ * files of the native dep into apps/desktop/build/native-deps/ and ship
+ * THAT subtree via extraResources.  main.cjs falls back to require()-ing
+ * from process.resourcesPath when the hoisted-root require fails.
+ *
+ * Runs as part of `npm run build`. Idempotent -- always re-stages on each
+ * build to pick up native binary updates.
+ *
+ * Layout note: upstream node-pty (microsoft/node-pty 1.x) is N-API based
+ * and ships its prebuilts under `prebuilds/<platform>-<arch>/` instead of
+ * `build/Release/`.  Its runtime resolver (lib/utils.js) checks
+ * build/Release first and falls through to the per-arch prebuilds dir, so
+ * shipping only the latter is sufficient for packaged runs.  Per-arch
+ * staging keeps the resource bundle lean -- we only need the target
+ * arch's prebuilt, not all of them.
+ */
+
+const fs = require('node:fs')
+const path = require('node:path')
+
+const APP_ROOT = path.resolve(__dirname, '..')
+const REPO_ROOT = path.resolve(APP_ROOT, '..', '..')
+const STAGE_ROOT = path.join(APP_ROOT, 'build', 'native-deps')
+
+// The target arch may be overridden by electron-builder via npm_config_arch
+// (e.g. `npm run dist -- --arm64`); fall back to the build host's arch.
+const TARGET_ARCH = process.env.npm_config_arch || process.arch
+const TARGET_PLATFORM = process.platform
+
+// Modules to stage. The "from" path is the hoisted location in the workspace
+// root; "to" is the layout we want inside build/native-deps/.  The "include"
+// globs (relative to "from") select the runtime-essential files.  Anything
+// outside the include list is left behind (source, deps/, scripts/, etc.).
+const NATIVE_DEPS = [
+  {
+    from: path.join(REPO_ROOT, 'node_modules', 'node-pty'),
+    to: path.join(STAGE_ROOT, 'node-pty'),
+    include: [
+      'package.json',
+      'lib/*.js',
+      'lib/**/*.js',
+      'build/Release/*.node',
+      // Per-arch runtime payload. Explicit file types so we don't ship the
+      // ~25 MB of .pdb debug symbols that prebuild-install bundles for
+      // Windows crash analysis -- not used at runtime, would just bloat
+      // the installer.
+      `prebuilds/${TARGET_PLATFORM}-${TARGET_ARCH}/*.node`,
+      `prebuilds/${TARGET_PLATFORM}-${TARGET_ARCH}/*.dll`,
+      `prebuilds/${TARGET_PLATFORM}-${TARGET_ARCH}/*.exe`,
+      `prebuilds/${TARGET_PLATFORM}-${TARGET_ARCH}/spawn-helper`,
+      `prebuilds/${TARGET_PLATFORM}-${TARGET_ARCH}/conpty/*`
+    ]
+  }
+]
+
+function rmrf(target) {
+  fs.rmSync(target, { recursive: true, force: true })
+}
+
+function ensureDir(target) {
+  fs.mkdirSync(target, { recursive: true })
+}
+
+function walk(root) {
+  const results = []
+  const stack = [root]
+  while (stack.length) {
+    const current = stack.pop()
+    let entries
+    try {
+      entries = fs.readdirSync(current, { withFileTypes: true })
+    } catch {
+      continue
+    }
+    for (const entry of entries) {
+      const full = path.join(current, entry.name)
+      if (entry.isDirectory()) {
+        stack.push(full)
+      } else if (entry.isFile()) {
+        results.push(full)
+      }
+    }
+  }
+  return results
+}
+
+// Match a relative path against simple ** and * glob patterns. Implementation
+// is intentionally tiny -- the include lists are small and don't need full
+// minimatch support.
+function matchGlob(rel, pattern) {
+  const r = rel.replace(/\\/g, '/')
+  const re = new RegExp(
+    '^' +
+      pattern
+        .replace(/\\/g, '/')
+        .replace(/[.+^${}()|[\]\\]/g, '\\$&')
+        .replace(/\*\*/g, '__DOUBLE_STAR__')
+        .replace(/\*/g, '[^/]*')
+        .replace(/__DOUBLE_STAR__/g, '.*') +
+      '$'
+  )
+  return re.test(r)
+}
+
+function stageOne(spec) {
+  if (!fs.existsSync(spec.from)) {
+    throw new Error(
+      `stage-native-deps: source missing at ${spec.from}.  Run \`npm install\` ` +
+        `at the workspace root first.`
+    )
+  }
+  rmrf(spec.to)
+  ensureDir(spec.to)
+
+  const files = walk(spec.from)
+  let copied = 0
+  for (const abs of files) {
+    const rel = path.relative(spec.from, abs)
+    const included = spec.include.some(g => matchGlob(rel, g))
+    if (!included) continue
+    const dest = path.join(spec.to, rel)
+    ensureDir(path.dirname(dest))
+    fs.copyFileSync(abs, dest)
+    // node-pty's darwin spawn-helper and the Windows helper binaries
+    // (OpenConsole.exe, winpty-agent.exe) are invoked via posix_spawn /
+    // CreateProcess at runtime, so they must remain executable in the
+    // staged tree.  fs.copyFileSync preserves source mode on POSIX, but we
+    // re-assert +x defensively for the darwin spawn-helper (no extension
+    // means a stripped mode would be silently broken at runtime).
+    if (path.basename(rel) === 'spawn-helper' && process.platform !== 'win32') {
+      try { fs.chmodSync(dest, 0o755) } catch { /* best-effort */ }
+    }
+    copied += 1
+  }
+  console.log(`[stage-native-deps] ${path.relative(APP_ROOT, spec.to)}: ${copied} files`)
+}
+
+function main() {
+  rmrf(STAGE_ROOT)
+  ensureDir(STAGE_ROOT)
+  for (const spec of NATIVE_DEPS) {
+    stageOne(spec)
+  }
+}
+
+main()

apps/desktop/scripts/test-desktop.mjs

@@ -0,0 +1,425 @@
+import fs from 'node:fs'
+import os from 'node:os'
+import path from 'node:path'
+import { spawn, spawnSync } from 'node:child_process'
+import { fileURLToPath } from 'node:url'
+import { listPackage } from '@electron/asar'
+
+const DESKTOP_ROOT = path.resolve(path.dirname(fileURLToPath(import.meta.url)), '..')
+const PACKAGE_JSON = JSON.parse(fs.readFileSync(path.join(DESKTOP_ROOT, 'package.json'), 'utf8'))
+const MODE = process.argv[2] || 'help'
+const ARCH = process.arch === 'arm64' ? 'arm64' : 'x64'
+const RELEASE_ROOT = path.join(DESKTOP_ROOT, 'release')
+const PLATFORM = process.platform
+
+// Platform-specific packaged-app layout. The thin installer ships an Electron
+// app shell plus extraResources (install-stamp.json + native-deps/) -- it
+// no longer bundles the Hermes Agent Python payload (that's fetched at first
+// launch via install.ps1 / install.sh, per the Phase 1 thin-installer flow).
+const APP = (() => {
+  if (PLATFORM === 'darwin') {
+    const appPath = path.join(RELEASE_ROOT, `mac-${ARCH}`, 'Hermes.app')
+    return {
+      appPath,
+      binary: path.join(appPath, 'Contents', 'MacOS', 'Hermes'),
+      resourcesPath: path.join(appPath, 'Contents', 'Resources'),
+      asarPath: path.join(appPath, 'Contents', 'Resources', 'app.asar'),
+      unpackedDistIndex: path.join(appPath, 'Contents', 'Resources', 'app.asar.unpacked', 'dist', 'index.html')
+    }
+  }
+  if (PLATFORM === 'win32') {
+    const unpacked = path.join(RELEASE_ROOT, 'win-unpacked')
+    return {
+      appPath: unpacked,
+      binary: path.join(unpacked, 'Hermes.exe'),
+      resourcesPath: path.join(unpacked, 'resources'),
+      asarPath: path.join(unpacked, 'resources', 'app.asar'),
+      unpackedDistIndex: path.join(unpacked, 'resources', 'app.asar.unpacked', 'dist', 'index.html')
+    }
+  }
+  // linux unpacked layout matches windows but with different binary name
+  const unpacked = path.join(RELEASE_ROOT, 'linux-unpacked')
+  return {
+    appPath: unpacked,
+    binary: path.join(unpacked, 'hermes'),
+    resourcesPath: path.join(unpacked, 'resources'),
+    asarPath: path.join(unpacked, 'resources', 'app.asar'),
+    unpackedDistIndex: path.join(unpacked, 'resources', 'app.asar.unpacked', 'dist', 'index.html')
+  }
+})()
+
+// Default HERMES_HOME for non-sandboxed runs -- matches main.cjs's
+// resolveHermesHome(). On Windows it's %LOCALAPPDATA%\hermes; elsewhere
+// it's ~/.hermes. The fresh-install sandbox launchFresh() sets its own
+// HERMES_HOME and never touches this.
+const DEFAULT_HERMES_HOME = (() => {
+  if (PLATFORM === 'win32' && process.env.LOCALAPPDATA) {
+    return path.join(process.env.LOCALAPPDATA, 'hermes')
+  }
+  return path.join(os.homedir(), '.hermes')
+})()
+const VENV_ROOT = path.join(DEFAULT_HERMES_HOME, 'hermes-agent', 'venv')
+const FRESH_SANDBOX_ROOT = path.join(os.tmpdir(), 'hermes-desktop-fresh-install')
+
+function die(message) {
+  console.error(`\n${message}`)
+  process.exit(1)
+}
+
+function run(command, args, options = {}) {
+  const result = spawnSync(command, args, {
+    cwd: options.cwd || DESKTOP_ROOT,
+    env: options.env || process.env,
+    shell: Boolean(options.shell) || PLATFORM === 'win32',
+    stdio: 'inherit'
+  })
+
+  if (result.status !== 0) {
+    die(`${command} ${args.join(' ')} failed`)
+  }
+}
+
+function exists(target) {
+  return fs.existsSync(target)
+}
+
+// Match nodepty native binding location to what main.cjs's resolver fallback
+// expects (apps/desktop/electron/main.cjs, packaged-build branch).  Upstream
+// node-pty 1.x is N-API based and ships per-arch prebuilts under
+// prebuilds/<platform>-<arch>/ instead of build/Release/.  We check the
+// per-arch dir since that's what stage-native-deps actually copies.
+function expectedNativeDepPaths() {
+  const root = path.join(APP.resourcesPath, 'native-deps', 'node-pty')
+  const prebuildsDir = path.join(root, 'prebuilds', `${PLATFORM}-${ARCH}`)
+  return {
+    packageJson: path.join(root, 'package.json'),
+    prebuildsDir,
+    libIndex: path.join(root, 'lib', 'index.js')
+  }
+}
+
+function ensurePlatformBuilds() {
+  if (PLATFORM === 'darwin') return
+  if (PLATFORM === 'win32') return
+  die(
+    `Desktop bundle validation is only wired for darwin / win32 today; platform=${PLATFORM} ` +
+      `is not yet supported. The thin-installer story for Linux ships in Phase 2 alongside ` +
+      `install.sh's stage protocol.`
+  )
+}
+
+function ensurePackagedApp() {
+  if (process.env.HERMES_DESKTOP_SKIP_BUILD === '1' && exists(APP.binary)) {
+    return
+  }
+
+  run('npm', ['run', 'pack'])
+}
+
+function resolveDmgPath() {
+  if (!exists(RELEASE_ROOT)) {
+    return path.join(RELEASE_ROOT, `Hermes-${PACKAGE_JSON.version}-${ARCH}.dmg`)
+  }
+
+  const prefix = `Hermes-${PACKAGE_JSON.version}`
+  const candidates = fs
+    .readdirSync(RELEASE_ROOT)
+    .filter(name => name.endsWith('.dmg'))
+    .filter(name => name.startsWith(prefix))
+    .filter(name => name.includes(ARCH))
+    .sort((a, b) => {
+      const aMtime = fs.statSync(path.join(RELEASE_ROOT, a)).mtimeMs
+      const bMtime = fs.statSync(path.join(RELEASE_ROOT, b)).mtimeMs
+      return bMtime - aMtime
+    })
+
+  return candidates.length > 0
+    ? path.join(RELEASE_ROOT, candidates[0])
+    : path.join(RELEASE_ROOT, `Hermes-${PACKAGE_JSON.version}-${ARCH}.dmg`)
+}
+
+function resolveNsisPath() {
+  // electron-builder NSIS artifactName template is 'Hermes-${version}-${os}-${arch}.${ext}'
+  if (!exists(RELEASE_ROOT)) return null
+  const candidates = fs
+    .readdirSync(RELEASE_ROOT)
+    .filter(name => /\.exe$/i.test(name) && /win/i.test(name))
+    .sort((a, b) => {
+      const aMtime = fs.statSync(path.join(RELEASE_ROOT, a)).mtimeMs
+      const bMtime = fs.statSync(path.join(RELEASE_ROOT, b)).mtimeMs
+      return bMtime - aMtime
+    })
+  return candidates.length > 0 ? path.join(RELEASE_ROOT, candidates[0]) : null
+}
+
+function ensureDmg() {
+  if (PLATFORM !== 'darwin') {
+    die('DMG mode is macOS-only; on Windows use the `nsis` mode instead.')
+  }
+  if (process.env.HERMES_DESKTOP_SKIP_BUILD === '1' && exists(resolveDmgPath())) {
+    return
+  }
+  run('npm', ['run', 'dist:mac:dmg'])
+}
+
+function ensureNsis() {
+  if (PLATFORM !== 'win32') {
+    die('NSIS mode is win32-only; on macOS use the `dmg` mode instead.')
+  }
+  if (process.env.HERMES_DESKTOP_SKIP_BUILD === '1' && resolveNsisPath()) {
+    return
+  }
+  run('npm', ['run', 'dist:win:nsis'])
+}
+
+function openApp() {
+  if (!exists(APP.binary)) {
+    die(`Missing packaged app: ${APP.binary}`)
+  }
+
+  if (PLATFORM === 'darwin') {
+    run('open', ['-n', APP.appPath])
+  } else if (PLATFORM === 'win32') {
+    // Spawn detached so the test script exits while the app keeps running.
+    spawn(APP.binary, [], { detached: true, stdio: 'ignore' }).unref()
+  } else {
+    spawn(APP.binary, [], { detached: true, stdio: 'ignore' }).unref()
+  }
+}
+
+function openDmg() {
+  if (PLATFORM !== 'darwin') {
+    die('DMG mode is macOS-only.')
+  }
+  const dmgPath = resolveDmgPath()
+  if (!exists(dmgPath)) {
+    die(`Missing DMG: ${dmgPath}`)
+  }
+  run('open', [dmgPath])
+}
+
+const CREDENTIAL_ENV_SUFFIXES = [
+  '_API_KEY',
+  '_TOKEN',
+  '_SECRET',
+  '_PASSWORD',
+  '_CREDENTIALS',
+  '_ACCESS_KEY',
+  '_PRIVATE_KEY',
+  '_OAUTH_TOKEN'
+]
+
+const CREDENTIAL_ENV_NAMES = new Set([
+  'ANTHROPIC_BASE_URL',
+  'ANTHROPIC_TOKEN',
+  'AWS_ACCESS_KEY_ID',
+  'AWS_SECRET_ACCESS_KEY',
+  'AWS_SESSION_TOKEN',
+  'CUSTOM_API_KEY',
+  'GEMINI_BASE_URL',
+  'OPENAI_BASE_URL',
+  'OPENROUTER_BASE_URL',
+  'OLLAMA_BASE_URL',
+  'GROQ_BASE_URL',
+  'XAI_BASE_URL'
+])
+
+function isCredentialEnvVar(name) {
+  if (CREDENTIAL_ENV_NAMES.has(name)) return true
+  return CREDENTIAL_ENV_SUFFIXES.some(suffix => name.endsWith(suffix))
+}
+
+function launchFresh() {
+  if (!exists(APP.binary)) {
+    die(`Missing app executable: ${APP.binary}`)
+  }
+
+  const sandbox = fs.mkdtempSync(`${FRESH_SANDBOX_ROOT}-`)
+  const userDataDir = path.join(sandbox, 'electron-user-data')
+  const hermesHome = path.join(sandbox, 'hermes-home')
+  const cwd = path.join(sandbox, 'workspace')
+
+  fs.mkdirSync(userDataDir, { recursive: true })
+  fs.mkdirSync(hermesHome, { recursive: true })
+  fs.mkdirSync(cwd, { recursive: true })
+
+  // Strip every credential-shaped env var so the sandbox is actually fresh.
+  const env = {}
+  for (const [key, value] of Object.entries(process.env)) {
+    if (isCredentialEnvVar(key)) continue
+    env[key] = value
+  }
+
+  env.HERMES_DESKTOP_CWD = cwd
+  env.HERMES_DESKTOP_IGNORE_EXISTING = '1'
+  env.HERMES_DESKTOP_TEST_MODE = 'fresh-install'
+  env.HERMES_DESKTOP_USER_DATA_DIR = userDataDir
+  env.HERMES_HOME = hermesHome
+  delete env.HERMES_DESKTOP_HERMES
+  delete env.HERMES_DESKTOP_HERMES_ROOT
+
+  const child = spawn(APP.binary, [], {
+    cwd: os.homedir(),
+    detached: true,
+    env,
+    stdio: 'ignore'
+  })
+  child.unref()
+
+  console.log('\nFresh install sandbox:')
+  console.log(`  root: ${sandbox}`)
+  console.log(`  electron userData: ${userDataDir}`)
+  console.log(`  HERMES_HOME: ${hermesHome}`)
+  console.log(`  cwd: ${cwd}`)
+
+  return { runtimeRoot: path.join(hermesHome, 'hermes-agent', 'venv') }
+}
+
+// Validate the packaged bundle matches the thin-installer architecture:
+//   - The Hermes Agent Python payload is NOT shipped (it's fetched at first
+//     launch via install.ps1's stage protocol).
+//   - install-stamp.json IS shipped in resources/ with a valid commit + branch.
+//   - native-deps/@homebridge/node-pty-prebuilt-multiarch/ IS shipped with
+//     the package.json + lib/ + at least one .node binary (the renderer's
+//     integrated terminal needs this; see Phase 1F.6).
+//   - The renderer's dist/index.html is reachable (either unpacked or
+//     inside app.asar).
+function validateBundle() {
+  if (!exists(APP.binary)) {
+    die(`Missing packaged app binary: ${APP.binary}`)
+  }
+
+  // Negative assertion: the OLD fat-installer factory payload must NOT be
+  // present anymore. If a stray ship of hermes_cli sneaks back in we want
+  // to fail loudly rather than re-introduce the 400MB delta we just removed.
+  const staleFactoryMarker = path.join(APP.resourcesPath, 'hermes-agent', 'hermes_cli', 'main.py')
+  if (exists(staleFactoryMarker)) {
+    die(
+      `Thin-installer regression: factory-payload file should NOT be in the package: ${staleFactoryMarker}`
+    )
+  }
+
+  // Positive assertion: install-stamp.json carries a sane commit + branch
+  const stampPath = path.join(APP.resourcesPath, 'install-stamp.json')
+  if (!exists(stampPath)) {
+    die(`Missing install-stamp.json (required for first-launch bootstrap pinning): ${stampPath}`)
+  }
+  let stamp
+  try {
+    stamp = JSON.parse(fs.readFileSync(stampPath, 'utf8'))
+  } catch (err) {
+    die(`install-stamp.json is not valid JSON: ${err.message}`)
+  }
+  if (!stamp.commit || typeof stamp.commit !== 'string' || stamp.commit.length < 7) {
+    die(`install-stamp.json is missing a usable commit field: ${JSON.stringify(stamp)}`)
+  }
+  if (!stamp.branch || typeof stamp.branch !== 'string') {
+    die(`install-stamp.json is missing the branch field: ${JSON.stringify(stamp)}`)
+  }
+
+  // Positive assertion: node-pty native deps shipped
+  const native = expectedNativeDepPaths()
+  if (!exists(native.packageJson)) {
+    die(`Missing node-pty package.json in resources/native-deps: ${native.packageJson}`)
+  }
+  if (!exists(native.libIndex)) {
+    die(`Missing node-pty lib/index.js in resources/native-deps: ${native.libIndex}`)
+  }
+  if (!exists(native.prebuildsDir)) {
+    die(`Missing node-pty prebuilds dir for ${PLATFORM}-${ARCH}: ${native.prebuildsDir}`)
+  }
+  const nodeBinaries = fs.readdirSync(native.prebuildsDir).filter(name => name.endsWith('.node'))
+  if (nodeBinaries.length === 0) {
+    die(`No .node native binaries found in: ${native.prebuildsDir}`)
+  }
+  // Darwin requires a runtime-execed spawn-helper alongside pty.node; missing
+  // it manifests as "ENOENT: spawn-helper" on first pty.spawn() call.
+  if (PLATFORM === 'darwin') {
+    const spawnHelper = path.join(native.prebuildsDir, 'spawn-helper')
+    if (!exists(spawnHelper)) {
+      die(`Missing node-pty spawn-helper (required on darwin): ${spawnHelper}`)
+    }
+  }
+
+  // Renderer payload check (either unpacked or in the asar)
+  if (exists(APP.unpackedDistIndex)) {
+    return { stamp, nodeBinaries }
+  }
+  if (!exists(APP.asarPath)) {
+    die(`Missing renderer payload: neither ${APP.unpackedDistIndex} nor ${APP.asarPath} exists`)
+  }
+  const files = listPackage(APP.asarPath)
+  // Normalize separators because @electron/asar's listPackage returns
+  // backslash-prefixed entries on Windows ('\\dist\\index.html') and
+  // forward-slash on Unix.
+  const normalized = files.map(f => f.replace(/\\/g, '/').replace(/^\/+/, ''))
+  if (!normalized.includes('dist/index.html')) {
+    die(`Missing renderer payload file in app.asar: ${APP.asarPath} (expected dist/index.html)`)
+  }
+  return { stamp, nodeBinaries }
+}
+
+function printArtifacts(options = {}) {
+  const runtimeRoot = options.runtimeRoot || VENV_ROOT
+  const stamp = options.stamp
+
+  console.log('\nDesktop artifacts:')
+  console.log(`  app: ${APP.appPath}`)
+  if (PLATFORM === 'darwin') {
+    console.log(`  dmg: ${resolveDmgPath()}`)
+  } else if (PLATFORM === 'win32') {
+    const exe = resolveNsisPath()
+    if (exe) console.log(`  installer: ${exe}`)
+  }
+  console.log(`  runtime: ${runtimeRoot}`)
+  if (stamp) {
+    console.log(`  install-stamp: ${stamp.commit.slice(0, 12)} on ${stamp.branch}`)
+  }
+  if (options.nodeBinaries && options.nodeBinaries.length > 0) {
+    console.log(`  node-pty binaries: ${options.nodeBinaries.join(', ')}`)
+  }
+}
+
+function help() {
+  console.log(`Usage:
+  npm run test:desktop:existing  # build packaged app, launch with normal PATH/existing Hermes
+  npm run test:desktop:fresh     # build packaged app, launch with temp userData + HERMES_HOME
+  npm run test:desktop:dmg       # (macOS only) build DMG and open it
+  npm run test:desktop:nsis      # (win32 only) build NSIS installer
+  npm run test:desktop:all       # build installer, validate app payload, print paths
+
+Fast rerun (skip rebuild if the packaged app already exists):
+  HERMES_DESKTOP_SKIP_BUILD=1 npm run test:desktop:fresh
+`)
+}
+
+ensurePlatformBuilds()
+
+if (MODE === 'existing') {
+  ensurePackagedApp()
+  const result = validateBundle()
+  openApp()
+  printArtifacts(result)
+} else if (MODE === 'fresh') {
+  ensurePackagedApp()
+  const result = validateBundle()
+  printArtifacts({ ...launchFresh(), ...result })
+} else if (MODE === 'dmg') {
+  ensureDmg()
+  openDmg()
+  printArtifacts()
+} else if (MODE === 'nsis') {
+  ensureNsis()
+  printArtifacts(validateBundle())
+} else if (MODE === 'all') {
+  if (PLATFORM === 'darwin') {
+    ensureDmg()
+  } else if (PLATFORM === 'win32') {
+    ensureNsis()
+  } else {
+    ensurePackagedApp()
+  }
+  printArtifacts(validateBundle())
+} else {
+  help()
+}

apps/desktop/scripts/write-build-stamp.cjs

@@ -0,0 +1,126 @@
+"use strict"
+
+/**
+ * Writes apps/desktop/build/install-stamp.json with the git ref the desktop
+ * .exe should pin to at first-launch bootstrap time.  This file ships inside
+ * the packaged app via electron-builder's extraResources entry and is read
+ * by electron/main.cjs to drive the install.ps1 stage bootstrap flow.
+ *
+ * Schema (subject to bump via STAMP_SCHEMA_VERSION):
+ *   {
+ *     "schemaVersion": 1,
+ *     "commit":        "<40-char SHA>",
+ *     "branch":        "<branch name>",
+ *     "builtAt":       "<ISO 8601 UTC timestamp>",
+ *     "dirty":         true|false,
+ *     "source":        "ci" | "local"
+ *   }
+ *
+ * Source preference order:
+ *   1. CI env vars ($GITHUB_SHA / $GITHUB_REF_NAME) -- avoid edge cases with
+ *      shallow clones, detached HEADs, etc. in CI.
+ *   2. Local `git rev-parse` against the parent repo (../..).
+ *
+ * Dev / out-of-repo builds without git produce an explicit error rather than
+ * silently writing an unstamped manifest -- the packaged app refuses to
+ * bootstrap without a stamp.
+ */
+
+const fs = require("fs")
+const path = require("path")
+const { execSync } = require("child_process")
+
+const STAMP_SCHEMA_VERSION = 1
+
+const DESKTOP_ROOT = path.resolve(__dirname, "..")
+const REPO_ROOT = path.resolve(DESKTOP_ROOT, "..", "..")
+const OUT_DIR = path.join(DESKTOP_ROOT, "build")
+const OUT_FILE = path.join(OUT_DIR, "install-stamp.json")
+
+function tryExec(cmd, opts) {
+  try {
+    return execSync(cmd, { encoding: "utf8", stdio: ["ignore", "pipe", "ignore"], ...opts }).trim()
+  } catch {
+    return null
+  }
+}
+
+function fromCI() {
+  const sha = process.env.GITHUB_SHA
+  if (!sha) return null
+  const branch = process.env.GITHUB_REF_NAME || process.env.GITHUB_HEAD_REF || null
+  return {
+    commit: sha,
+    branch: branch,
+    dirty: false, // CI builds from a checkout-of-ref by definition
+    source: "ci"
+  }
+}
+
+function fromLocalGit() {
+  const sha = tryExec("git rev-parse HEAD", { cwd: REPO_ROOT })
+  if (!sha) return null
+  const branch = tryExec("git rev-parse --abbrev-ref HEAD", { cwd: REPO_ROOT })
+  // `git status --porcelain -uno` is empty iff tracked files match HEAD.
+  // We exclude untracked files (-uno) intentionally: a developer who's
+  // checked out an installer scratch dir alongside the repo shouldn't
+  // poison every local build with a [DIRTY] stamp.  We DO care about
+  // tracked-but-modified files because those mean the .exe content
+  // differs from the commit being pinned.
+  const status = tryExec("git status --porcelain -uno", { cwd: REPO_ROOT })
+  const dirty = status !== null && status.length > 0
+  return {
+    commit: sha,
+    branch: branch === "HEAD" ? null : branch, // detached HEAD -> null
+    dirty: dirty,
+    source: "local"
+  }
+}
+
+function main() {
+  const stamp = fromCI() || fromLocalGit()
+  if (!stamp || !stamp.commit) {
+    console.error(
+      "[write-build-stamp] ERROR: could not determine git commit.\n" +
+        "  - $GITHUB_SHA not set\n" +
+        "  - `git rev-parse HEAD` failed at " +
+        REPO_ROOT +
+        "\n" +
+        "Packaged builds require a git ref to pin first-launch install.ps1\n" +
+        "against. Run from a git checkout or set $GITHUB_SHA explicitly."
+    )
+    process.exit(1)
+  }
+
+  if (stamp.dirty) {
+    console.warn(
+      "[write-build-stamp] WARNING: working tree is dirty.\n" +
+        "  Pinning to " +
+        stamp.commit.slice(0, 12) +
+        " but the packaged code may differ from that commit.\n" +
+        "  Commit your changes before publishing this build."
+    )
+  }
+
+  const payload = {
+    schemaVersion: STAMP_SCHEMA_VERSION,
+    commit: stamp.commit,
+    branch: stamp.branch,
+    builtAt: new Date().toISOString(),
+    dirty: stamp.dirty,
+    source: stamp.source
+  }
+
+  fs.mkdirSync(OUT_DIR, { recursive: true })
+  fs.writeFileSync(OUT_FILE, JSON.stringify(payload, null, 2) + "\n", "utf8")
+  console.log(
+    "[write-build-stamp] wrote " +
+      path.relative(REPO_ROOT, OUT_FILE) +
+      " -> " +
+      stamp.commit.slice(0, 12) +
+      (stamp.branch ? " (" + stamp.branch + ")" : "") +
+      (stamp.dirty ? " [DIRTY]" : "")
+  )
+}
+
+main()