Use mise to install zig + go in release CI; cut 0.0.4

`mlugg/setup-zig` was chasing mirrors for ~4 minutes on every run
(see v0.0.1 / v0.0.2 logs) and `actions/setup-go` was spending
another ~4 minutes downloading Go before patterm started building.
mise already manages the project's zig pin; adding `go = "1.26.3"`
to `.mise.toml` (matching go.mod) lets `jdx/mise-action@v2` install
both with one cached step. Subsequent runs reuse the mise cache
instead of re-resolving mirror URLs and re-downloading toolchains.

Also adds an `actions/cache@v4` step for `~/.cache/go-build` and
`~/go/pkg/mod` keyed on `go.sum` so `go build` itself doesn't
re-pull modules every tag push.

.gitea/workflows/release.yml

@@ -11,14 +11,19 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      - uses: actions/setup-go@v5
+      - uses: jdx/mise-action@v2
         with:
-          go-version-file: go.mod
           cache: true
 
-      - uses: mlugg/setup-zig@v1
+      - name: Cache Go modules
+        uses: actions/cache@v4
         with:
-          version: 0.15.2
+          path: |
+            ~/.cache/go-build
+            ~/go/pkg/mod
+          key: ${{ runner.os }}-go-${{ hashFiles('**/go.sum') }}
+          restore-keys: |
+            ${{ runner.os }}-go-
 
       - name: Build libghostty-vt
         run: make deps

.mise.toml

@@ -0,0 +1,10 @@
+# mise config — `mise install` provisions the tools `make deps` needs.
+#
+# libghostty-vt is built from a pinned upstream Ghostty commit; that
+# commit's build.zig.zon pins minimum_zig_version = 0.15.2. We match
+# it here so contributors don't have to puzzle out the version from
+# a deep upstream file. The go pin matches go.mod so CI and local
+# builds use the same toolchain.
+[tools]
+zig = "0.15.2"
+go = "1.26.3"

CHANGELOG.md

@@ -6,7 +6,207 @@ loosely follows [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## [Unreleased]
 
+## [0.0.4] - 2026-05-15
+
+### Changed
+- Release workflow (`.gitea/workflows/release.yml`) now provisions
+  Zig and Go through `jdx/mise-action@v2`, reading the versions from
+  `.mise.toml` (zig 0.15.2, go 1.26.3). Both toolchains were
+  previously installed via `mlugg/setup-zig` and `actions/setup-go`,
+  whose mirror chase / GitHub fetch combined for ~8 minutes per run
+  before any patterm code compiled. mise pulls each tool once and
+  caches the install dir, so subsequent runs hit the cache instead of
+  re-downloading. `make deps` still resolves zig via `mise which zig`
+  with a PATH fallback; `go.mod` already pinned `go 1.26.3`, so the
+  new `go` entry in `.mise.toml` just keeps CI and local builds on
+  the same toolchain.
+- A Go module/build cache step (`actions/cache@v4`, keyed on
+  `go.sum`) was added so `go build` doesn't re-download dependencies
+  on every tag push.
+
+## [0.0.3] - 2026-05-15
+
 ### Added
+- Auto-summarization for top-level agent tabs. patterm now loads
+  `$XDG_CONFIG_HOME/patterm/settings.json`, enables Codex-based
+  summaries by default (`gpt-5.4-mini`; OpenCode defaults to
+  `opencode-go/minimax-m2.7`), and can run Codex, OpenCode, or opt-in
+  Claude summarizers with configurable model names. Summary
+  attempts are armed by meaningful human input, wait for recent output
+  to go quiet, and respect a minimum cadence so unchanged tabs are not
+  summarized on a timer. The active thread summary appears under the
+  top tab title and in the sidebar below the Agent Tree section.
+- Settings overlay reachable from the command palette via
+  `Open Settings`. The searchable Settings picker opens
+  `Agents / Auto-summarization`, where users can enable/disable
+  summaries, choose provider, edit provider model names, cycle cadence,
+  test the selected summarizer (`patterm okay`), summarize the current
+  top-level agent immediately, and explicitly save or cancel draft
+  settings changes. Cadence choices match Solo: `15s`, `30s`, and
+  `1m`; the value is a minimum quiet/activity gap before another
+  summary attempt for the same top-level agent, not a background
+  periodic timer.
+
+### Changed
+- Command palette UX overhaul. The single flat list grew section
+  bands (`── Focused ──`, `── Open ──`, `── Spawn ──`, `── Quit ──`)
+  so the rows are scannable at a glance; cursor navigation skips
+  the dim header rows transparently. A chip strip — `[All] Open
+  Spawn Close` — sits below the query line and tracks the active
+  macro filter; `Tab` / `Shift-Tab` cycle through the chips, and
+  the typed-prefix macros (`sw `, `sp `, `k `) still work and now
+  collapse the whole prefix on a single backspace instead of
+  leaving a stray `sw` behind. The title bar surfaces the current
+  focus subject (`on: <child>` / `pad: <name>`) so the user knows
+  which Focused row is targeting what. The duplicate global Close
+  list is gone — close is reachable via the Focused-section action,
+  the `k ` macro / `[Close]` chip, or the new `Ctrl-X` inline close
+  on a Switch row. The "(current)" marker on the focused Switch row
+  became a leading `▶`. The empty-state hint now reads `no matches
+  · ⌫ to widen` instead of bare `no matches`. The middle divider
+  shows a `▼ N more` / `▲ N above` scroll indicator when the list
+  overflows, and the footer carries a `cursor/total` counter.
+- Spawn verbs are unified on **Spawn**: `Run process: …` →
+  `Spawn process: …`, `New Terminal` → `Spawn terminal`, and the
+  freeform-form row is now `Spawn process… (custom)` so the
+  trailing ellipsis still signals it opens a form.
+- Filtering switched from binary fuzzy-include to scored ranking.
+  Prefix matches beat word-boundary matches beat substring matches
+  beat scattered-fuzzy matches; ties fall back to section order so
+  a Focused-section hit always outranks an equally tight Spawn
+  hit. The matched characters in the rendered label render in
+  accent+bold so the user can see why a row matched.
+- Rename forms split the long subject (`scratchpad:
+  some-really-long-name.md`) onto its own dim row above the input
+  so the title bar no longer truncates with an ellipsis when the
+  subject name is wide.
+- New palette accelerators: `Alt-1` … `Alt-9` quick-pick the Nth
+  visible row, `Home` / `End` jump to first / last selectable row,
+  `?` (with empty query) opens an inline keybinding cheat-sheet
+  which any further keystroke dismisses, and `Ctrl-R` inside the
+  Spawn-process form toggles "Relaunch on exit" without leaving
+  the command field.
+
+### Fixed
+- Error/status flashes now restore the currently focused pane instead
+  of drawing the empty-state hint over a running agent or process.
+- Release workflow (`.gitea/workflows/release.yml`) now uses
+  `mlugg/setup-zig@v2` instead of the deprecated `@v1`. v1 hard-coded
+  the pre-0.14 tarball name (`zig-linux-x86_64-<ver>.tar.xz`), so
+  every mirror and the official `ziglang.org/builds` returned 404 for
+  Zig 0.15.2 and the v0.0.1 / v0.0.2 tag pushes never produced a
+  release asset. v2 uses the post-0.14 `zig-x86_64-linux-<ver>.tar.xz`
+  layout, so the runner can fetch Zig and build patterm.
+- Typing into a focused child while its emulator viewport is
+  scrolled up into scrollback history now auto-snaps the viewport
+  back to the live area. Previously the keystroke reached the
+  child PTY but the input box was off-screen below the visible
+  region, so it looked like typing did nothing. Wheel scrolling
+  and Ctrl-B are unchanged; only forwarded keystrokes snap.
+- Top tab bar now keeps the top-level agent's tab highlighted
+  when focus is on one of its sub-agents (or on a Processes pane
+  entry, matching the existing agent-tree behavior). Previously
+  the tab would lose its highlight as soon as you stepped into a
+  child agent, even though you were still within that thread.
+
+### Changed
+- MCP tool descriptions and `help('coordination')` /
+  `help('readiness')` now spell out that a sub-agent's reply to
+  `send_message` lands in the caller's own pane (tagged
+  `[sub-agent:<name>]`), not in the sub-agent's output. The canonical
+  wait-for-reply pattern — `send_message` → `timer_fire_when_idle_any`
+  on the sub-agent → read your own pane — is now called out on
+  `send_message`, `wait_for_pattern`, both `timer_fire_when_idle_*`,
+  the help topics, and the server-instructions preamble every agent
+  reads at startup. Previously `wait_for_pattern` was the obvious
+  blocking primitive in the catalog, and agents routinely called it
+  against the sub-agent for a reply that had already arrived in their
+  own pane, deadlocking until the wait timed out. No behaviour
+  changes; descriptions only.
+- Agent-initiated `spawn_agent` and `spawn_process` MCP calls no
+  longer steal viewport focus from the currently active tab. The
+  new child still appears in the sidebar and tab bar; switch to it
+  explicitly via the palette or `select_process`. Palette-initiated
+  spawns and persistence restores are unchanged — they still auto-
+  focus the new pane.
+- Sidebar rows (Processes, Agent Tree, Scratchpads) now truncate
+  overflowing names with a trailing `…` instead of spilling into
+  the main viewport. The focused row marquees its name when it
+  overflows — 1 s hold on the head, ~150 ms per cell scroll until
+  the tail is visible, 1 s hold on the tail, snap back. Row
+  position never moves while the marquee animates. When budget is
+  tight, the trailing timer indicator drops before the name
+  ellipses, since the name is the only identifier the row carries.
+
+## [0.0.2] - 2026-05-15
+
+### Added
+- `.mise.toml` pinning `zig = "0.15.2"` (the minimum version the
+  vendored Ghostty commit requires). Contributors run
+  `mise install` once; the Makefile picks up the resulting `zig`
+  binary automatically via `mise which zig` and falls back to
+  PATH when mise isn't available, so the existing build flow
+  still works.
+- ASCII-video stress benchmarks (`internal/app/bench_test.go`):
+  per-frame and per-stream variants at 30 / 60 / 120 fps targets,
+  three workload fixtures (8-colour cells, 24-bit truecolor cells,
+  and a Bad-Apple-style 1-bit pattern). Each stream benchmark
+  reports `µs/frame`, an achievable `fps_ceiling`, and `budget_pct`
+  so you can read off "do we hit N fps?" directly. A matching
+  Pipeline_ASCIIVideo_* set includes libghostty-vt's em.Write CGO
+  and an io.Discard stdout write so the FPS claim reflects the
+  whole pipeline, not just the renderer.
+- MCP `initialize.instructions`, the `spawn_agent` tool description
+  (visible to LLMs via `tools/list`), and the `help('spawning')`
+  topic now spell out — in the three places vendor TUIs actually
+  consult — that the connected `patterm` MCP server is the only
+  correct way to drive the host. Anti-patterns called out by name:
+  (a) trying to launch `patterm` / `patterm mcp-stdio` themselves,
+  (b) piping JSON-RPC into the per-PID Unix socket via `perl` /
+  `nc` / `socat` / `curl`, and (c) shelling out to `claude` /
+  `codex` / `opencode` to start a peer. Each of those bypasses
+  caller identity, so a sub-agent spawned that way reads back as
+  a stray top-level tab instead of a child under the spawning
+  agent. Codex was hitting (b) and (c) in practice — this is the
+  fix.
+- `--debug[=DIR]` flag captures detailed run artefacts for offline
+  analysis: a verbose `patterm.log` (the existing `PATTERM_DEBUG_LOG`
+  stream), an `events.jsonl` lifecycle log (spawn / exit / idle-state
+  changes with timestamps), and per-child `<id>.raw` files containing
+  the raw PTY byte stream. With no argument, the dated subdir
+  `$XDG_STATE_HOME/patterm/debug/YYYYMMDD-HHMMSS` is used; pass an
+  explicit path to override. All output goes to files — stdout/stderr
+  are untouched.
+- `--profile[=DIR]` flag captures pprof data plus concrete
+  performance counters for performance work: `cpu.pprof` (running
+  for the lifetime of the session), plus `heap.pprof` and
+  `goroutine.pprof` snapshots written on shutdown; alongside them,
+  a per-hot-path metrics tracker writes `metrics.jsonl` (one row
+  per second with chunk/byte rates, per-stage mean and max
+  latencies, and cache hit rates) plus a final `metrics.json`
+  aggregate and a human-readable `summary.txt` on exit.
+  Instrumented hot paths: `OnPTYOut`, viewport `renderer.Render`,
+  host stdout writes, libghostty-vt `emulator.Write` / `Title`,
+  sidebar / tab bar / status line draws (with cache-hit
+  accounting), snapshot replays, and the chrome ticker (so you can
+  see how often it fires with nothing to do). Defaults to
+  `$XDG_STATE_HOME/patterm/profile/YYYYMMDD-HHMMSS`. All
+  diagnostics (startup, errors) are written to `profile.log`
+  inside the dir, never to stdout/stderr.
+- Renderer benchmark suite (`internal/app/bench_test.go`). Three
+  workload fixtures — plain ASCII, SGR-styled lines, and a
+  ratatui-style cursor-shuffling burst — plus an OSC-gate
+  micro-benchmark. Run via `go test -bench=. -benchmem
+  ./internal/app/`. Gives a stable reference for the per-chunk
+  cost of the viewport renderer so future changes can be compared
+  apples-to-apples.
+- "New Terminal" entry in the command palette spawns a bare interactive
+  `$SHELL` pane (kind `terminal`). Unlike "Run process: …" presets,
+  which are session-persistent and reachable via `restart_process`,
+  terminals are ephemeral — once they exit they vanish from the
+  Processes sidebar instead of lingering as a dead row. The default
+  `shell` process preset that previously seeded on first run has been
+  removed; this entry replaces it.
 - Per-child idle-state classifier with five states (`idle`, `working`,
   `thinking`, `permission`, `error`) and three pluggable strategies:
   `output_activity` (claude / opencode defaults), `osc_title_stability`
@@ -98,6 +298,11 @@ loosely follows [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
   after a child program disables mouse tracking.
 
 ### Changed
+- The palette's per-child "Kill <name>" action is now labelled
+  "Close <name>". The underlying signal (SIGTERM) and behaviour are
+  unchanged; the new label matches the existing "Close agent: …"
+  context entry and reads less violent for what is really just a
+  graceful termination.
 - `timer_wait` is now a thin wrapper over the shared timer manager
   (`timer_set` semantics). Existing callers see no behavioural change;
   the timer is visible in `timer_list` while it's pending.
@@ -108,6 +313,47 @@ loosely follows [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
   renders the canonical `--flag` form.
 
 ### Fixed
+- `make deps` now builds libghostty-vt with `-Doptimize=ReleaseFast`
+  instead of zig's silent `Debug` default, and resolves `zig`
+  through `mise` when a project `.mise.toml` pins it. The
+  default-Debug build shipped an unoptimised CSI/SGR parser that
+  ate 16-29 ms per 30-70 KiB full-screen frame in benchmarks,
+  capping the entire PTY-to-host pipeline at 34-63 fps. After the
+  rebuild the same pipeline runs at **930-2030 fps**: 27-32× the
+  prior throughput, and 7-16× margin over 120 fps for full-screen
+  truecolor ASCII video. Static library size drops from 33 MiB to
+  13 MiB. Override with `make deps GHOSTTY_VT_OPTIMIZE=Debug` only
+  when debugging the upstream library itself. Apply on existing
+  checkouts with `mise install && make clean-deps && make deps`.
+- Long claude session resume (and codex steady-state rendering) is
+  noticeably faster. Two costs that scaled per-PTY-chunk are now
+  deferred or short-circuited: (1) `drawSidebar()` used to run
+  synchronously for every chunk that scrolled — on a session
+  resume where every chunk scrolls, this rebuilt the full sidebar
+  string hundreds of times for a frame that was almost always
+  cache-equal. The sidebar now signals dirty and the chrome ticker
+  (60 Hz) handles the repaint. (2) `pumpChild` polled the
+  emulator's OSC title after every PTY chunk via CGO, even for
+  chunks (the common case under codex/ratatui) that carry no OSC
+  bytes at all. The poll is now gated on a containsOSC scan over
+  the chunk.
+- Click-and-drag text selection from alt-screen TUIs (codex in
+  particular) now works. Patterm used to keep host SGR mouse
+  reporting armed continuously, which forced the host terminal to
+  forward every click as an escape sequence and prevented native
+  selection. The host's mouse mode now follows the focused child's
+  screen side: primary-screen children keep mouse armed (so wheel
+  scrollback works), alt-screen children get host mouse disabled by
+  default. Alt-screen TUIs that need mouse events (vim, less, etc.)
+  re-enable mouse-mode themselves; the viewport renderer forwards
+  those toggles to the host while the child is on alt. Leaving alt
+  re-arms host mouse reporting so wheel scrollback resumes.
+- Exited terminal panes (kind `terminal`, including those launched via
+  the new "New Terminal" palette entry or MCP `spawn_process` with
+  `kind=terminal`) are now removed from the session and the Processes
+  sidebar as soon as they exit. Previously they stuck around as a
+  greyed-out row indistinguishable from an exited command process,
+  even though terminals have no restart path.
 - `whoami` and `help("timers")` now advertise the full Solo-parity timer
   surface (`timer_set`, `timer_fire_when_idle_any`,
   `timer_fire_when_idle_all`, `timer_cancel`, `timer_pause`,

Makefile

@@ -20,10 +20,30 @@ $(SOURCE)/.git/HEAD:
 
 deps-fetch: $(SOURCE)/.git/HEAD
 
+# Zig's `standardOptimizeOption` defaults to .Debug when no
+# -Doptimize is passed, which makes libghostty-vt's CSI/SGR parser
+# an order of magnitude slower — truecolor full-screen frames spend
+# ~16-29 ms each in em.Write under Debug (see
+# internal/app/bench_test.go BenchmarkEmulator_Write_*), which caps
+# the full PTY-to-host pipeline at ~60 fps. ReleaseFast is the
+# right default for the shipped artefact. Override with
+# `make deps GHOSTTY_VT_OPTIMIZE=Debug` when you actually want a
+# debug build of the upstream lib.
+GHOSTTY_VT_OPTIMIZE ?= ReleaseFast
+
+# Resolve zig via the project's mise pin (.mise.toml) when available,
+# falling back to whatever's on PATH. mise keeps the zig version in
+# lockstep with what the pinned ghostty commit requires; without it,
+# contributors have to chase the version requirement themselves.
+ZIG := $(shell command -v mise >/dev/null && mise which zig 2>/dev/null || command -v zig 2>/dev/null)
+
 $(INSTALL)/lib/libghostty-vt.a: $(SOURCE)/.git/HEAD
-	@command -v zig >/dev/null || { echo "ERROR: zig not on PATH (need >=0.15.2 to build libghostty-vt)"; exit 1; }
-	@echo ">> building libghostty-vt with zig"
-	@cd $(SOURCE) && zig build -Demit-lib-vt --prefix $(INSTALL)
+	@if [ -z "$(ZIG)" ]; then \
+		echo "ERROR: zig not available. Run \`mise install\` (see .mise.toml — needs zig 0.15.2) or install zig manually."; \
+		exit 1; \
+	fi
+	@echo ">> building libghostty-vt with $(ZIG) (optimize=$(GHOSTTY_VT_OPTIMIZE))"
+	@cd $(SOURCE) && $(ZIG) build -Demit-lib-vt -Doptimize=$(GHOSTTY_VT_OPTIMIZE) --prefix $(INSTALL)
 	@test -f $(INSTALL)/lib/libghostty-vt.a || { echo "ERROR: expected static lib at $(INSTALL)/lib/libghostty-vt.a"; exit 1; }
 	@echo ">> libghostty-vt installed under $(INSTALL)"
 

TODO.md

@@ -1,16 +1,102 @@
-- [ ] We should probably rename the Kill <Process> terminology to Close <Process> instead, across processes and agents.
-- [ ] Exited shells are still being treated as active processes. They should be removed from the process list when they exit.
-- [ ] Shells should be renamed to terminals. "New Terminal" etc.
-- [ ] Codex seemed to think that it needed to launch patterm itself to get the mcp working
-- [ ] I cant click and drag to select text from codex
-- [ ] codex uses perl to interact with the socket rather than calling mcp tools
-  - when it _did_ open a sub claude it opened it as a separate tab rather than a sub-agent.
-- [ ] codex rendering is VERY slow
-  - maybe we need to use diffing rather than rendering the entire viewport for performance
-- We should add a --debug and --profile flag, so we can get detailed performance data and full logs of the agent output to be debugged later on.
-  - I don't mind what format this is in, ideally easy for LLMs to understand
-- [ ] Resuming a long claude session takes a couple of seconds for the entire buffer to load in, it looks like it's scrolling down for a couple seconds.
-  - In raw alacritty this is instant, so there's some sort of performance issue with patterm's terminal emulation.
+# Perf Audit (reviewed 2026-05-15)
+Findings that survived the 2026-05-15 review pass. Low and marginal
+items from the original sweep were removed; remaining items have enough
+measured or workflow evidence to justify action.
+
+Baseline benchmark numbers (`go test -bench=. ./internal/app/`, AMD
+Ryzen 7 7800X3D, libghostty-vt **ReleaseFast** after the Makefile
+fix landed):
+
+```
+# Renderer alone
+ViewportRenderer_PlainASCII       229 MB/s     1.3 KB/op    6 allocs/op
+ViewportRenderer_StyledLines       89 MB/s    91   KB/op  4325 allocs/op
+ViewportRenderer_RatatuiBurst      40 MB/s   365   KB/op 17306 allocs/op
+RendererThroughput_ReuseInstance   90 MB/s   316   KB/op 17380 allocs/op
+ContainsOSC_NoOSC                3050 MB/s     0   B/op     0 allocs/op
+
+# ASCII-video stream (renderer only — 3 sec at the target fps)
+ASCIIVideo_Stream_8Color_120fps     260 µs/frame  3845 fps_ceiling   3.1% budget
+ASCIIVideo_Stream_TrueColor_120fps  576 µs/frame  1735 fps_ceiling   6.9% budget
+
+# Full pipeline (em.Write + renderer + io.Discard write)
+Pipeline_ASCIIVideo_8Color_120fps     493 µs/frame  2030 fps_ceiling   5.9% budget
+Pipeline_ASCIIVideo_TrueColor_120fps 1075 µs/frame   931 fps_ceiling  12.9% budget
+
+# Emulator alone (libghostty-vt CSI/SGR parser)
+Emulator_Write_Stream_8Color_120fps    257 µs/frame  3890 fps_ceiling
+Emulator_Write_Stream_TrueColor_120fps 488 µs/frame  2051 fps_ceiling
+```
+
+The current pipeline still has large 120 fps headroom. The remaining
+renderer concern is multi-MiB styled replay latency and allocation
+churn, not normal steady-state frame budget.
+
+
+- [ ] **viewport renderer allocates heavily on SGR/CSI-heavy chunks.** [MEDIUM]
+  - Review evidence: five benchmark reps confirmed
+    `ViewportRenderer_StyledLines` at about 4,325 allocs per 16 KiB
+    chunk (~91.5 KB/op, roughly 1 alloc per 3.8 input bytes), and
+    `ViewportRenderer_RatatuiBurst` at about 17,306 allocs per chunk
+    (~365 KB/op). A 5 MiB styled resume benchmark allocated about
+    31 MB across 1.38M objects.
+  - Likely hot paths: generic CSI/SGR output in
+    `internal/app/viewport_renderer.go` sends many sequences through
+    `vr.shifter.Shift(vr.buf)`, while `internal/app/cursorshift.go`
+    returns a fresh `[]byte` via `pending.String()` on every
+    `Shift` call and parses CSI params through `string(raw)` /
+    `strings.Split`. The mode-helper `string(params)` conversions
+    are real, but probably not the main SGR-heavy cost.
+  - Fix direction: make `cursorShifter` write into caller-owned
+    scratch output or directly into the viewport renderer's pending
+    builder; parse CSI params from byte slices; pre-grow/reuse
+    renderer and shifter buffers. Re-run styled-lines, ratatui, and
+    5 MiB resume benchmarks; use pprof when available to confirm the
+    top allocation sites.
+
+- [ ] **large styled resume/replay dumps spend visible time in viewport rendering.** [MEDIUM]
+  - Review evidence: `BenchmarkSessionResume_5MiBStyled` measured
+    about 58 ms median and 63 ms p95 over five reps. The plain 5 MiB
+    benchmark was about 23-24 ms with only 21 allocs. The live path
+    renders focused PTY chunks through `renderer.Render`, then still
+    pays emulator writes, ring writes, event dispatch, stdout writes,
+    and real terminal paint.
+  - Scope: this is not a Codex steady-state throughput limit. A
+    100 KB/s stream is far below the styled renderer's ~80-90 MB/s
+    ceiling. It matters for multi-MiB burst replay, resume/startup
+    dumps, and dense full-screen churn.
+  - Fix direction: do the allocation fix first, since it should also
+    improve throughput. After that, invest further only if styled
+    resume traces remain user-visible or the styled-lines benchmark
+    is still under roughly 300 MB/s.
+
+- [ ] **wait_for_pattern re-scans the entire stream/grid while waiting.** [MEDIUM]
+  - `internal/app/host.go:476-493` (the `check` closure). On
+    `scope="scrollback"` it calls `c.StreamRead(0)` followed by
+    `stripANSIBytes(nil, b)`, so each check can copy, strip, and
+    search the full 1 MiB ring. On `scope="grid"` it calls
+    `PlainText()` and runs the regex against the full grid string.
+  - Caveat from review: the current chunk notifier coalesces bursts
+    with a buffered channel and has a 500 ms fallback, so this is not
+    necessarily one full scan per PTY chunk. It is still meaningful
+    for active waits on chatty panes.
+  - Fix direction: for `scrollback`, track the last checked stream
+    offset and search only new output plus a bounded overlap/scratch
+    buffer so matches spanning chunks are not missed. For `grid`,
+    dedupe on `ScreenVersion()` and skip work when the version has
+    not changed.
+
+- [ ] **search_output rebuilds and searches whole scrollback on every call.** [MEDIUM]
+  - `internal/app/host.go:428-437` compiles a fresh regex, reads the
+    stream from offset 0, strips ANSI for `kind="rendered"`, converts
+    the full buffer to a string, and splits it into lines before
+    applying `limit`. This is meaningful when agents poll the same
+    pattern; it is low impact for ad hoc searches.
+  - Fix direction: cache compiled regexes by pattern; cache stripped
+    rendered output by child id and stream end offset; avoid
+    `strings.Split` over the whole ring when only the first `limit`
+    matches are needed. Prefer an incremental search shape if this
+    becomes the standard "watch for marker" path.
 
 # On Hold
 - [ ] There's a unicode <?> being displayed in opencode [ON HOLD]

cmd/patterm/main.go

@@ -16,7 +16,10 @@ import (
 	"context"
 	"fmt"
 	"os"
+	"path/filepath"
+	"runtime"
 	"runtime/debug"
+	"runtime/pprof"
 	"time"
 
 	flag "github.com/spf13/pflag"
@@ -49,7 +52,13 @@ func main() {
 	var (
 		projectDir  = flag.String("project", "", "project directory (default $PWD)")
 		showVersion = flag.Bool("version", false, "print version and exit")
+		debugDir    = flag.String("debug", "", "write debug logs + per-child raw PTY output to DIR (auto-picks a dated subdir under $XDG_STATE_HOME/patterm/debug when DIR is omitted)")
+		profileDir  = flag.String("profile", "", "write pprof files (cpu/heap/goroutine) and live perf counters (metrics.jsonl per-second, metrics.json + summary.txt on exit) to DIR (auto-picks a dated subdir under $XDG_STATE_HOME/patterm/profile when DIR is omitted)")
 	)
+	// Allow bare `--debug` / `--profile` with no value — pflag treats
+	// them as boolean-shaped strings, picking a sensible default dir.
+	flag.Lookup("debug").NoOptDefVal = "auto"
+	flag.Lookup("profile").NoOptDefVal = "auto"
 	flag.Parse()
 
 	if *showVersion {
@@ -73,15 +82,104 @@ func main() {
 		die("chdir %s: %v", cwd, err)
 	}
 
+	resolvedDebug, err := resolveDiagDir(*debugDir, "debug")
+	if err != nil {
+		die("debug: %v", err)
+	}
+	resolvedProfile, err := resolveDiagDir(*profileDir, "profile")
+	if err != nil {
+		die("profile: %v", err)
+	}
+
+	stopProfile := startProfile(resolvedProfile)
+	defer stopProfile()
+
 	ctx := context.Background()
 	if err := app.Run(ctx, app.Options{
 		ProjectDir: cwd,
 		ProjectKey: key,
+		DebugDir:   resolvedDebug,
+		ProfileDir: resolvedProfile,
 	}); err != nil {
 		die("%v", err)
 	}
 }
 
+// resolveDiagDir turns the raw flag value into an absolute directory
+// path. Empty string disables the feature. The sentinel "auto" (set by
+// NoOptDefVal on bare flags) picks $XDG_STATE_HOME/patterm/<kind>/<ts>.
+// Any other value is treated as a literal path.
+func resolveDiagDir(raw, kind string) (string, error) {
+	if raw == "" {
+		return "", nil
+	}
+	if raw == "auto" {
+		base := os.Getenv("XDG_STATE_HOME")
+		if base == "" {
+			home, err := os.UserHomeDir()
+			if err != nil {
+				return "", err
+			}
+			base = filepath.Join(home, ".local", "state")
+		}
+		ts := time.Now().Format("20060102-150405")
+		return filepath.Join(base, "patterm", kind, ts), nil
+	}
+	return raw, nil
+}
+
+// startProfile begins a CPU profile under dir and returns a stop func
+// that writes heap + goroutine snapshots before flushing the CPU file.
+// Returns a no-op stop func when dir is empty. All diagnostics are
+// written to <dir>/profile.log — never to stdout/stderr — so the TUI
+// stays uncluttered.
+func startProfile(dir string) func() {
+	if dir == "" {
+		return func() {}
+	}
+	if err := os.MkdirAll(dir, 0o700); err != nil {
+		return func() {}
+	}
+	logPath := filepath.Join(dir, "profile.log")
+	plog := func(format string, args ...any) {
+		f, err := os.OpenFile(logPath, os.O_CREATE|os.O_WRONLY|os.O_APPEND, 0o600)
+		if err != nil {
+			return
+		}
+		defer f.Close()
+		fmt.Fprintf(f, format+"\n", args...)
+	}
+	cpuPath := filepath.Join(dir, "cpu.pprof")
+	f, err := os.Create(cpuPath)
+	if err != nil {
+		plog("cpu open: %v", err)
+		return func() {}
+	}
+	if err := pprof.StartCPUProfile(f); err != nil {
+		plog("cpu start: %v", err)
+		_ = f.Close()
+		return func() {}
+	}
+	plog("profiling started at %s", time.Now().Format(time.RFC3339Nano))
+	return func() {
+		pprof.StopCPUProfile()
+		_ = f.Close()
+		// Heap and goroutine snapshots at exit. Heap captures
+		// steady-state allocation; goroutine catches stragglers
+		// that didn't get cleaned up.
+		runtime.GC()
+		if hf, err := os.Create(filepath.Join(dir, "heap.pprof")); err == nil {
+			_ = pprof.Lookup("heap").WriteTo(hf, 0)
+			_ = hf.Close()
+		}
+		if gf, err := os.Create(filepath.Join(dir, "goroutine.pprof")); err == nil {
+			_ = pprof.Lookup("goroutine").WriteTo(gf, 0)
+			_ = gf.Close()
+		}
+		plog("profiling stopped at %s", time.Now().Format(time.RFC3339Nano))
+	}
+}
+
 func runMCPProxy() {
 	var (
 		socket   = flag.String("socket", "", "path to the running patterm process's MCP socket")

internal/app/app.go

@@ -29,6 +29,17 @@ import (
 type Options struct {
 	ProjectDir string
 	ProjectKey string
+	// DebugDir, when non-empty, enables verbose debug logging to
+	// <DebugDir>/patterm.log and per-child raw PTY output capture to
+	// <DebugDir>/<child-id>.raw. The dir is created if missing. Events
+	// (spawn / exit / state change) land in <DebugDir>/events.jsonl.
+	DebugDir string
+	// ProfileDir, when non-empty, enables in-process performance
+	// counters. patterm writes a per-second JSONL snapshot stream to
+	// <ProfileDir>/metrics.jsonl, a final aggregate to metrics.json,
+	// and a human-readable summary.txt on shutdown. The pprof files
+	// written by --profile sit alongside these in the same dir.
+	ProfileDir string
 }
 
 const keyCtrlK byte = 0x0b
@@ -44,6 +55,10 @@ func Run(ctx context.Context, opts Options) error {
 	if err != nil {
 		return fmt.Errorf("app: load presets: %w", err)
 	}
+	appSettings, settingsPath, err := loadSettings()
+	if err != nil {
+		logf("settings load: %v", err)
+	}
 
 	// Ensure the per-project scratchpad dir exists so MCP and the UI
 	// can read/write into it. SPEC §3.
@@ -77,6 +92,22 @@ func Run(ctx context.Context, opts Options) error {
 
 	sess := NewSession(opts.ProjectDir, opts.ProjectKey)
 	defer sess.Shutdown()
+
+	// Debug capture: when --debug=<dir> is set, write a verbose log
+	// (patterm.log), per-child raw PTY output (<id>.raw), and a
+	// JSONL event stream (events.jsonl). Installed before the TUI
+	// listener so the very first OnChildSpawned / OnPTYOut event
+	// is captured.
+	if opts.DebugDir != "" {
+		dc, err := openDebugCapture(opts.DebugDir)
+		if err != nil {
+			return fmt.Errorf("app: debug capture: %w", err)
+		}
+		os.Setenv("PATTERM_DEBUG_LOG", dc.LogPath())
+		sess.Subscribe(dc)
+		defer dc.Close()
+		logf("debug capture enabled at %s", opts.DebugDir)
+	}
 	// Snapshot persisted processes BEFORE attaching the store: Spawn
 	// mints fresh ids, so the old records would otherwise linger
 	// alongside the new ones. Drop them up front; the restore loop
@@ -113,29 +144,61 @@ func Run(ctx context.Context, opts Options) error {
 	ctx, cancel := context.WithCancel(ctx)
 	defer cancel()
 
+	// Performance tracker — instrumented hot-path timings written to
+	// <ProfileDir>. nil when --profile is off, in which case every
+	// record*() call is a fast nil check.
+	metrics, err := newMetricsTracker(opts.ProfileDir)
+	if err != nil {
+		return fmt.Errorf("app: metrics tracker: %w", err)
+	}
+	if metrics != nil {
+		go metrics.run(ctx)
+		defer metrics.close()
+	}
+
 	// Per-session idle-detection classifier. One goroutine ticks every
 	// 250ms over every live child and updates IdleState. It stops when
 	// ctx is cancelled.
 	go sess.runClassifier(ctx)
 
 	st := &uiState{
-		sess:       sess,
-		presets:    presets,
-		launcher:   launcher,
-		pads:       pads,
-		chromeWake: make(chan struct{}, 1),
-		trust:      trustStore,
-		timers:     host.timers,
-		hostCols:   cols,
-		hostRows:   rows,
-		stdinTTY:   term.IsTerminal(int(os.Stdin.Fd())),
+		sess:         sess,
+		presets:      presets,
+		launcher:     launcher,
+		pads:         pads,
+		chromeWake:   make(chan struct{}, 1),
+		trust:        trustStore,
+		timers:       host.timers,
+		hostCols:     cols,
+		hostRows:     rows,
+		stdinTTY:     term.IsTerminal(int(os.Stdin.Fd())),
+		metrics:      metrics,
+		settings:     appSettings,
+		settingsPath: settingsPath,
+		ctx:          ctx,
 	}
+	st.summaries = newSummaryManager(sess, opts.ProjectDir, presets, func() autoSummarySettings {
+		st.settingsMu.Lock()
+		defer st.settingsMu.Unlock()
+		return st.settings.AutoSummary.clone()
+	}, func() {
+		st.markChromeDirty()
+		st.markSidebarDirty()
+	}, func(_ string, result summaryState) {
+		if result.Error != "" {
+			st.flashError(fmt.Sprintf("summary: %v", result.Error))
+			return
+		}
+		st.flashTransient("summary updated")
+	})
+	sess.SetMetrics(metrics)
 	host.attention = st
 	host.focus = st
 	host.prompter = st
 	host.scratch = st
 	st.lastExit.Store(-1)
 	sess.Subscribe(st)
+	go st.summaries.run(ctx)
 
 	st.enterScreen()
 	st.renderEmptyState()
@@ -248,11 +311,42 @@ func Run(ctx context.Context, opts Options) error {
 			case <-st.chromeWake:
 			case <-ticker.C:
 			}
-			if !st.chromeDirty.Swap(false) {
+			chromeChanged := st.chromeDirty.Swap(false)
+			sidebarChanged := st.sidebarDirty.Swap(false)
+			didWork := chromeChanged || sidebarChanged
+			st.metrics.recordTickerFire(didWork)
+			if !didWork {
 				continue
 			}
-			st.drawTabBar()
-			st.drawStatusLine()
+			if chromeChanged {
+				st.drawTabBar()
+				st.drawStatusLine()
+			}
+			if sidebarChanged {
+				st.drawSidebar()
+			}
+		}
+	}()
+
+	// Marquee ticker: while a focused sidebar row's name overflows the
+	// rail width, advance the pause-scroll-pause animation by marking
+	// the sidebar dirty every marqueeStep. The chrome ticker above does
+	// the actual repaint. When no row is animating, this is a single
+	// cheap wakeup with no work.
+	wg.Add(1)
+	go func() {
+		defer wg.Done()
+		ticker := time.NewTicker(marqueeStep)
+		defer ticker.Stop()
+		for {
+			select {
+			case <-ctx.Done():
+				return
+			case <-ticker.C:
+			}
+			if st.marquee.active() {
+				st.markSidebarDirty()
+			}
 		}
 	}()
 
@@ -326,7 +420,6 @@ type uiState struct {
 	// switch resets the offset cleanly.
 	padOffsetName string
 
-
 	// activeAgentID tracks which top-level agent tab "owns" the agent
 	// tree section of the sidebar. It only updates when focus lands on
 	// an agent (or one of its sub-agents), so the agent tree stays
@@ -355,6 +448,17 @@ type uiState struct {
 	hostCols, hostRows uint16
 	stdinTTY           bool
 
+	// metrics is the optional performance tracker. nil when --profile
+	// is off. Hot paths call metrics.recordX which is a fast nil
+	// check on the disabled path.
+	metrics *metricsTracker
+
+	settingsMu   sync.Mutex
+	settings     settings
+	settingsPath string
+	ctx          context.Context
+	summaries    *summaryManager
+
 	// chromeCacheMu guards the last-rendered byte cache for each chrome
 	// element. The tab bar, sidebar, and status line all repaint on
 	// many state changes and on every PTY chunk, but their content
@@ -372,7 +476,19 @@ type uiState struct {
 	// sensitive paths (owner flip, attention, trust, focus change)
 	// continue to call drawStatusLine / drawTabBar synchronously.
 	chromeDirty atomic.Bool
-	chromeWake  chan struct{}
+	// sidebarDirty defers sidebar repaints off the per-chunk hot path
+	// in the same way. A long claude session resume — where every PTY
+	// chunk scrolls the viewport — used to call drawSidebar()
+	// synchronously per chunk, which dominated the resume's wall time
+	// (hundreds of full-sidebar rebuilds for a frame that was almost
+	// always cache-equal).
+	sidebarDirty atomic.Bool
+	chromeWake   chan struct{}
+
+	// marquee animates the focused sidebar row's name when it overflows
+	// the rail width. The dedicated 150ms ticker below flips
+	// sidebarDirty while a row is animating; idle case is free.
+	marquee marqueeState
 
 	// padsCacheMu guards the cached scratchpad listing. The sidebar
 	// and palette/sidebar nav helpers read it on every chunk-driven
@@ -389,6 +505,33 @@ func (st *uiState) dbgf(format string, args ...any) {
 	logf(format, args...)
 }
 
+func (st *uiState) activeSummaryText(width int) string {
+	if width <= 0 || st.summaries == nil {
+		return ""
+	}
+	st.settingsMu.Lock()
+	enabled := st.settings.AutoSummary.Enabled
+	st.settingsMu.Unlock()
+	if !enabled {
+		return ""
+	}
+	st.mu.Lock()
+	active := st.activeAgentID
+	st.mu.Unlock()
+	if active == "" {
+		return ""
+	}
+	sum := st.summaries.Summary(active)
+	text := strings.TrimSpace(sum.Text)
+	if text == "" {
+		return ""
+	}
+	if visibleLen(text) > width {
+		text = clipRunes(text, width-1) + "…"
+	}
+	return text
+}
+
 // trustRequest is one outstanding SPEC §7 trust prompt: an agent tried
 // to spawn / start / restart against an untrusted command preset and
 // the host wants user confirmation before the next attempt succeeds.
@@ -414,15 +557,20 @@ func (st *uiState) focusProcess(processID string) {
 	if c == nil {
 		return
 	}
+	st.marquee.reset()
 	layout := st.layoutSnapshot()
+	onAlt := childIsOnAlt(c)
 	st.mu.Lock()
 	leavingPad := st.focusedPad != ""
 	st.focusedPad = ""
 	st.focusedID = c.ID
 	st.focusedName = c.DisplayName()
 	st.updateActiveAgentLocked(c)
-	st.renderer = newViewportRenderer(layout)
+	r := newViewportRenderer(layout)
+	r.SetChildOnAlt(onAlt)
+	st.renderer = r
 	st.mu.Unlock()
+	st.syncHostMouseForChild(onAlt)
 	// Wipe whatever the previous focus (PTY child or pad view) left in
 	// the viewport before painting the new child's snapshot.
 	if leavingPad {
@@ -434,6 +582,41 @@ func (st *uiState) focusProcess(processID string) {
 	st.drawStatusLine()
 }
 
+// childIsOnAlt reports whether the child's emulator is currently on
+// its alternate screen. Returns false if the emulator is gone or the
+// query fails.
+func childIsOnAlt(c *Child) bool {
+	if c == nil {
+		return false
+	}
+	em := c.Emulator()
+	if em == nil {
+		return false
+	}
+	sc, err := em.ActiveScreen()
+	if err != nil {
+		return false
+	}
+	return sc == vt.ScreenAlternate
+}
+
+// syncHostMouseForChild emits the host mouse-reporting toggle that
+// matches a newly-focused child's screen side. Primary-screen children
+// want host mouse armed so the wheel drives inline scrollback; alt-
+// screen children get host mouse disabled by default so click-and-drag
+// selection works. Alt-screen TUIs that need mouse (vim, ranger, etc.)
+// re-enable it themselves, and the viewport renderer forwards those
+// toggles back to the host.
+func (st *uiState) syncHostMouseForChild(onAlt bool) {
+	st.outMu.Lock()
+	defer st.outMu.Unlock()
+	if onAlt {
+		_, _ = os.Stdout.WriteString("\x1b[?1000l\x1b[?1006l")
+	} else {
+		_, _ = os.Stdout.WriteString("\x1b[?1000h\x1b[?1006h")
+	}
+}
+
 // focusScratchpad shifts focus to a scratchpad. The main viewport
 // renders the pad's text instead of any child PTY; PTY output for the
 // previously focused child is dropped until focus moves back to a
@@ -442,6 +625,7 @@ func (st *uiState) focusScratchpad(name string) {
 	if name == "" {
 		return
 	}
+	st.marquee.reset()
 	st.mu.Lock()
 	if st.padOffsetName != name {
 		st.padOffset = 0
@@ -485,6 +669,7 @@ func (st *uiState) restartFocusedCommand(processID string) {
 	if c == nil || c.Kind != KindCommand {
 		return
 	}
+	st.marquee.reset()
 	layout := st.layoutSnapshot()
 	renderer := newViewportRenderer(layout)
 	st.mu.Lock()
@@ -569,15 +754,39 @@ func (st *uiState) scratchpadsChanged() {
 	}
 }
 
-// OnChildSpawned auto-focuses the new child.
+// OnChildSpawned auto-focuses the new child when the spawn came from
+// the user (palette, persistence restore, or an external MCP client with
+// no resolved identity). When ParentID is set — meaning a patterm-managed
+// agent spawned this child via spawn_agent/spawn_process — focus stays
+// on whatever the user was watching; the new child is still surfaced in
+// the sidebar/tab bar so it's reachable via the palette or select_process.
 func (st *uiState) OnChildSpawned(c *Child) {
+	if st.summaries != nil {
+		st.summaries.RegisterChild(c)
+	}
+	if c.ParentID != "" {
+		st.mu.Lock()
+		if st.palette != nil {
+			st.palette.children = st.sess.Children()
+			st.palette.focused = st.focusedID
+			st.palette.rebuild()
+			st.renderPaletteLocked()
+		}
+		st.mu.Unlock()
+		st.drawTabBar()
+		st.drawSidebar()
+		return
+	}
+	st.marquee.reset()
 	layout := st.layoutSnapshot()
+	onAlt := childIsOnAlt(c)
 	st.mu.Lock()
 	st.focusedPad = ""
 	st.focusedID = c.ID
 	st.focusedName = c.DisplayName()
 	st.updateActiveAgentLocked(c)
 	renderer := newViewportRenderer(layout)
+	renderer.SetChildOnAlt(onAlt)
 	st.renderer = renderer
 	palOpen := st.palette != nil
 	if palOpen {
@@ -611,6 +820,7 @@ func (st *uiState) OnChildSpawned(c *Child) {
 		st.outMu.Unlock()
 	}
 
+	st.syncHostMouseForChild(onAlt)
 	st.moveToViewportOrigin()
 	st.drawTabBar()
 	st.drawSidebar()
@@ -628,7 +838,11 @@ func (st *uiState) OnChildStateChanged(string, IdleState) {
 // OnChildExited drops focus and shows the empty state if it was the
 // focused child.
 func (st *uiState) OnChildExited(c *Child) {
+	if st.summaries != nil {
+		st.summaries.UnregisterChild(c.ID)
+	}
 	st.lastExit.Store(int32(c.ExitCode()))
+	st.marquee.reset()
 	layout := st.layoutSnapshot()
 	renderEmpty := false
 	st.mu.Lock()
@@ -710,6 +924,13 @@ func (st *uiState) scheduleAutoRestart(c *Child) {
 // disabled only around the replay so long styled runs cannot wrap into
 // the right rail.
 func (st *uiState) OnPTYOut(childID string, chunk []byte) {
+	var entry time.Time
+	if st.metrics != nil {
+		entry = time.Now()
+	}
+	if st.summaries != nil {
+		st.summaries.ObserveOutput(childID)
+	}
 	layout := st.layoutSnapshot()
 	st.mu.Lock()
 	focus := st.focusedID
@@ -726,16 +947,31 @@ func (st *uiState) OnPTYOut(childID string, chunk []byte) {
 	}
 	st.mu.Unlock()
 	if palOpen || focus != childID || renderer == nil {
+		st.metrics.recordPTYOutDrop()
 		return
 	}
 	var out []byte
 	if forceRepaint {
+		var snapStart time.Time
+		if st.metrics != nil {
+			snapStart = time.Now()
+		}
 		out = st.renderFocusedSnapshot(childID, renderer, layout)
+		if st.metrics != nil {
+			st.metrics.recordSnapshot(time.Since(snapStart))
+		}
 		if len(out) == 0 {
 			return
 		}
 	} else {
+		var rstart time.Time
+		if st.metrics != nil {
+			rstart = time.Now()
+		}
 		out = renderer.Render(chunk)
+		if st.metrics != nil {
+			st.metrics.recordRender(time.Since(rstart))
+		}
 	}
 	// One write covers the autowrap-disable prelude, the chunk, and the
 	// autowrap-restore postlude — three syscalls collapsed into one
@@ -745,9 +981,16 @@ func (st *uiState) OnPTYOut(childID string, chunk []byte) {
 	wrapped = append(wrapped, "\x1b[?7l"...)
 	wrapped = append(wrapped, out...)
 	wrapped = append(wrapped, "\x1b[?7h"...)
+	var wstart time.Time
+	if st.metrics != nil {
+		wstart = time.Now()
+	}
 	st.outMu.Lock()
 	_, _ = os.Stdout.Write(wrapped)
 	st.outMu.Unlock()
+	if st.metrics != nil {
+		st.metrics.recordStdout(time.Since(wstart), len(wrapped))
+	}
 	// RI / IND / NEL / SU / SD / IL / DL and bottom-margin LF / VT / FF
 	// scroll content within the host's scroll region, which spans every
 	// column — so any of them drags the right-hand sidebar's session-tree
@@ -760,15 +1003,23 @@ func (st *uiState) OnPTYOut(childID string, chunk []byte) {
 		st.chromeCacheMu.Lock()
 		st.sidebarCache = ""
 		st.chromeCacheMu.Unlock()
-		// Scrolled chunks can clobber the sidebar columns; repaint
-		// synchronously so the gap fills before the next chunk lands.
-		st.drawSidebar()
+		// Defer the sidebar repaint to the chrome ticker. On a long
+		// session resume every PTY chunk scrolls, and a synchronous
+		// drawSidebar() per chunk dominates wall time even when the
+		// frame ends up cache-equal — the rebuild work is unconditional.
+		// The chrome ticker drains the dirty flag at ~60 Hz, so the
+		// visible gap a scrolled chunk can leave in the sidebar columns
+		// is bounded by one frame.
+		st.markSidebarDirty()
 	}
 	// Defer the tab bar + status line repaint to the chrome ticker.
 	// The cached frame already short-circuits the wire write, but
 	// avoiding the string build, FindChild, and locking on every
 	// chunk pulls steady-state CPU off the hot path.
 	st.markChromeDirty()
+	if st.metrics != nil {
+		st.metrics.recordPTYOut(time.Since(entry), len(chunk))
+	}
 }
 
 func (st *uiState) enterScreen() {
@@ -866,6 +1117,18 @@ func (st *uiState) markChromeDirty() {
 	}
 }
 
+// markSidebarDirty schedules a sidebar repaint on the next ticker
+// frame. Hot path — every scrolled PTY chunk lands here. Synchronous
+// repaints from latency-sensitive sites (spawn, exit, focus, state
+// change, trust) keep calling drawSidebar directly.
+func (st *uiState) markSidebarDirty() {
+	st.sidebarDirty.Store(true)
+	select {
+	case st.chromeWake <- struct{}{}:
+	default:
+	}
+}
+
 func (st *uiState) invalidateChromeCache() {
 	st.chromeCacheMu.Lock()
 	st.tabBarCache = ""
@@ -896,6 +1159,10 @@ func (st *uiState) renderPaletteLocked() {
 // attention ask. Right side: palette hint. The PTY child occupies
 // host_rows-1 rows so this row is exclusively ours.
 func (st *uiState) drawStatusLine() {
+	var entry time.Time
+	if st.metrics != nil {
+		entry = time.Now()
+	}
 	st.mu.Lock()
 	palOpen := st.palette != nil
 	focusID := st.focusedID
@@ -982,10 +1249,16 @@ func (st *uiState) drawStatusLine() {
 	st.chromeCacheMu.Lock()
 	if line == st.statusLineCache {
 		st.chromeCacheMu.Unlock()
+		if st.metrics != nil {
+			st.metrics.recordStatus(time.Since(entry), true)
+		}
 		return
 	}
 	st.statusLineCache = line
 	st.chromeCacheMu.Unlock()
+	if st.metrics != nil {
+		defer func() { st.metrics.recordStatus(time.Since(entry), false) }()
+	}
 
 	st.outMu.Lock()
 	defer st.outMu.Unlock()
@@ -1131,6 +1404,15 @@ func (st *uiState) processStdin(chunk []byte) {
 	}
 
 	forward := make([]byte, 0, len(chunk))
+
+	var pendingAction *paletteAction
+	var pendingNav navEntry
+	var pendingRestartID string
+	var pendingViewportDelta int
+	var pendingViewportBottom bool
+	var pendingPadStep int
+	var pendingPadExit bool
+
 	flushForward := func() {
 		if len(forward) == 0 {
 			return
@@ -1142,22 +1424,22 @@ func (st *uiState) processStdin(chunk []byte) {
 				// writes so claude / codex / opencode don't treat a
 				// "text\r" batch as a paste.
 				_ = c.InjectAsUser(forward)
+				if st.summaries != nil {
+					st.summaries.ObserveHumanInput(c.ID, forward)
+				}
 				if prev != OwnerUser {
 					go st.drawStatusLine()
 				}
+				// Auto-snap the emulator viewport to the live area
+				// on any forwarded keystroke. Without this, typing
+				// while scrolled into history leaves the cursor /
+				// echoed bytes off-screen below the visible region.
+				pendingViewportBottom = true
 			}
 		}
 		forward = forward[:0]
 	}
 
-	var pendingAction *paletteAction
-	var pendingNav navEntry
-	var pendingRestartID string
-	var pendingViewportDelta int
-	var pendingViewportBottom bool
-	var pendingPadStep int
-	var pendingPadExit bool
-
 	// childOnPrimary captures whether the focused child is on its primary
 	// screen at the start of this chunk. Wheel events on the primary
 	// screen scroll the emulator viewport (inline scrollback); on the
@@ -1547,7 +1829,10 @@ func (st *uiState) scrollFocusedViewportToBottom() {
 }
 
 func (st *uiState) openPaletteLocked() {
-	st.palette = newPalette(st.sess.Children(), st.focusedID, st.focusedPad, st.presets)
+	st.settingsMu.Lock()
+	appSettings := st.settings.clone()
+	st.settingsMu.Unlock()
+	st.palette = newPalette(st.sess.Children(), st.focusedID, st.focusedPad, st.presets, appSettings)
 	// Push a "no kitty flags" entry onto the host terminal's keyboard
 	// stack so palette input arrives in plain legacy form regardless of
 	// what the focused child pushed. Codex/ratatui enables kitty mode
@@ -1622,6 +1907,13 @@ func (st *uiState) closePalette(action paletteAction) {
 			st.flashError(fmt.Sprintf("spawn %s: %v", action.preset.Name, err))
 		}
 
+	case "spawn-terminal":
+		l := st.layoutSnapshot()
+		st.launcher.SetSize(l.childCols(), l.childRows())
+		if _, err := st.launcher.LaunchTerminal(nil, "terminal", "", "", nil); err != nil {
+			st.flashError(fmt.Sprintf("spawn terminal: %v", err))
+		}
+
 	case "spawn-process-submit":
 		if action.command == "" {
 			restoreView()
@@ -1713,9 +2005,85 @@ func (st *uiState) closePalette(action paletteAction) {
 
 	case "proc-restart":
 		st.handleProcRestart(action.childID)
+
+	case "settings-close":
+		st.applySettingsAction(action)
+		restoreView()
+		st.drawTabBar()
+		st.drawSidebar()
+		st.drawStatusLine()
+
+	case "settings-test":
+		st.applySettingsAction(action)
+		restoreView()
+		st.drawTabBar()
+		st.drawSidebar()
+		st.drawStatusLine()
+		go st.testSummarizer()
+
+	case "settings-run-now":
+		st.applySettingsAction(action)
+		restoreView()
+		st.drawTabBar()
+		st.drawSidebar()
+		st.drawStatusLine()
+		st.runSummaryNow()
 	}
 }
 
+func (st *uiState) applySettingsAction(action paletteAction) {
+	if action.settings == nil {
+		return
+	}
+	next := action.settings.clone()
+	st.settingsMu.Lock()
+	path := st.settingsPath
+	st.settingsMu.Unlock()
+	if err := saveSettings(path, next); err != nil {
+		st.flashError(fmt.Sprintf("save settings: %v", err))
+		return
+	}
+	st.settingsMu.Lock()
+	st.settings = next
+	st.settingsMu.Unlock()
+}
+
+func (st *uiState) testSummarizer() {
+	if st.summaries == nil {
+		return
+	}
+	base := st.ctx
+	if base == nil {
+		base = context.Background()
+	}
+	ctx, cancel := context.WithTimeout(base, summaryTimeout)
+	defer cancel()
+	if err := st.summaries.Test(ctx); err != nil {
+		st.flashError(fmt.Sprintf("summarizer test: %v", err))
+		return
+	}
+	st.flashTransient("summarizer test passed")
+}
+
+func (st *uiState) runSummaryNow() {
+	if st.summaries == nil {
+		return
+	}
+	st.mu.Lock()
+	active := st.activeAgentID
+	st.mu.Unlock()
+	if active == "" {
+		st.flashError("no active top-level agent to summarize")
+		return
+	}
+	ctx := st.ctx
+	if ctx == nil {
+		ctx = context.Background()
+	}
+	st.summaries.RunNow(ctx, active)
+	st.flashTransient("summary requested")
+}
+
 func (st *uiState) handlePadDelete(name string) {
 	if name == "" || st.pads == nil {
 		st.repaintFocused()
@@ -1900,8 +2268,17 @@ func (st *uiState) flashError(msg string) {
 	st.mu.Lock()
 	st.attentionText = msg
 	st.attentionAt = "" // shows on every focus until cleared
+	focusedPad := st.focusedPad
+	focusedID := st.focusedID
 	st.mu.Unlock()
-	st.renderEmptyState()
+	switch {
+	case focusedPad != "":
+		st.repaintFocusedPad()
+	case focusedID != "":
+		st.repaintFocused()
+	default:
+		st.renderEmptyState()
+	}
 	st.drawTabBar()
 	st.drawSidebar()
 	st.drawStatusLine()

internal/app/bench_test.go

@@ -0,0 +1,546 @@
+package app
+
+import (
+	"fmt"
+	"io"
+	"strings"
+	"testing"
+
+	"github.com/hjbdev/patterm/internal/vt"
+)
+
+// Benchmarks for patterm's hot paths. Run with:
+//
+//	go test -bench=. -benchmem ./internal/app/
+//
+// or target one:
+//
+//	go test -bench=BenchmarkViewportRenderer_PlainASCII -benchmem ./internal/app/
+//
+// The fixtures below model the three workloads we care about most:
+//
+//   - PlainASCII: long-running text output (claude streaming a code
+//     diff, codex outputting a tool result body). Fast-path territory.
+//   - StyledLines: SGR-heavy output (claude/codex chat history with
+//     coloured tokens). State-machine path.
+//   - RatatuiBurst: many short cursor-positioning / SGR transitions in
+//     a tight chunk, matching codex/ratatui's incremental diff
+//     updates.
+//   - SnapshotReplay: full styled-grid replay (focus switch).
+
+// buildPlainASCIIChunk returns a roughly N-byte chunk of pure
+// printable ASCII text with the occasional newline — the cheapest
+// workload, exercises the fast path in viewport_renderer.Render.
+func buildPlainASCIIChunk(n int) []byte {
+	var b strings.Builder
+	b.Grow(n)
+	line := "The quick brown fox jumps over the lazy dog 0123456789 "
+	for b.Len() < n {
+		b.WriteString(line)
+		if b.Len()%80 < len(line) {
+			b.WriteByte('\n')
+		}
+	}
+	return []byte(b.String()[:n])
+}
+
+// buildStyledLinesChunk simulates SGR-heavy output: every word wears
+// a colour, so the renderer breaks out of its fast path on every
+// escape sequence.
+func buildStyledLinesChunk(n int) []byte {
+	var b strings.Builder
+	b.Grow(n)
+	colours := []string{"31", "32", "33", "34", "35", "36"}
+	words := []string{"package", "func", "return", "import", "struct", "type", "const", "var"}
+	i := 0
+	for b.Len() < n {
+		fmt.Fprintf(&b, "\x1b[%sm%s\x1b[0m ", colours[i%len(colours)], words[i%len(words)])
+		if i%10 == 9 {
+			b.WriteByte('\n')
+		}
+		i++
+	}
+	return []byte(b.String()[:n])
+}
+
+// buildRatatuiBurst simulates a single ratatui-style diff frame:
+// CUP, SGR, a few chars, CUP, SGR, a few chars… for a viewport's
+// worth of cells.
+func buildRatatuiBurst(cells int) []byte {
+	var b strings.Builder
+	for i := 0; i < cells; i++ {
+		row := (i / 80) + 1
+		col := (i % 80) + 1
+		fmt.Fprintf(&b, "\x1b[%d;%dH\x1b[3%dm%c", row, col, i%8, byte('A'+(i%26)))
+	}
+	b.WriteString("\x1b[0m")
+	return []byte(b.String())
+}
+
+// BenchmarkViewportRenderer_PlainASCII drives a 16 KiB plain-text
+// chunk through Render once per iteration. Reports ns/op,
+// allocations, and B/op.
+func BenchmarkViewportRenderer_PlainASCII(b *testing.B) {
+	chunk := buildPlainASCIIChunk(16 * 1024)
+	b.SetBytes(int64(len(chunk)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		_ = vr.Render(chunk)
+	}
+}
+
+// BenchmarkViewportRenderer_StyledLines exercises the per-byte CSI
+// path on SGR-heavy output. Most claude/codex chat resume traffic
+// looks like this — coloured prose with frequent style toggles.
+func BenchmarkViewportRenderer_StyledLines(b *testing.B) {
+	chunk := buildStyledLinesChunk(16 * 1024)
+	b.SetBytes(int64(len(chunk)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		_ = vr.Render(chunk)
+	}
+}
+
+// BenchmarkViewportRenderer_RatatuiBurst measures the worst-case
+// cursor-shuffling workload: full-frame diff updates dominated by
+// CUP + SGR + single-char writes.
+func BenchmarkViewportRenderer_RatatuiBurst(b *testing.B) {
+	chunk := buildRatatuiBurst(80 * 24) // one screenful of cells
+	b.SetBytes(int64(len(chunk)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		_ = vr.Render(chunk)
+	}
+}
+
+// BenchmarkContainsOSC measures the OSC-gate fast path used by
+// pumpChild before deciding whether to fire the per-chunk Title()
+// CGO call. Inputs:
+//   - "hot": SGR-styled output without OSC — the common case for
+//     codex/ratatui. We want this near zero.
+//   - "cold": chunk with an OSC sequence in the middle.
+func BenchmarkContainsOSC_NoOSC(b *testing.B) {
+	chunk := buildStyledLinesChunk(8 * 1024)
+	b.SetBytes(int64(len(chunk)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		_ = containsOSC(chunk)
+	}
+}
+
+func BenchmarkContainsOSC_WithOSC(b *testing.B) {
+	chunk := append(buildStyledLinesChunk(8*1024), []byte("\x1b]0;new title\x07")...)
+	b.SetBytes(int64(len(chunk)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		_ = containsOSC(chunk)
+	}
+}
+
+// BenchmarkRendererThroughput_ReuseInstance approximates real
+// session behaviour: a single viewport renderer fed many chunks in
+// sequence, no per-iteration allocation. Reports a throughput
+// closer to the steady-state OnPTYOut path. Chunks are 4 KiB to
+// match typical PTY read sizes; the renderer is reset every
+// benchmark run.
+func BenchmarkRendererThroughput_ReuseInstance(b *testing.B) {
+	chunks := make([][]byte, 16)
+	for i := range chunks {
+		chunks[i] = buildStyledLinesChunk(4 * 1024)
+	}
+	totalBytes := 0
+	for _, c := range chunks {
+		totalBytes += len(c)
+	}
+	b.SetBytes(int64(totalBytes))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		for _, c := range chunks {
+			_ = vr.Render(c)
+		}
+	}
+}
+
+// Stress workloads — these model the worst things a real session
+// can throw at us. The headline target is "ASCII video": every cell
+// of an 80x40 viewport carries an SGR colour change and a printable
+// character, rendered as one chunk per frame. Real ASCII-video CLIs
+// (ascii-image-converter, asciinema-render, towel.blinkenlights, the
+// Bad Apple meme) hit patterm with exactly this pattern at 24-30 fps
+// for minutes at a time.
+//
+// We synthesise the workload rather than ship a captured corpus so
+// the benchmarks stay deterministic and the repo doesn't carry tens
+// of MiB of fixture data. The encoding is faithful to what those
+// tools actually emit.
+
+// buildASCIIVideoFrame builds a single full-viewport frame with
+// 8-colour SGR per cell (`\x1b[3Nm`). One frame ≈ 30 KiB for an
+// 80x40 viewport, which lines up with what ascii-video tools emit.
+func buildASCIIVideoFrame(cols, rows int) []byte {
+	var b strings.Builder
+	b.WriteString("\x1b[H") // home cursor before the frame starts
+	for r := 0; r < rows; r++ {
+		for c := 0; c < cols; c++ {
+			fmt.Fprintf(&b, "\x1b[3%dm%c", (r+c)%8, byte(' '+(r*c)%(0x7e-' ')))
+		}
+		b.WriteString("\x1b[0m\r\n")
+	}
+	return []byte(b.String())
+}
+
+// buildASCIIVideoFrameTrueColor builds the same frame but with
+// 24-bit RGB SGR (`\x1b[38;2;R;G;Bm`). Every cell is ~20 bytes of
+// escape + 1 byte glyph, so a frame is ≈ 70 KiB. This is what
+// chafa --colors=full and modern terminal video players emit, and
+// it's the heaviest SGR variant the renderer's CSI path sees.
+func buildASCIIVideoFrameTrueColor(cols, rows int) []byte {
+	var b strings.Builder
+	b.WriteString("\x1b[H")
+	for r := 0; r < rows; r++ {
+		for c := 0; c < cols; c++ {
+			rd := (r * 7) % 256
+			gd := (c * 11) % 256
+			bd := ((r + c) * 13) % 256
+			fmt.Fprintf(&b, "\x1b[38;2;%d;%d;%dm%c", rd, gd, bd, byte(' '+(r*c)%(0x7e-' ')))
+		}
+		b.WriteString("\x1b[0m\r\n")
+	}
+	return []byte(b.String())
+}
+
+// buildBadApplePattern builds the simplest possible ASCII video
+// frame: alternating black/white cells (the Bad Apple meme is
+// essentially a 1-bit silhouette video). This is the pattern that
+// stresses the SGR state-machine without exercising truecolor parse
+// — useful for isolating "is the cost in the colour parsing or in
+// the cell-by-cell switching?"
+func buildBadApplePattern(cols, rows int) []byte {
+	var b strings.Builder
+	b.WriteString("\x1b[H")
+	for r := 0; r < rows; r++ {
+		for c := 0; c < cols; c++ {
+			if (r+c)%2 == 0 {
+				b.WriteString("\x1b[37m█")
+			} else {
+				b.WriteString("\x1b[30m█")
+			}
+		}
+		b.WriteString("\x1b[0m\r\n")
+	}
+	return []byte(b.String())
+}
+
+// BenchmarkASCIIVideo_Frame_8Color renders a single full-screen
+// frame as one chunk. The headline number is MB/s — at 30 fps a
+// frame is one PTY chunk every ~33 ms, so this should comfortably
+// stay well under 1 ms.
+func BenchmarkASCIIVideo_Frame_8Color(b *testing.B) {
+	frame := buildASCIIVideoFrame(80, 40)
+	b.SetBytes(int64(len(frame)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		_ = vr.Render(frame)
+	}
+}
+
+// BenchmarkASCIIVideo_Frame_TrueColor renders a single truecolor
+// frame. ~70 KiB per frame. Compare this to the 8-colour number to
+// see how much extra cost the truecolor SGR parse imposes — the
+// `\x1b[38;2;R;G;Bm` form is the longest and most parameter-rich
+// CSI patterm sees in practice.
+func BenchmarkASCIIVideo_Frame_TrueColor(b *testing.B) {
+	frame := buildASCIIVideoFrameTrueColor(80, 40)
+	b.SetBytes(int64(len(frame)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		_ = vr.Render(frame)
+	}
+}
+
+// BenchmarkASCIIVideo_Frame_BadApple is the 1-bit pattern: simplest
+// SGR (two colours, alternating). Isolates the renderer's cell-by-
+// cell SGR cycling cost from the truecolor parse cost.
+func BenchmarkASCIIVideo_Frame_BadApple(b *testing.B) {
+	frame := buildBadApplePattern(80, 40)
+	b.SetBytes(int64(len(frame)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		_ = vr.Render(frame)
+	}
+}
+
+// runStreamBench is the shared body for the per-fps stream
+// benchmarks. It feeds a fixed frame N times through a single
+// renderer instance and reports µs/frame + an achievable-fps
+// ceiling alongside the standard ns/op + MB/s. The fps value in
+// the benchmark name is the *target* — the workload itself doesn't
+// rate-limit; we just decide how many frames make a benchmark op
+// (3 seconds' worth) so steady-state cost dominates warm-up.
+func runStreamBench(b *testing.B, frame []byte, fps int) {
+	frames := fps * 3 // 3 seconds at the target rate
+	totalBytes := int64(len(frame) * frames)
+	b.SetBytes(totalBytes)
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		for f := 0; f < frames; f++ {
+			_ = vr.Render(frame)
+		}
+	}
+	nsPerFrame := float64(b.Elapsed().Nanoseconds()) / float64(b.N*frames)
+	b.ReportMetric(nsPerFrame/1000.0, "µs/frame")
+	b.ReportMetric(1e9/nsPerFrame, "fps_ceiling")
+	// budget_pct = how much of the per-frame budget at the target
+	// rate we burn. Under 100 means we can hit the target; over
+	// means we can't.
+	budgetNs := 1e9 / float64(fps)
+	b.ReportMetric(nsPerFrame/budgetNs*100, "budget_pct")
+}
+
+// BenchmarkASCIIVideo_Stream_8Color_30fps / _60fps / _120fps reuse
+// one renderer across (3 × fps) frames. The headline numbers are
+// µs/frame, fps_ceiling (= 1e9 / ns/frame), and budget_pct (=
+// percent of the per-frame budget at the target rate we consume).
+//
+// 30 fps is the typical ASCII-video baseline (towel, chafa, Bad
+// Apple ports). 60 is the "smooth playback" target. 120 is a
+// future-proofing stress level matching modern high-refresh
+// terminals.
+func BenchmarkASCIIVideo_Stream_8Color_30fps(b *testing.B) {
+	runStreamBench(b, buildASCIIVideoFrame(80, 40), 30)
+}
+func BenchmarkASCIIVideo_Stream_8Color_60fps(b *testing.B) {
+	runStreamBench(b, buildASCIIVideoFrame(80, 40), 60)
+}
+func BenchmarkASCIIVideo_Stream_8Color_120fps(b *testing.B) {
+	runStreamBench(b, buildASCIIVideoFrame(80, 40), 120)
+}
+
+// BenchmarkASCIIVideo_Stream_TrueColor_* same set but with the
+// truecolor frames. Compare against the 8-colour numbers to see
+// what the longer `\x1b[38;2;R;G;Bm` parse costs us.
+func BenchmarkASCIIVideo_Stream_TrueColor_30fps(b *testing.B) {
+	runStreamBench(b, buildASCIIVideoFrameTrueColor(80, 40), 30)
+}
+func BenchmarkASCIIVideo_Stream_TrueColor_60fps(b *testing.B) {
+	runStreamBench(b, buildASCIIVideoFrameTrueColor(80, 40), 60)
+}
+func BenchmarkASCIIVideo_Stream_TrueColor_120fps(b *testing.B) {
+	runStreamBench(b, buildASCIIVideoFrameTrueColor(80, 40), 120)
+}
+
+// BenchmarkASCIIVideo_Stream_BadApple_* tracks the 1-bit alternating
+// pattern. Isolates per-cell SGR cycling cost from the truecolor
+// parse cost above — useful when reading the diff between the two
+// stream variants.
+func BenchmarkASCIIVideo_Stream_BadApple_30fps(b *testing.B) {
+	runStreamBench(b, buildBadApplePattern(80, 40), 30)
+}
+func BenchmarkASCIIVideo_Stream_BadApple_60fps(b *testing.B) {
+	runStreamBench(b, buildBadApplePattern(80, 40), 60)
+}
+func BenchmarkASCIIVideo_Stream_BadApple_120fps(b *testing.B) {
+	runStreamBench(b, buildBadApplePattern(80, 40), 120)
+}
+
+// BenchmarkEmulator_Write_8Color / _TrueColor isolate the
+// libghostty-vt CGO cost — same frames the Pipeline benchmarks use,
+// but feeding only the emulator. The delta between this and
+// BenchmarkASCIIVideo_Stream_… is the renderer's share; the rest
+// is libghostty-vt.
+func BenchmarkEmulator_Write_8Color_Frame(b *testing.B) {
+	frame := buildASCIIVideoFrame(80, 40)
+	b.SetBytes(int64(len(frame)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		em, err := vt.NewGhosttyEmulator(80, 40)
+		if err != nil {
+			b.Fatalf("emulator: %v", err)
+		}
+		if _, werr := em.Write(frame); werr != nil {
+			b.Fatalf("emulator.Write: %v", werr)
+		}
+		_ = em.Close()
+	}
+}
+
+func BenchmarkEmulator_Write_TrueColor_Frame(b *testing.B) {
+	frame := buildASCIIVideoFrameTrueColor(80, 40)
+	b.SetBytes(int64(len(frame)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		em, err := vt.NewGhosttyEmulator(80, 40)
+		if err != nil {
+			b.Fatalf("emulator: %v", err)
+		}
+		if _, werr := em.Write(frame); werr != nil {
+			b.Fatalf("emulator.Write: %v", werr)
+		}
+		_ = em.Close()
+	}
+}
+
+// BenchmarkEmulator_Write_Stream_120fps reuses one emulator across
+// 360 frames (3 sec × 120 fps). This is the cleanest measurement
+// of em.Write steady-state cost.
+func BenchmarkEmulator_Write_Stream_8Color_120fps(b *testing.B) {
+	frame := buildASCIIVideoFrame(80, 40)
+	const frames = 360
+	b.SetBytes(int64(len(frame) * frames))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		em, err := vt.NewGhosttyEmulator(80, 40)
+		if err != nil {
+			b.Fatalf("emulator: %v", err)
+		}
+		for f := 0; f < frames; f++ {
+			if _, werr := em.Write(frame); werr != nil {
+				b.Fatalf("emulator.Write: %v", werr)
+			}
+		}
+		_ = em.Close()
+	}
+	nsPerFrame := float64(b.Elapsed().Nanoseconds()) / float64(b.N*frames)
+	b.ReportMetric(nsPerFrame/1000.0, "µs/frame")
+	b.ReportMetric(1e9/nsPerFrame, "fps_ceiling")
+}
+
+func BenchmarkEmulator_Write_Stream_TrueColor_120fps(b *testing.B) {
+	frame := buildASCIIVideoFrameTrueColor(80, 40)
+	const frames = 360
+	b.SetBytes(int64(len(frame) * frames))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		em, err := vt.NewGhosttyEmulator(80, 40)
+		if err != nil {
+			b.Fatalf("emulator: %v", err)
+		}
+		for f := 0; f < frames; f++ {
+			if _, werr := em.Write(frame); werr != nil {
+				b.Fatalf("emulator.Write: %v", werr)
+			}
+		}
+		_ = em.Close()
+	}
+	nsPerFrame := float64(b.Elapsed().Nanoseconds()) / float64(b.N*frames)
+	b.ReportMetric(nsPerFrame/1000.0, "µs/frame")
+	b.ReportMetric(1e9/nsPerFrame, "fps_ceiling")
+}
+
+// runPipelineStreamBench includes the libghostty-vt emulator.Write
+// CGO call and a stdout write to io.Discard alongside the renderer
+// — i.e. everything OnPTYOut does in production except the host
+// terminal's own paint time (which patterm doesn't control). This
+// is the honest "can we hit N fps end-to-end?" measurement.
+func runPipelineStreamBench(b *testing.B, frame []byte, fps int) {
+	frames := fps * 3
+	totalBytes := int64(len(frame) * frames)
+	b.SetBytes(totalBytes)
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		em, err := vt.NewGhosttyEmulator(80, 40)
+		if err != nil {
+			b.Fatalf("emulator: %v", err)
+		}
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		for f := 0; f < frames; f++ {
+			if _, werr := em.Write(frame); werr != nil {
+				b.Fatalf("emulator.Write: %v", werr)
+			}
+			out := vr.Render(frame)
+			// Match OnPTYOut's autowrap prelude/postlude wrapping so
+			// the byte count is faithful.
+			_, _ = io.Discard.Write([]byte("\x1b[?7l"))
+			_, _ = io.Discard.Write(out)
+			_, _ = io.Discard.Write([]byte("\x1b[?7h"))
+		}
+		_ = em.Close()
+	}
+	nsPerFrame := float64(b.Elapsed().Nanoseconds()) / float64(b.N*frames)
+	b.ReportMetric(nsPerFrame/1000.0, "µs/frame")
+	b.ReportMetric(1e9/nsPerFrame, "fps_ceiling")
+	budgetNs := 1e9 / float64(fps)
+	b.ReportMetric(nsPerFrame/budgetNs*100, "budget_pct")
+}
+
+// BenchmarkPipeline_ASCIIVideo_* — the FULL OnPTYOut path
+// (emulator.Write CGO + viewport renderer + a stdout write to
+// io.Discard) running at 30/60/120 fps targets. These are the
+// numbers to trust when asking "can we sustain N fps?" The
+// renderer-only Stream benchmarks above isolate one stage and
+// understate the real cost.
+//
+// 120 fps is the explicit baseline: anything under 100% of the
+// per-frame budget here means we hit 120 fps with margin to spare.
+func BenchmarkPipeline_ASCIIVideo_8Color_30fps(b *testing.B) {
+	runPipelineStreamBench(b, buildASCIIVideoFrame(80, 40), 30)
+}
+func BenchmarkPipeline_ASCIIVideo_8Color_60fps(b *testing.B) {
+	runPipelineStreamBench(b, buildASCIIVideoFrame(80, 40), 60)
+}
+func BenchmarkPipeline_ASCIIVideo_8Color_120fps(b *testing.B) {
+	runPipelineStreamBench(b, buildASCIIVideoFrame(80, 40), 120)
+}
+
+func BenchmarkPipeline_ASCIIVideo_TrueColor_30fps(b *testing.B) {
+	runPipelineStreamBench(b, buildASCIIVideoFrameTrueColor(80, 40), 30)
+}
+func BenchmarkPipeline_ASCIIVideo_TrueColor_60fps(b *testing.B) {
+	runPipelineStreamBench(b, buildASCIIVideoFrameTrueColor(80, 40), 60)
+}
+func BenchmarkPipeline_ASCIIVideo_TrueColor_120fps(b *testing.B) {
+	runPipelineStreamBench(b, buildASCIIVideoFrameTrueColor(80, 40), 120)
+}
+
+// BenchmarkSessionResume_5MiBStyled simulates the user's
+// motivating case: claude resuming a long chat session and dumping
+// the whole history. 5 MiB of styled output as a single Render
+// call. Numbers here tell us how long the visible "scrolling
+// while resume loads" window will be.
+func BenchmarkSessionResume_5MiBStyled(b *testing.B) {
+	chunk := buildStyledLinesChunk(5 * 1024 * 1024)
+	b.SetBytes(int64(len(chunk)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		_ = vr.Render(chunk)
+	}
+}
+
+// BenchmarkSessionResume_5MiBPlain same as above but pure text.
+// Lower bound — what we'd hit if the resume content were styling-
+// free.
+func BenchmarkSessionResume_5MiBPlain(b *testing.B) {
+	chunk := buildPlainASCIIChunk(5 * 1024 * 1024)
+	b.SetBytes(int64(len(chunk)))
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		vr := newViewportRenderer(newTerminalLayout(120, 40))
+		_ = vr.Render(chunk)
+	}
+}

internal/app/debug.go

@@ -0,0 +1,155 @@
+package app
+
+import (
+	"encoding/json"
+	"fmt"
+	"os"
+	"path/filepath"
+	"sync"
+	"time"
+)
+
+// debugCapture implements ChildEventListener and writes structured
+// debug artefacts under a single directory:
+//
+//   - patterm.log    — the existing logf() stream
+//   - events.jsonl   — one JSON object per lifecycle event
+//   - <id>.raw       — raw PTY bytes for each child, by id+name
+//
+// The capture is installed only when --debug=<dir> is set, so default
+// runs pay nothing.
+type debugCapture struct {
+	dir     string
+	logPath string
+
+	mu      sync.Mutex
+	events  *os.File
+	rawByID map[string]*os.File
+}
+
+func openDebugCapture(dir string) (*debugCapture, error) {
+	if err := os.MkdirAll(dir, 0o700); err != nil {
+		return nil, err
+	}
+	logPath := filepath.Join(dir, "patterm.log")
+	// Truncate-style fresh log per run is friendlier for grep'ing one
+	// session. The existing logf opens O_APPEND though, so concurrent
+	// runs against the same dir would interleave — that's on the user.
+	if f, err := os.Create(logPath); err != nil {
+		return nil, err
+	} else {
+		_ = f.Close()
+	}
+	ev, err := os.Create(filepath.Join(dir, "events.jsonl"))
+	if err != nil {
+		return nil, err
+	}
+	dc := &debugCapture{
+		dir:     dir,
+		logPath: logPath,
+		events:  ev,
+		rawByID: make(map[string]*os.File),
+	}
+	dc.writeEvent("session_start", map[string]any{
+		"time": time.Now().Format(time.RFC3339Nano),
+		"pid":  os.Getpid(),
+	})
+	return dc, nil
+}
+
+func (d *debugCapture) LogPath() string { return d.logPath }
+
+func (d *debugCapture) Close() error {
+	d.mu.Lock()
+	defer d.mu.Unlock()
+	d.writeEventLocked("session_end", map[string]any{
+		"time": time.Now().Format(time.RFC3339Nano),
+	})
+	for _, f := range d.rawByID {
+		_ = f.Close()
+	}
+	d.rawByID = nil
+	if d.events != nil {
+		_ = d.events.Close()
+		d.events = nil
+	}
+	return nil
+}
+
+func (d *debugCapture) OnChildSpawned(c *Child) {
+	d.writeEvent("child_spawned", map[string]any{
+		"time":      time.Now().Format(time.RFC3339Nano),
+		"id":        c.ID,
+		"name":      c.Name,
+		"kind":      string(c.Kind),
+		"parent_id": c.ParentID,
+		"preset":    c.PresetRef,
+		"argv":      c.Argv,
+	})
+}
+
+func (d *debugCapture) OnChildExited(c *Child) {
+	d.writeEvent("child_exited", map[string]any{
+		"time":      time.Now().Format(time.RFC3339Nano),
+		"id":        c.ID,
+		"name":      c.Name,
+		"exit_code": c.ExitCode(),
+	})
+	d.mu.Lock()
+	defer d.mu.Unlock()
+	if f, ok := d.rawByID[c.ID]; ok {
+		_ = f.Close()
+		delete(d.rawByID, c.ID)
+	}
+}
+
+func (d *debugCapture) OnChildStateChanged(id string, state IdleState) {
+	d.writeEvent("child_state", map[string]any{
+		"time":  time.Now().Format(time.RFC3339Nano),
+		"id":    id,
+		"state": string(state),
+	})
+}
+
+func (d *debugCapture) OnPTYOut(childID string, chunk []byte) {
+	if len(chunk) == 0 {
+		return
+	}
+	d.mu.Lock()
+	defer d.mu.Unlock()
+	f, ok := d.rawByID[childID]
+	if !ok {
+		path := filepath.Join(d.dir, childID+".raw")
+		nf, err := os.Create(path)
+		if err != nil {
+			return
+		}
+		f = nf
+		d.rawByID[childID] = nf
+	}
+	// Listener contract: don't retain chunk past return. Writing now
+	// is fine; the slice's backing buffer is reused for the next read
+	// only after this listener chain completes.
+	_, _ = f.Write(chunk)
+}
+
+func (d *debugCapture) writeEvent(kind string, fields map[string]any) {
+	d.mu.Lock()
+	defer d.mu.Unlock()
+	d.writeEventLocked(kind, fields)
+}
+
+func (d *debugCapture) writeEventLocked(kind string, fields map[string]any) {
+	if d.events == nil {
+		return
+	}
+	if fields == nil {
+		fields = map[string]any{}
+	}
+	fields["event"] = kind
+	enc, err := json.Marshal(fields)
+	if err != nil {
+		return
+	}
+	_, _ = fmt.Fprintln(d.events, string(enc))
+}

internal/app/host.go

@@ -1111,7 +1111,7 @@ func helpFor(topic string) mcp.HelpResponse {
 	case "spawning":
 		return mcp.HelpResponse{
 			Topic:        "spawning",
-			Content:      "spawn_agent launches another vendor LLM CLI as a sub-agent (orchestrator only). spawn_process(kind: command, preset: …) starts a stored command; spawn_process(kind: terminal) opens a shell. Command presets need trust the first time — you'll get needs_trust until the human accepts. Whatever you spawn is yours to clean up — see help('lifecycle').",
+			Content:      "spawn_agent launches another vendor LLM CLI as a sub-agent (orchestrator only). spawn_process(kind: command, preset: …) starts a stored command; spawn_process(kind: terminal) opens a shell. Command presets need trust the first time — you'll get needs_trust until the human accepts. ANTI-PATTERNS: do not shell out to `claude` / `codex` / `opencode` (or any other agent CLI) yourself, and do not pipe JSON-RPC into patterm's Unix socket via perl / nc / socat / curl. Either path bypasses caller-identity and the new agent reads back as a stray top-level tab instead of your child — call spawn_agent through the MCP transport you were initialised on. Whatever you spawn is yours to clean up — see help('lifecycle').",
 			RelatedTools: []string{"spawn_agent", "spawn_process", "start_process", "restart_process", "close_process"},
 		}
 	case "lifecycle":
@@ -1134,9 +1134,10 @@ func helpFor(topic string) mcp.HelpResponse {
 		}
 	case "coordination":
 		return mcp.HelpResponse{
-			Topic:        "coordination",
-			Content:      "send_message tags the message with the caller's role (parent → [orchestrator], child → [sub-agent:<name>]). Siblings must route through their parent. request_human_attention raises a UI notification when you can't safely decide.",
-			RelatedTools: []string{"send_message", "request_human_attention"},
+			Topic: "coordination",
+			Content: "send_message tags the message with the caller's role (parent → [orchestrator], child → [sub-agent:<name>]). Siblings must route through their parent. request_human_attention raises a UI notification when you can't safely decide.\n\n" +
+				"Reply routing: a sub-agent's reply to your send_message lands in YOUR pane tagged `[sub-agent:<name>]`, not in the sub-agent's output. Anti-pattern: `wait_for_pattern(sub_agent, …)` to wait for a reply — the sub-agent is already idle, its output won't change, and the call spins to timeout. Pattern: send_message → timer_fire_when_idle_any([sub_agent_id], body=\"[system] sub-agent finished\") → when the timer fires, the reply is already queued as your next user turn (or visible via get_process_output on your own pane).",
+			RelatedTools: []string{"send_message", "request_human_attention", "timer_fire_when_idle_any", "timer_fire_when_idle_all"},
 		}
 	case "scratchpads":
 		return mcp.HelpResponse{
@@ -1161,9 +1162,14 @@ func helpFor(topic string) mcp.HelpResponse {
 		}
 	case "readiness":
 		return mcp.HelpResponse{
-			Topic:        "readiness",
-			Content:      "A pane is 'idle' once nothing has been written to its PTY for ~1s (SPEC §11). Treat idle as a signal to read, not a guarantee of completion. wait_for_pattern lets you wait on a known terminal marker for stronger evidence.",
-			RelatedTools: []string{"wait_for_pattern", "get_process_status"},
+			Topic: "readiness",
+			Content: "A pane is 'idle' once nothing has been written to its PTY for ~1s (SPEC §11). Treat idle as a signal to read, not a guarantee of completion.\n\n" +
+				"Waiting for a sub-agent's reply (canonical pattern):\n" +
+				"  1. send_message(sub_agent_id, request)\n" +
+				"  2. timer_fire_when_idle_any(watched=[sub_agent_id], body=\"[system] sub-agent done\")\n" +
+				"  3. When the timer fires you re-enter as a fresh user turn; the sub-agent's reply is already in your own pane tagged `[sub-agent:<name>]` (read via get_process_output on yourself if you need it explicitly).\n\n" +
+				"wait_for_pattern is for waiting on text a process emits in its OWN output (a shell prompt, a build's \"tests passed\" line). It does NOT see send_message replies, because those land in the caller's pane, not the target's — calling wait_for_pattern on a sub-agent to wait for its reply deadlocks until timeout.",
+			RelatedTools: []string{"wait_for_pattern", "get_process_status", "timer_fire_when_idle_any", "send_message"},
 		}
 	case "permissions":
 		return mcp.HelpResponse{

internal/app/layout_test.go

@@ -14,10 +14,10 @@ func TestTerminalLayoutWideUsesMainViewport(t *testing.T) {
 	if l.childCols() != 91 {
 		t.Fatalf("child cols: got %d want 91", l.childCols())
 	}
-	if l.childRows() != 37 {
-		t.Fatalf("child rows: got %d want 37", l.childRows())
+	if l.childRows() != 36 {
+		t.Fatalf("child rows: got %d want 36", l.childRows())
 	}
-	if l.mainTop != 3 || l.statusRow != 40 {
+	if l.mainTop != 4 || l.statusRow != 40 {
 		t.Fatalf("unexpected vertical chrome: mainTop=%d statusRow=%d", l.mainTop, l.statusRow)
 	}
 }
@@ -30,8 +30,8 @@ func TestTerminalLayoutNarrowHidesSidebar(t *testing.T) {
 	if l.childCols() != 38 {
 		t.Fatalf("child cols: got %d want 38", l.childCols())
 	}
-	if l.childRows() != 9 {
-		t.Fatalf("child rows: got %d want 9", l.childRows())
+	if l.childRows() != 8 {
+		t.Fatalf("child rows: got %d want 8", l.childRows())
 	}
 }
 
@@ -46,13 +46,13 @@ func TestSpawnSizingUsesViewportDimensions(t *testing.T) {
 	l := newTerminalLayout(120, 40)
 	launcher := NewLauncher(nil, "", l.childCols(), l.childRows())
 	cols, rows := launcher.size()
-	if cols != 91 || rows != 37 {
-		t.Fatalf("launcher size: got %dx%d want 91x37", cols, rows)
+	if cols != 91 || rows != 36 {
+		t.Fatalf("launcher size: got %dx%d want 91x36", cols, rows)
 	}
 
 	host := newToolHost(nil, nil, nil, preset.Set{}, nil, l.childCols(), l.childRows())
 	cols, rows = host.size()
-	if cols != 91 || rows != 37 {
-		t.Fatalf("tool host size: got %dx%d want 91x37", cols, rows)
+	if cols != 91 || rows != 36 {
+		t.Fatalf("tool host size: got %dx%d want 91x36", cols, rows)
 	}
 }

internal/app/marquee.go

@@ -0,0 +1,123 @@
+package app
+
+import (
+	"sync"
+	"time"
+)
+
+// Phase ordering of the marquee state machine: hold the head, scroll
+// one cell per marqueeStep until the tail is visible, hold the tail,
+// snap back to the head.
+const (
+	phaseHoldStart = iota
+	phaseScroll
+	phaseHoldEnd
+)
+
+const (
+	marqueeHoldStart = time.Second
+	marqueeStep      = 150 * time.Millisecond
+	marqueeHoldEnd   = time.Second
+)
+
+// marqueeState drives the focused sidebar row's pause-scroll-pause
+// animation. State is wall-clock anchored (since), not tick-count
+// anchored, so a missed tick yields a slightly later frame rather
+// than a skipped one.
+type marqueeState struct {
+	mu      sync.Mutex
+	id      string
+	nameLen int
+	budget  int
+	state   int
+	offset  int
+	since   time.Time
+}
+
+// step advances the state machine for the row identified by id with
+// the given visible name length (in runes) and column budget. It
+// returns the current scroll offset, whether the row is animating
+// (i.e. nameLen > budget), and how long until the next visual change.
+//
+// When id changes, or nameLen <= budget, the state machine resets to
+// phaseHoldStart with offset 0 anchored at now.
+func (m *marqueeState) step(id string, nameLen, budget int, now time.Time) (offset int, animating bool, nextWake time.Duration) {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+
+	if id != m.id || nameLen != m.nameLen || budget != m.budget {
+		m.id = id
+		m.nameLen = nameLen
+		m.budget = budget
+		m.state = phaseHoldStart
+		m.offset = 0
+		m.since = now
+	}
+
+	if nameLen <= budget || budget <= 0 {
+		return 0, false, 0
+	}
+
+	maxOffset := nameLen - budget
+
+	for {
+		elapsed := now.Sub(m.since)
+		switch m.state {
+		case phaseHoldStart:
+			if elapsed < marqueeHoldStart {
+				return 0, true, marqueeHoldStart - elapsed
+			}
+			m.state = phaseScroll
+			m.since = m.since.Add(marqueeHoldStart)
+			continue
+		case phaseScroll:
+			steps := int(elapsed / marqueeStep)
+			if steps >= maxOffset {
+				m.offset = maxOffset
+				m.state = phaseHoldEnd
+				m.since = m.since.Add(time.Duration(maxOffset) * marqueeStep)
+				continue
+			}
+			m.offset = steps
+			rem := marqueeStep - (elapsed % marqueeStep)
+			return m.offset, true, rem
+		case phaseHoldEnd:
+			if elapsed < marqueeHoldEnd {
+				return maxOffset, true, marqueeHoldEnd - elapsed
+			}
+			m.state = phaseHoldStart
+			m.offset = 0
+			m.since = m.since.Add(marqueeHoldEnd)
+			continue
+		default:
+			m.state = phaseHoldStart
+			m.offset = 0
+			m.since = now
+			return 0, true, marqueeHoldStart
+		}
+	}
+}
+
+// active reports whether the marquee currently has an overflowing row
+// to animate. The marquee ticker goroutine uses this to gate dirty
+// flag flips so an idle sidebar costs nothing.
+func (m *marqueeState) active() bool {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+	return m.id != "" && m.nameLen > m.budget && m.budget > 0
+}
+
+// reset clears all state, forcing the next step() call to start a
+// fresh phaseHoldStart. Call this when focus changes so the newly
+// focused row begins with a full head-hold instead of inheriting
+// whatever phase the previous focus was in.
+func (m *marqueeState) reset() {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+	m.id = ""
+	m.nameLen = 0
+	m.budget = 0
+	m.state = phaseHoldStart
+	m.offset = 0
+	m.since = time.Time{}
+}

internal/app/marquee_test.go

@@ -0,0 +1,161 @@
+package app
+
+import (
+	"testing"
+	"time"
+)
+
+func TestMarqueeStepFits(t *testing.T) {
+	var m marqueeState
+	now := time.Unix(0, 0)
+	off, animating, _ := m.step("a", 5, 10, now)
+	if animating {
+		t.Fatalf("expected no animation when name fits in budget")
+	}
+	if off != 0 {
+		t.Fatalf("expected offset 0, got %d", off)
+	}
+}
+
+func TestMarqueePhaseProgression(t *testing.T) {
+	var m marqueeState
+	// name 10 runes, budget 5 → maxOffset = 5.
+	const nameLen, budget = 10, 5
+	t0 := time.Unix(0, 0)
+
+	// At t0: phaseHoldStart, offset 0, animating.
+	off, anim, wake := m.step("row", nameLen, budget, t0)
+	if off != 0 || !anim || wake != marqueeHoldStart {
+		t.Fatalf("t0: off=%d anim=%v wake=%v", off, anim, wake)
+	}
+
+	// Just before hold expires: still offset 0.
+	off, anim, _ = m.step("row", nameLen, budget, t0.Add(marqueeHoldStart-time.Millisecond))
+	if off != 0 || !anim {
+		t.Fatalf("pre-expiry hold: off=%d anim=%v", off, anim)
+	}
+
+	// At hold expiry + 1 step: should have transitioned to scroll, offset 1.
+	off, anim, _ = m.step("row", nameLen, budget, t0.Add(marqueeHoldStart+marqueeStep))
+	if !anim || off != 1 {
+		t.Fatalf("first scroll step: off=%d anim=%v", off, anim)
+	}
+
+	// Mid-scroll: offset == 3.
+	off, _, _ = m.step("row", nameLen, budget, t0.Add(marqueeHoldStart+3*marqueeStep))
+	if off != 3 {
+		t.Fatalf("mid scroll: off=%d", off)
+	}
+
+	// Tail reached: offset == maxOffset == 5.
+	off, _, _ = m.step("row", nameLen, budget, t0.Add(marqueeHoldStart+5*marqueeStep+time.Millisecond))
+	if off != 5 {
+		t.Fatalf("tail: off=%d", off)
+	}
+
+	// Hold-end window still pegged at maxOffset.
+	off, _, _ = m.step("row", nameLen, budget, t0.Add(marqueeHoldStart+5*marqueeStep+marqueeHoldEnd/2))
+	if off != 5 {
+		t.Fatalf("hold-end mid: off=%d", off)
+	}
+
+	// After hold-end: snap back to offset 0.
+	off, _, _ = m.step("row", nameLen, budget, t0.Add(marqueeHoldStart+5*marqueeStep+marqueeHoldEnd+time.Millisecond))
+	if off != 0 {
+		t.Fatalf("snap back: off=%d", off)
+	}
+}
+
+func TestMarqueeIDChangeResets(t *testing.T) {
+	var m marqueeState
+	t0 := time.Unix(0, 0)
+	_, _, _ = m.step("a", 10, 5, t0)
+	// Advance well into scroll for row "a".
+	_, _, _ = m.step("a", 10, 5, t0.Add(marqueeHoldStart+3*marqueeStep))
+	// Now focus moves to "b": offset must reset to 0 and phase to hold-start.
+	off, anim, wake := m.step("b", 10, 5, t0.Add(marqueeHoldStart+3*marqueeStep))
+	if off != 0 || !anim || wake != marqueeHoldStart {
+		t.Fatalf("id reset: off=%d anim=%v wake=%v", off, anim, wake)
+	}
+}
+
+func TestMarqueeActive(t *testing.T) {
+	var m marqueeState
+	if m.active() {
+		t.Fatalf("fresh marquee should not be active")
+	}
+	_, _, _ = m.step("row", 10, 5, time.Unix(0, 0))
+	if !m.active() {
+		t.Fatalf("expected active after overflow step")
+	}
+	_, _, _ = m.step("row", 4, 5, time.Unix(0, 0))
+	if m.active() {
+		t.Fatalf("should not be active when name fits")
+	}
+}
+
+func TestMarqueeReset(t *testing.T) {
+	var m marqueeState
+	_, _, _ = m.step("row", 10, 5, time.Unix(0, 0))
+	m.reset()
+	if m.active() {
+		t.Fatalf("expected inactive after reset")
+	}
+	// After reset, stepping the same id starts fresh.
+	off, _, wake := m.step("row", 10, 5, time.Unix(5, 0))
+	if off != 0 || wake != marqueeHoldStart {
+		t.Fatalf("post-reset start: off=%d wake=%v", off, wake)
+	}
+}
+
+func TestFitName(t *testing.T) {
+	cases := []struct {
+		name, in string
+		budget   int
+		want     string
+	}{
+		{"fits", "abc", 5, "abc"},
+		{"exact", "abcde", 5, "abcde"},
+		{"truncate", "abcdef", 5, "abcd…"},
+		{"budget1", "abcdef", 1, "…"},
+		{"budget0", "abc", 0, ""},
+		{"unicode", "αβγδεζη", 4, "αβγ…"},
+	}
+	for _, c := range cases {
+		t.Run(c.name, func(t *testing.T) {
+			got := fitName(c.in, c.budget)
+			if got != c.want {
+				t.Fatalf("fitName(%q, %d) = %q want %q", c.in, c.budget, got, c.want)
+			}
+		})
+	}
+}
+
+func TestMarqueeWindow(t *testing.T) {
+	got := marqueeWindow("abcdefgh", 4, 2)
+	if got != "cdef" {
+		t.Fatalf("window = %q", got)
+	}
+	// Clamp end-of-string overflow.
+	got = marqueeWindow("abcdef", 4, 10)
+	if got != "cdef" {
+		t.Fatalf("clamped window = %q", got)
+	}
+}
+
+func TestClampVisible(t *testing.T) {
+	// Plain string longer than width.
+	if got := clampVisible("abcdef", 3); visibleLen(got) != 3 {
+		t.Fatalf("plain clamp visible = %d (%q)", visibleLen(got), got)
+	}
+	// Already-fitting string is unchanged.
+	if got := clampVisible("abc", 5); got != "abc" {
+		t.Fatalf("unchanged = %q", got)
+	}
+	// SGR-wrapped string: visible portion must be <= width.
+	in := "\x1b[1mhello\x1b[0m world"
+	got := clampVisible(in, 5)
+	if visibleLen(got) != 5 {
+		t.Fatalf("sgr clamp visible = %d (%q)", visibleLen(got), got)
+	}
+}

internal/app/metrics.go

@@ -0,0 +1,462 @@
+package app
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"os"
+	"path/filepath"
+	"sync/atomic"
+	"time"
+)
+
+// metricsTracker collects per-hot-path counters and timings. All
+// fields are atomic so callers can record from the per-PTY-chunk path
+// without taking a lock. Enabled only when --profile is set.
+//
+// Sampled rates ("X per second", "p99 latency") are not tracked here
+// directly — the snapshotter goroutine writes a row to metrics.jsonl
+// every second, and analysis tools compute rates from the deltas.
+// Aggregate totals are written to metrics.json on shutdown.
+type metricsTracker struct {
+	startedAt time.Time
+
+	// PTY chunk arrival → stdout write pipeline (per OnPTYOut call).
+	ptyChunks      atomic.Int64
+	ptyBytes       atomic.Int64
+	onPTYOutNs     atomic.Int64
+	onPTYOutMaxNs  atomic.Int64
+	onPTYOutDrops  atomic.Int64 // chunks for non-focused children — fast-path returns
+	stdoutWrites   atomic.Int64
+	stdoutBytes    atomic.Int64
+	stdoutNs       atomic.Int64
+	stdoutMaxNs    atomic.Int64
+
+	// Viewport renderer (state-machine over child PTY bytes).
+	renderCalls atomic.Int64
+	renderNs    atomic.Int64
+	renderMaxNs atomic.Int64
+
+	// CGO into libghostty-vt (counted from pumpChild).
+	emuWriteCalls atomic.Int64
+	emuWriteNs    atomic.Int64
+	emuWriteMaxNs atomic.Int64
+	emuTitleCalls atomic.Int64
+	emuTitleNs    atomic.Int64
+	emuTitleSkips atomic.Int64 // OSC-gate fast path — title poll skipped
+
+	// Chrome paint pipeline.
+	sidebarDraws     atomic.Int64
+	sidebarCacheHits atomic.Int64
+	sidebarNs        atomic.Int64
+	sidebarMaxNs     atomic.Int64
+
+	tabbarDraws     atomic.Int64
+	tabbarCacheHits atomic.Int64
+	tabbarNs        atomic.Int64
+
+	statusDraws     atomic.Int64
+	statusCacheHits atomic.Int64
+	statusNs        atomic.Int64
+
+	// Snapshot replay (focus / spawn / nudge).
+	snapshotReplays atomic.Int64
+	snapshotNs      atomic.Int64
+	snapshotMaxNs   atomic.Int64
+
+	// Chrome ticker — distinguishes useful work from idle wakeups.
+	tickerFires      atomic.Int64
+	tickerIdleFires  atomic.Int64 // nothing dirty when the ticker fired
+
+	// Output destination (set when enabled).
+	rowFile *os.File // metrics.jsonl
+	dir     string
+}
+
+// newMetricsTracker creates an enabled tracker writing to <dir>/.
+// Returns nil + nil err if dir is empty (feature off). Caller must
+// call tracker.run(ctx) in a goroutine and tracker.close() at exit.
+func newMetricsTracker(dir string) (*metricsTracker, error) {
+	if dir == "" {
+		return nil, nil
+	}
+	if err := os.MkdirAll(dir, 0o700); err != nil {
+		return nil, err
+	}
+	row, err := os.Create(filepath.Join(dir, "metrics.jsonl"))
+	if err != nil {
+		return nil, err
+	}
+	return &metricsTracker{
+		startedAt: time.Now(),
+		rowFile:   row,
+		dir:       dir,
+	}, nil
+}
+
+// observeMax updates dst to max(dst, v) using a CAS loop. Atomic max
+// isn't a hardware primitive on most CPUs; this is the standard idiom.
+// Spurious wakeups can race but the result settles at the true max.
+func observeMax(dst *atomic.Int64, v int64) {
+	for {
+		old := dst.Load()
+		if v <= old {
+			return
+		}
+		if dst.CompareAndSwap(old, v) {
+			return
+		}
+	}
+}
+
+// recordPTYOut is called once at the end of each OnPTYOut invocation.
+// `dur` is the full per-chunk wall time (renderer + stdout + chrome
+// signals); `bytes` is the chunk's byte count.
+func (m *metricsTracker) recordPTYOut(dur time.Duration, bytes int) {
+	if m == nil {
+		return
+	}
+	m.ptyChunks.Add(1)
+	m.ptyBytes.Add(int64(bytes))
+	ns := dur.Nanoseconds()
+	m.onPTYOutNs.Add(ns)
+	observeMax(&m.onPTYOutMaxNs, ns)
+}
+
+func (m *metricsTracker) recordPTYOutDrop() {
+	if m == nil {
+		return
+	}
+	m.onPTYOutDrops.Add(1)
+}
+
+func (m *metricsTracker) recordRender(dur time.Duration) {
+	if m == nil {
+		return
+	}
+	m.renderCalls.Add(1)
+	ns := dur.Nanoseconds()
+	m.renderNs.Add(ns)
+	observeMax(&m.renderMaxNs, ns)
+}
+
+func (m *metricsTracker) recordStdout(dur time.Duration, bytes int) {
+	if m == nil {
+		return
+	}
+	m.stdoutWrites.Add(1)
+	m.stdoutBytes.Add(int64(bytes))
+	ns := dur.Nanoseconds()
+	m.stdoutNs.Add(ns)
+	observeMax(&m.stdoutMaxNs, ns)
+}
+
+func (m *metricsTracker) recordEmuWrite(dur time.Duration) {
+	if m == nil {
+		return
+	}
+	m.emuWriteCalls.Add(1)
+	ns := dur.Nanoseconds()
+	m.emuWriteNs.Add(ns)
+	observeMax(&m.emuWriteMaxNs, ns)
+}
+
+func (m *metricsTracker) recordEmuTitle(dur time.Duration, skipped bool) {
+	if m == nil {
+		return
+	}
+	if skipped {
+		m.emuTitleSkips.Add(1)
+		return
+	}
+	m.emuTitleCalls.Add(1)
+	m.emuTitleNs.Add(dur.Nanoseconds())
+}
+
+func (m *metricsTracker) recordSidebar(dur time.Duration, cacheHit bool) {
+	if m == nil {
+		return
+	}
+	m.sidebarDraws.Add(1)
+	if cacheHit {
+		m.sidebarCacheHits.Add(1)
+	}
+	ns := dur.Nanoseconds()
+	m.sidebarNs.Add(ns)
+	observeMax(&m.sidebarMaxNs, ns)
+}
+
+func (m *metricsTracker) recordTabbar(dur time.Duration, cacheHit bool) {
+	if m == nil {
+		return
+	}
+	m.tabbarDraws.Add(1)
+	if cacheHit {
+		m.tabbarCacheHits.Add(1)
+	}
+	m.tabbarNs.Add(dur.Nanoseconds())
+}
+
+func (m *metricsTracker) recordStatus(dur time.Duration, cacheHit bool) {
+	if m == nil {
+		return
+	}
+	m.statusDraws.Add(1)
+	if cacheHit {
+		m.statusCacheHits.Add(1)
+	}
+	m.statusNs.Add(dur.Nanoseconds())
+}
+
+func (m *metricsTracker) recordSnapshot(dur time.Duration) {
+	if m == nil {
+		return
+	}
+	m.snapshotReplays.Add(1)
+	ns := dur.Nanoseconds()
+	m.snapshotNs.Add(ns)
+	observeMax(&m.snapshotMaxNs, ns)
+}
+
+func (m *metricsTracker) recordTickerFire(didWork bool) {
+	if m == nil {
+		return
+	}
+	m.tickerFires.Add(1)
+	if !didWork {
+		m.tickerIdleFires.Add(1)
+	}
+}
+
+// snapshot captures the tracker's current state as a JSON-serialisable
+// map. Suitable for both the per-second JSONL row and the final
+// metrics.json aggregate.
+type metricsSnapshot struct {
+	WallSeconds   float64 `json:"wall_seconds"`
+	PTYChunks     int64   `json:"pty_chunks"`
+	PTYBytes      int64   `json:"pty_bytes"`
+	OnPTYOutNs    int64   `json:"on_pty_out_ns_total"`
+	OnPTYOutMaxNs int64   `json:"on_pty_out_ns_max"`
+	OnPTYOutDrops int64   `json:"on_pty_out_drops"`
+	StdoutWrites  int64   `json:"stdout_writes"`
+	StdoutBytes   int64   `json:"stdout_bytes"`
+	StdoutNs      int64   `json:"stdout_ns_total"`
+	StdoutMaxNs   int64   `json:"stdout_ns_max"`
+
+	RenderCalls int64 `json:"render_calls"`
+	RenderNs    int64 `json:"render_ns_total"`
+	RenderMaxNs int64 `json:"render_ns_max"`
+
+	EmuWriteCalls int64 `json:"emu_write_calls"`
+	EmuWriteNs    int64 `json:"emu_write_ns_total"`
+	EmuWriteMaxNs int64 `json:"emu_write_ns_max"`
+	EmuTitleCalls int64 `json:"emu_title_calls"`
+	EmuTitleNs    int64 `json:"emu_title_ns_total"`
+	EmuTitleSkips int64 `json:"emu_title_skips"`
+
+	SidebarDraws     int64 `json:"sidebar_draws"`
+	SidebarCacheHits int64 `json:"sidebar_cache_hits"`
+	SidebarNs        int64 `json:"sidebar_ns_total"`
+	SidebarMaxNs     int64 `json:"sidebar_ns_max"`
+
+	TabbarDraws     int64 `json:"tabbar_draws"`
+	TabbarCacheHits int64 `json:"tabbar_cache_hits"`
+	TabbarNs        int64 `json:"tabbar_ns_total"`
+
+	StatusDraws     int64 `json:"status_draws"`
+	StatusCacheHits int64 `json:"status_cache_hits"`
+	StatusNs        int64 `json:"status_ns_total"`
+
+	SnapshotReplays int64 `json:"snapshot_replays"`
+	SnapshotNs      int64 `json:"snapshot_ns_total"`
+	SnapshotMaxNs   int64 `json:"snapshot_ns_max"`
+
+	TickerFires     int64 `json:"ticker_fires"`
+	TickerIdleFires int64 `json:"ticker_idle_fires"`
+
+	// Derived rates (computed at snapshot time so consumers don't have
+	// to). All "per_second" values are averaged over wall_seconds.
+	PTYChunksPerSec      float64 `json:"pty_chunks_per_sec"`
+	PTYBytesPerSec       float64 `json:"pty_bytes_per_sec"`
+	OnPTYOutMeanUs       float64 `json:"on_pty_out_mean_us"`
+	StdoutMeanUs         float64 `json:"stdout_mean_us"`
+	EmuWriteMeanUs       float64 `json:"emu_write_mean_us"`
+	SidebarMeanUs        float64 `json:"sidebar_mean_us"`
+	SidebarCacheHitRate  float64 `json:"sidebar_cache_hit_rate"`
+	TabbarCacheHitRate   float64 `json:"tabbar_cache_hit_rate"`
+	StatusCacheHitRate   float64 `json:"status_cache_hit_rate"`
+	EmuTitleSkipRate     float64 `json:"emu_title_skip_rate"`
+	TickerIdleRate       float64 `json:"ticker_idle_rate"`
+	Timestamp            string  `json:"timestamp"`
+}
+
+func (m *metricsTracker) snapshotNow() metricsSnapshot {
+	wall := time.Since(m.startedAt).Seconds()
+	if wall <= 0 {
+		wall = 1
+	}
+	chunks := m.ptyChunks.Load()
+	bytes := m.ptyBytes.Load()
+	onptyTotal := m.onPTYOutNs.Load()
+	stdW := m.stdoutWrites.Load()
+	stdNs := m.stdoutNs.Load()
+	emuW := m.emuWriteCalls.Load()
+	emuWNs := m.emuWriteNs.Load()
+	sbDraws := m.sidebarDraws.Load()
+	sbHits := m.sidebarCacheHits.Load()
+	sbNs := m.sidebarNs.Load()
+	tbDraws := m.tabbarDraws.Load()
+	tbHits := m.tabbarCacheHits.Load()
+	stDraws := m.statusDraws.Load()
+	stHits := m.statusCacheHits.Load()
+	emuTC := m.emuTitleCalls.Load()
+	emuTS := m.emuTitleSkips.Load()
+	tickerF := m.tickerFires.Load()
+	tickerI := m.tickerIdleFires.Load()
+
+	div := func(num, denom int64) float64 {
+		if denom == 0 {
+			return 0
+		}
+		return float64(num) / float64(denom)
+	}
+
+	return metricsSnapshot{
+		WallSeconds:   wall,
+		PTYChunks:     chunks,
+		PTYBytes:      bytes,
+		OnPTYOutNs:    onptyTotal,
+		OnPTYOutMaxNs: m.onPTYOutMaxNs.Load(),
+		OnPTYOutDrops: m.onPTYOutDrops.Load(),
+		StdoutWrites:  stdW,
+		StdoutBytes:   m.stdoutBytes.Load(),
+		StdoutNs:      stdNs,
+		StdoutMaxNs:   m.stdoutMaxNs.Load(),
+
+		RenderCalls: m.renderCalls.Load(),
+		RenderNs:    m.renderNs.Load(),
+		RenderMaxNs: m.renderMaxNs.Load(),
+
+		EmuWriteCalls: emuW,
+		EmuWriteNs:    emuWNs,
+		EmuWriteMaxNs: m.emuWriteMaxNs.Load(),
+		EmuTitleCalls: emuTC,
+		EmuTitleNs:    m.emuTitleNs.Load(),
+		EmuTitleSkips: emuTS,
+
+		SidebarDraws:     sbDraws,
+		SidebarCacheHits: sbHits,
+		SidebarNs:        sbNs,
+		SidebarMaxNs:     m.sidebarMaxNs.Load(),
+
+		TabbarDraws:     tbDraws,
+		TabbarCacheHits: tbHits,
+		TabbarNs:        m.tabbarNs.Load(),
+
+		StatusDraws:     stDraws,
+		StatusCacheHits: stHits,
+		StatusNs:        m.statusNs.Load(),
+
+		SnapshotReplays: m.snapshotReplays.Load(),
+		SnapshotNs:      m.snapshotNs.Load(),
+		SnapshotMaxNs:   m.snapshotMaxNs.Load(),
+
+		TickerFires:     tickerF,
+		TickerIdleFires: tickerI,
+
+		PTYChunksPerSec:     float64(chunks) / wall,
+		PTYBytesPerSec:      float64(bytes) / wall,
+		OnPTYOutMeanUs:      div(onptyTotal/1000, chunks),
+		StdoutMeanUs:        div(stdNs/1000, stdW),
+		EmuWriteMeanUs:      div(emuWNs/1000, emuW),
+		SidebarMeanUs:       div(sbNs/1000, sbDraws),
+		SidebarCacheHitRate: div(sbHits, sbDraws),
+		TabbarCacheHitRate:  div(tbHits, tbDraws),
+		StatusCacheHitRate:  div(stHits, stDraws),
+		EmuTitleSkipRate:    div(emuTS, emuTC+emuTS),
+		TickerIdleRate:      div(tickerI, tickerF),
+		Timestamp:           time.Now().Format(time.RFC3339Nano),
+	}
+}
+
+// run is the snapshotter goroutine: write a JSONL row every second
+// until ctx is cancelled. Stops cleanly without flushing partial
+// rows.
+func (m *metricsTracker) run(ctx context.Context) {
+	if m == nil {
+		return
+	}
+	enc := json.NewEncoder(m.rowFile)
+	ticker := time.NewTicker(time.Second)
+	defer ticker.Stop()
+	for {
+		select {
+		case <-ctx.Done():
+			return
+		case <-ticker.C:
+			snap := m.snapshotNow()
+			_ = enc.Encode(snap)
+		}
+	}
+}
+
+// close writes the final aggregate snapshot to metrics.json + a
+// short human-readable summary.txt, then closes the row file. Safe
+// to call on a nil receiver.
+func (m *metricsTracker) close() {
+	if m == nil {
+		return
+	}
+	snap := m.snapshotNow()
+	if f, err := os.Create(filepath.Join(m.dir, "metrics.json")); err == nil {
+		enc := json.NewEncoder(f)
+		enc.SetIndent("", "  ")
+		_ = enc.Encode(snap)
+		_ = f.Close()
+	}
+	if f, err := os.Create(filepath.Join(m.dir, "summary.txt")); err == nil {
+		writeSummary(f, snap)
+		_ = f.Close()
+	}
+	if m.rowFile != nil {
+		_ = m.rowFile.Close()
+		m.rowFile = nil
+	}
+}
+
+// writeSummary renders a brief human-readable digest of a snapshot.
+// Designed for `cat summary.txt` after a session — quick orientation
+// before diving into metrics.json / pprof.
+func writeSummary(w *os.File, s metricsSnapshot) {
+	fmt.Fprintf(w, "patterm performance summary\n")
+	fmt.Fprintf(w, "===========================\n\n")
+	fmt.Fprintf(w, "session length:        %.1fs\n", s.WallSeconds)
+	fmt.Fprintf(w, "pty chunks:            %d  (%.1f /s)\n", s.PTYChunks, s.PTYChunksPerSec)
+	fmt.Fprintf(w, "pty bytes:             %d  (%.0f /s, %.1f KiB/s)\n",
+		s.PTYBytes, s.PTYBytesPerSec, s.PTYBytesPerSec/1024)
+	fmt.Fprintf(w, "pty chunks dropped:    %d  (focus not on caller — fast-path return)\n", s.OnPTYOutDrops)
+	fmt.Fprintf(w, "\n")
+	fmt.Fprintf(w, "OnPTYOut mean:         %.1fµs   max: %.1fms\n",
+		s.OnPTYOutMeanUs, float64(s.OnPTYOutMaxNs)/1e6)
+	fmt.Fprintf(w, "viewport.Render calls: %d  total %.1fms  max %.1fms\n",
+		s.RenderCalls, float64(s.RenderNs)/1e6, float64(s.RenderMaxNs)/1e6)
+	fmt.Fprintf(w, "stdout writes:         %d  mean %.1fµs  max %.1fms  bytes %d\n",
+		s.StdoutWrites, s.StdoutMeanUs, float64(s.StdoutMaxNs)/1e6, s.StdoutBytes)
+	fmt.Fprintf(w, "\n")
+	fmt.Fprintf(w, "emulator.Write (cgo):  %d  mean %.1fµs  max %.1fms\n",
+		s.EmuWriteCalls, s.EmuWriteMeanUs, float64(s.EmuWriteMaxNs)/1e6)
+	fmt.Fprintf(w, "emulator.Title polls:  %d real, %d gated   skip rate %.1f%%\n",
+		s.EmuTitleCalls, s.EmuTitleSkips, s.EmuTitleSkipRate*100)
+	fmt.Fprintf(w, "\n")
+	fmt.Fprintf(w, "sidebar draws:         %d  mean %.1fµs  max %.1fms  cache-hit %.1f%%\n",
+		s.SidebarDraws, s.SidebarMeanUs, float64(s.SidebarMaxNs)/1e6, s.SidebarCacheHitRate*100)
+	fmt.Fprintf(w, "tabbar draws:          %d  cache-hit %.1f%%\n",
+		s.TabbarDraws, s.TabbarCacheHitRate*100)
+	fmt.Fprintf(w, "status draws:          %d  cache-hit %.1f%%\n",
+		s.StatusDraws, s.StatusCacheHitRate*100)
+	fmt.Fprintf(w, "snapshot replays:      %d  total %.1fms  max %.1fms\n",
+		s.SnapshotReplays, float64(s.SnapshotNs)/1e6, float64(s.SnapshotMaxNs)/1e6)
+	fmt.Fprintf(w, "\n")
+	fmt.Fprintf(w, "chrome ticker:         %d fires, %d idle   idle rate %.1f%%\n",
+		s.TickerFires, s.TickerIdleFires, s.TickerIdleRate*100)
+}

internal/app/metrics_test.go

@@ -0,0 +1,116 @@
+package app
+
+import (
+	"encoding/json"
+	"os"
+	"path/filepath"
+	"testing"
+	"time"
+)
+
+func TestMetricsTrackerDisabledByEmptyDir(t *testing.T) {
+	m, err := newMetricsTracker("")
+	if err != nil {
+		t.Fatalf("newMetricsTracker(\"\") err: %v", err)
+	}
+	if m != nil {
+		t.Fatalf("expected nil tracker for empty dir, got %v", m)
+	}
+}
+
+func TestMetricsTrackerRecordsAndWrites(t *testing.T) {
+	dir := t.TempDir()
+	m, err := newMetricsTracker(dir)
+	if err != nil {
+		t.Fatalf("newMetricsTracker: %v", err)
+	}
+	if m == nil {
+		t.Fatal("expected non-nil tracker")
+	}
+
+	m.recordPTYOut(2*time.Millisecond, 1024)
+	m.recordPTYOut(5*time.Millisecond, 4096)
+	m.recordRender(800 * time.Microsecond)
+	m.recordStdout(300*time.Microsecond, 1100)
+	m.recordEmuWrite(150 * time.Microsecond)
+	m.recordEmuTitle(0, true)
+	m.recordEmuTitle(20*time.Microsecond, false)
+	m.recordSidebar(100*time.Microsecond, true)
+	m.recordSidebar(900*time.Microsecond, false)
+	m.recordTabbar(50*time.Microsecond, true)
+	m.recordStatus(40*time.Microsecond, true)
+	m.recordSnapshot(2 * time.Millisecond)
+	m.recordTickerFire(false)
+	m.recordTickerFire(true)
+	m.recordPTYOutDrop()
+
+	m.close()
+
+	// metrics.json should exist and parse, and reflect what we recorded.
+	raw, err := os.ReadFile(filepath.Join(dir, "metrics.json"))
+	if err != nil {
+		t.Fatalf("read metrics.json: %v", err)
+	}
+	var snap metricsSnapshot
+	if err := json.Unmarshal(raw, &snap); err != nil {
+		t.Fatalf("parse metrics.json: %v", err)
+	}
+	if snap.PTYChunks != 2 {
+		t.Errorf("PTYChunks = %d, want 2", snap.PTYChunks)
+	}
+	if snap.PTYBytes != 5120 {
+		t.Errorf("PTYBytes = %d, want 5120", snap.PTYBytes)
+	}
+	if snap.OnPTYOutMaxNs != (5 * time.Millisecond).Nanoseconds() {
+		t.Errorf("OnPTYOutMaxNs = %d, want %d",
+			snap.OnPTYOutMaxNs, (5 * time.Millisecond).Nanoseconds())
+	}
+	if snap.SidebarDraws != 2 {
+		t.Errorf("SidebarDraws = %d, want 2", snap.SidebarDraws)
+	}
+	if snap.SidebarCacheHits != 1 {
+		t.Errorf("SidebarCacheHits = %d, want 1", snap.SidebarCacheHits)
+	}
+	if snap.SidebarCacheHitRate != 0.5 {
+		t.Errorf("SidebarCacheHitRate = %v, want 0.5", snap.SidebarCacheHitRate)
+	}
+	if snap.EmuTitleCalls != 1 || snap.EmuTitleSkips != 1 {
+		t.Errorf("emu title accounting: calls=%d skips=%d, want 1/1",
+			snap.EmuTitleCalls, snap.EmuTitleSkips)
+	}
+	if snap.TickerFires != 2 || snap.TickerIdleFires != 1 {
+		t.Errorf("ticker accounting: fires=%d idle=%d, want 2/1",
+			snap.TickerFires, snap.TickerIdleFires)
+	}
+	if snap.OnPTYOutDrops != 1 {
+		t.Errorf("OnPTYOutDrops = %d, want 1", snap.OnPTYOutDrops)
+	}
+
+	// summary.txt should also be present and non-empty.
+	info, err := os.Stat(filepath.Join(dir, "summary.txt"))
+	if err != nil {
+		t.Fatalf("stat summary.txt: %v", err)
+	}
+	if info.Size() == 0 {
+		t.Fatal("summary.txt is empty")
+	}
+}
+
+func TestMetricsTrackerNilSafe(t *testing.T) {
+	// Every record* method must be safe to call on a nil receiver
+	// because the hot paths use that to avoid an enabled-check.
+	var m *metricsTracker
+	m.recordPTYOut(time.Millisecond, 100)
+	m.recordPTYOutDrop()
+	m.recordRender(time.Microsecond)
+	m.recordStdout(time.Microsecond, 50)
+	m.recordEmuWrite(time.Microsecond)
+	m.recordEmuTitle(time.Microsecond, false)
+	m.recordEmuTitle(0, true)
+	m.recordSidebar(time.Microsecond, true)
+	m.recordTabbar(time.Microsecond, false)
+	m.recordStatus(time.Microsecond, true)
+	m.recordSnapshot(time.Microsecond)
+	m.recordTickerFire(true)
+	m.close()
+}

internal/app/palette_context_test.go

@@ -31,8 +31,10 @@ func findItem(p *paletteState, want string) (int, *paletteItem) {
 
 func TestContextItemsScratchpad(t *testing.T) {
 	p := newPalette(nil, "", "notes.md", preset.Set{})
-	if i, _ := findItem(p, "pad-delete"); i != 0 {
-		t.Fatalf("pad-delete at %d; want top", i)
+	// pad-delete is the first selectable row; the Focused section header
+	// (a non-selectable row) sits above it.
+	if i, _ := findItem(p, "pad-delete"); i != 1 {
+		t.Fatalf("pad-delete at %d; want 1 (after Focused header)", i)
 	}
 	if _, it := findItem(p, "pad-rename-form"); it == nil || it.action.padName != "notes.md" {
 		t.Fatalf("pad-rename-form missing or wrong padName: %+v", it)

internal/app/palette_input_test.go

@@ -47,36 +47,50 @@ func TestPaletteBareEscCancels(t *testing.T) {
 	}
 }
 
+// firstSelectable returns the lowest item index whose action is
+// selectable (not a section header), or -1 if the palette has no
+// selectable rows.
+func firstSelectable(p *paletteState) int {
+	for i, it := range p.items {
+		if it.action.kind != "header" {
+			return i
+		}
+	}
+	return -1
+}
+
 func TestPaletteKittyArrowsNavigate(t *testing.T) {
 	pr := []*preset.Preset{{Name: "a"}, {Name: "b"}, {Name: "c"}}
 	p := newPalette(nil, "", "", preset.Set{Agents: pr})
-	if p.cursor != 0 {
-		t.Fatalf("initial cursor %d", p.cursor)
+	first := firstSelectable(p)
+	if first < 0 || p.cursor != first {
+		t.Fatalf("initial cursor %d, want first selectable %d", p.cursor, first)
 	}
 	// Kitty functional Down arrow.
 	_, _, adv := p.handleInput([]byte("\x1b[57353u"), 0)
 	if adv != 8 {
 		t.Fatalf("advance %d", adv)
 	}
-	if p.cursor != 1 {
-		t.Fatalf("cursor %d after Down, want 1", p.cursor)
+	if p.cursor != first+1 {
+		t.Fatalf("cursor %d after Down, want %d", p.cursor, first+1)
 	}
 	// Kitty functional Up arrow.
 	_, _, _ = p.handleInput([]byte("\x1b[57352u"), 0)
-	if p.cursor != 0 {
-		t.Fatalf("cursor %d after Up, want 0", p.cursor)
+	if p.cursor != first {
+		t.Fatalf("cursor %d after Up, want %d", p.cursor, first)
 	}
 }
 
 func TestPaletteLegacyArrowsStillWork(t *testing.T) {
 	pr := []*preset.Preset{{Name: "a"}, {Name: "b"}}
 	p := newPalette(nil, "", "", preset.Set{Agents: pr})
+	first := firstSelectable(p)
 	_, _, adv := p.handleInput([]byte("\x1b[B"), 0)
 	if adv != 3 {
 		t.Fatalf("advance %d", adv)
 	}
-	if p.cursor != 1 {
-		t.Fatalf("cursor %d, want 1", p.cursor)
+	if p.cursor != first+1 {
+		t.Fatalf("cursor %d, want %d", p.cursor, first+1)
 	}
 }
 

internal/app/palette_ux_test.go

@@ -0,0 +1,385 @@
+package app
+
+import (
+	"strings"
+	"testing"
+
+	"github.com/hjbdev/patterm/internal/preset"
+)
+
+// -- Phase 1: naming & dropped global Close list ---------------------
+
+func TestPaletteVerbsAreUnified(t *testing.T) {
+	procs := []*preset.Preset{{Name: "dev"}}
+	agents := []*preset.Preset{{Name: "claude"}}
+	p := newPalette(nil, "", "", preset.Set{Agents: agents, Processes: procs})
+	gotLabels := make([]string, 0, len(p.items))
+	for _, it := range p.items {
+		if it.action.kind == "header" {
+			continue
+		}
+		gotLabels = append(gotLabels, it.label)
+	}
+	joined := strings.Join(gotLabels, "\n")
+
+	mustContain := []string{
+		"Spawn agent: claude",
+		"Spawn process: dev",
+		"Spawn terminal",
+		"Spawn process… (custom)",
+	}
+	for _, want := range mustContain {
+		if !strings.Contains(joined, want) {
+			t.Errorf("missing unified-verb label %q in:\n%s", want, joined)
+		}
+	}
+	// The pre-overhaul verb forms must not appear anywhere.
+	mustNotContain := []string{"Run process:", "New Terminal", "Spawn process… (custom)"}
+	for _, bad := range mustNotContain {
+		if strings.Contains(joined, bad) {
+			t.Errorf("leftover legacy verb %q present in:\n%s", bad, joined)
+		}
+	}
+}
+
+func TestPaletteDropsGlobalCloseList(t *testing.T) {
+	c1 := makeFakeChild("a", "claude", KindAgent)
+	c2 := makeFakeChild("b", "dev", KindCommand)
+	p := newPalette([]*Child{c1, c2}, "", "", preset.Set{})
+	// No focus → no Focused context, so no "kill" / "agent-close" /
+	// "proc-stop" rows should exist at all.
+	for _, kind := range []string{"kill", "agent-close", "proc-stop", "proc-delete"} {
+		if i, _ := findItem(p, kind); i != -1 {
+			t.Fatalf("kind %q present at %d; global Close list should be gone", kind, i)
+		}
+	}
+}
+
+// -- Phase 2: section headers and cursor skip ------------------------
+
+func TestPaletteSectionHeadersPresent(t *testing.T) {
+	c := makeFakeChild("a", "claude", KindAgent)
+	p := newPalette([]*Child{c}, "a", "", preset.Set{Agents: []*preset.Preset{{Name: "codex"}}})
+	wantSections := []string{"Focused", "Open", "Spawn", "Quit"}
+	for _, w := range wantSections {
+		found := false
+		for _, it := range p.items {
+			if it.action.kind == "header" && strings.Contains(it.label, w) {
+				found = true
+				break
+			}
+		}
+		if !found {
+			t.Errorf("section header %q missing from items", w)
+		}
+	}
+}
+
+func TestPaletteCursorSkipsHeaders(t *testing.T) {
+	pr := []*preset.Preset{{Name: "a"}, {Name: "b"}}
+	p := newPalette(nil, "", "", preset.Set{Agents: pr})
+	// Initial cursor must land on a selectable row, never a header.
+	if p.items[p.cursor].action.kind == "header" {
+		t.Fatalf("initial cursor sits on a header: %+v", p.items[p.cursor])
+	}
+	// Walk to the end with cursorDown; every stop must be selectable.
+	for i := 0; i < len(p.items)*2; i++ {
+		p.cursorDown()
+		if p.items[p.cursor].action.kind == "header" {
+			t.Fatalf("cursorDown landed on a header at index %d", p.cursor)
+		}
+	}
+	// Walk back to top.
+	for i := 0; i < len(p.items)*2; i++ {
+		p.cursorUp()
+		if p.items[p.cursor].action.kind == "header" {
+			t.Fatalf("cursorUp landed on a header at index %d", p.cursor)
+		}
+	}
+}
+
+func TestPaletteEnterOnHeaderIsNoOp(t *testing.T) {
+	pr := []*preset.Preset{{Name: "a"}}
+	p := newPalette(nil, "", "", preset.Set{Agents: pr})
+	// Force the cursor onto a header.
+	for i, it := range p.items {
+		if it.action.kind == "header" {
+			p.cursor = i
+			break
+		}
+	}
+	_, done, _ := p.handleInput([]byte("\r"), 0)
+	if done {
+		t.Fatalf("Enter on header closed palette; expected no-op")
+	}
+}
+
+// -- Phase 3: filter chips & macro coexistence -----------------------
+
+func TestPaletteTabCyclesChip(t *testing.T) {
+	p := newTestPalette()
+	// All → Open
+	_, _, _ = p.handleInput([]byte{'\t'}, 0)
+	if string(p.query) != "sw " {
+		t.Fatalf("Tab #1: query %q, want %q", string(p.query), "sw ")
+	}
+	// Open → Spawn
+	_, _, _ = p.handleInput([]byte{'\t'}, 0)
+	if string(p.query) != "sp " {
+		t.Fatalf("Tab #2: query %q, want %q", string(p.query), "sp ")
+	}
+	// Spawn → Close
+	_, _, _ = p.handleInput([]byte{'\t'}, 0)
+	if string(p.query) != "k " {
+		t.Fatalf("Tab #3: query %q, want %q", string(p.query), "k ")
+	}
+	// Close → All (wraps)
+	_, _, _ = p.handleInput([]byte{'\t'}, 0)
+	if string(p.query) != "" {
+		t.Fatalf("Tab #4 wrap: query %q, want empty", string(p.query))
+	}
+}
+
+func TestPaletteShiftTabCyclesBackwards(t *testing.T) {
+	p := newTestPalette()
+	// Shift-Tab via legacy CSI Z: All → Close
+	_, _, _ = p.handleInput([]byte("\x1b[Z"), 0)
+	if string(p.query) != "k " {
+		t.Fatalf("Shift-Tab: query %q, want %q", string(p.query), "k ")
+	}
+}
+
+func TestPaletteBackspaceThroughTrailingMacro(t *testing.T) {
+	p := newTestPalette()
+	p.query = []rune("sw ")
+	p.rebuild()
+	p.backspace()
+	if string(p.query) != "" {
+		t.Fatalf("backspace through 'sw ' left %q; want empty", string(p.query))
+	}
+}
+
+func TestPaletteMacroPreservesQueryCase(t *testing.T) {
+	// Tab cycling shouldn't downcase the user-typed search text.
+	p := newTestPalette()
+	p.query = []rune("Foo")
+	p.rebuild()
+	_, _, _ = p.handleInput([]byte{'\t'}, 0)
+	if string(p.query) != "sw Foo" {
+		t.Fatalf("query after Tab over 'Foo' = %q; want 'sw Foo'", string(p.query))
+	}
+}
+
+// -- Phase 4: scored matching ----------------------------------------
+
+func TestFuzzyScorePrefixBeatsBoundaryBeatsSubstring(t *testing.T) {
+	prefix, _ := fuzzyScore("spawn agent: foo", "", "spa")
+	boundary, _ := fuzzyScore("hello spam", "", "spa")
+	substring, _ := fuzzyScore("escapade", "", "spa")
+	if !(prefix > boundary && boundary > substring) {
+		t.Fatalf("score ordering wrong: prefix=%d boundary=%d substring=%d", prefix, boundary, substring)
+	}
+}
+
+func TestFuzzyScoreReturnsMatchPositions(t *testing.T) {
+	_, pos := fuzzyScore("spawn process: dev", "", "dev")
+	want := []int{15, 16, 17}
+	if len(pos) != len(want) {
+		t.Fatalf("positions = %v, want %v", pos, want)
+	}
+	for i, p := range pos {
+		if p != want[i] {
+			t.Fatalf("pos[%d] = %d, want %d (full %v)", i, p, want[i], pos)
+		}
+	}
+}
+
+func TestPaletteScoredResultsDropHeaders(t *testing.T) {
+	pr := []*preset.Preset{{Name: "claude"}, {Name: "codex"}}
+	p := newPalette(nil, "", "", preset.Set{Agents: pr})
+	// Type a needle that matches both.
+	p.query = []rune("c")
+	p.rebuild()
+	for _, it := range p.items {
+		if it.action.kind == "header" {
+			t.Fatalf("scored mode should not emit header rows; got %+v", it)
+		}
+	}
+}
+
+func TestPaletteScoringFloatsPrefixMatchToTop(t *testing.T) {
+	// "x" is a prefix of "xtest" preset; it's a scattered-fuzzy match
+	// against many other rows. Scoring should land the prefix match at
+	// the top regardless of group order.
+	pr := []*preset.Preset{
+		{Name: "alpha"},
+		{Name: "xtest"},
+		{Name: "beta"},
+	}
+	p := newPalette(nil, "", "", preset.Set{Agents: pr})
+	p.query = []rune("xt")
+	p.rebuild()
+	if len(p.items) == 0 {
+		t.Fatalf("no scored items for needle 'xt'")
+	}
+	if !strings.Contains(p.items[0].label, "xtest") {
+		t.Fatalf("expected xtest at top of scored list, got %q", p.items[0].label)
+	}
+}
+
+// -- Phase 5: power-user accelerators --------------------------------
+
+func TestPaletteCtrlXOnSwitchKills(t *testing.T) {
+	c := makeFakeChild("a", "claude", KindAgent)
+	p := newPalette([]*Child{c}, "", "", preset.Set{})
+	// Cursor should already be on the switch row (it's the first
+	// selectable item with no Focused section).
+	idx, _ := findItem(p, "switch")
+	if idx < 0 {
+		t.Fatalf("no switch item in palette")
+	}
+	p.cursor = idx
+	action, done, _ := p.handleInput([]byte{0x18}, 0)
+	if !done {
+		t.Fatalf("Ctrl-X on switch row didn't close palette: action=%+v", action)
+	}
+	if action.kind != "kill" || action.childID != "a" {
+		t.Fatalf("Ctrl-X action = %+v, want kill of 'a'", action)
+	}
+}
+
+func TestPaletteCtrlXOnNonSwitchIsNoOp(t *testing.T) {
+	p := newPalette(nil, "", "", preset.Set{})
+	// Cursor parks on Quit or Spawn entries — neither is a switch row.
+	_, done, _ := p.handleInput([]byte{0x18}, 0)
+	if done {
+		t.Fatalf("Ctrl-X on non-switch closed palette")
+	}
+}
+
+func TestPaletteHelpToggle(t *testing.T) {
+	p := newTestPalette()
+	// `?` with empty query opens help.
+	_, done, _ := p.handleInput([]byte("?"), 0)
+	if done {
+		t.Fatalf("? closed palette")
+	}
+	if !p.showHelp {
+		t.Fatalf("? didn't open help")
+	}
+	// Next keystroke dismisses.
+	_, _, _ = p.handleInput([]byte("a"), 0)
+	if p.showHelp {
+		t.Fatalf("help still showing after dismissing keystroke")
+	}
+}
+
+func TestPaletteHelpDoesNotInterceptInQuery(t *testing.T) {
+	p := newTestPalette()
+	p.query = []rune("dev")
+	p.rebuild()
+	_, _, _ = p.handleInput([]byte("?"), 0)
+	if p.showHelp {
+		t.Fatalf("? with non-empty query incorrectly opened help")
+	}
+	if string(p.query) != "dev?" {
+		t.Fatalf("? with non-empty query failed to append: %q", string(p.query))
+	}
+}
+
+func TestPaletteHomeEndJumpsOverHeaders(t *testing.T) {
+	pr := []*preset.Preset{{Name: "a"}, {Name: "b"}}
+	p := newPalette(nil, "", "", preset.Set{Agents: pr})
+	// End jumps to last selectable.
+	p.cursorEnd()
+	if p.items[p.cursor].action.kind == "header" {
+		t.Fatalf("End landed on header: %+v", p.items[p.cursor])
+	}
+	if p.items[p.cursor].action.kind != "quit" {
+		t.Fatalf("End on simple palette should park on Quit; got %+v", p.items[p.cursor])
+	}
+	// Home returns to first selectable.
+	p.cursorHome()
+	if p.items[p.cursor].action.kind == "header" {
+		t.Fatalf("Home landed on header: %+v", p.items[p.cursor])
+	}
+}
+
+func TestPaletteAltDigitQuickPick(t *testing.T) {
+	pr := []*preset.Preset{{Name: "first"}, {Name: "second"}}
+	p := newPalette(nil, "", "", preset.Set{Agents: pr})
+	// Alt-1 picks the first selectable item (Spawn agent: first).
+	action, done, adv := p.handleInput([]byte("\x1b1"), 0)
+	if adv != 2 {
+		t.Fatalf("Alt-1 advance %d, want 2", adv)
+	}
+	if !done {
+		t.Fatalf("Alt-1 didn't close palette")
+	}
+	if action.kind != "spawn-agent" || action.preset == nil || action.preset.Name != "first" {
+		t.Fatalf("Alt-1 action = %+v, want spawn-agent first", action)
+	}
+}
+
+func TestAutoSummaryCadenceCyclesSoloValues(t *testing.T) {
+	p := newPalette(nil, "", "", preset.Set{}, defaultSettings())
+	p.mode = paletteModeAutoSummary
+	for i, row := range autoSummaryRows() {
+		if row.key == "cadence" {
+			p.cursor = i
+			break
+		}
+	}
+	if p.settings.AutoSummary.Cadence != "1m" {
+		t.Fatalf("initial cadence = %q", p.settings.AutoSummary.Cadence)
+	}
+	p.activateAutoSummaryRow()
+	if p.settings.AutoSummary.Cadence != "15s" {
+		t.Fatalf("first cycle cadence = %q", p.settings.AutoSummary.Cadence)
+	}
+	p.activateAutoSummaryRow()
+	if p.settings.AutoSummary.Cadence != "30s" {
+		t.Fatalf("second cycle cadence = %q", p.settings.AutoSummary.Cadence)
+	}
+	p.activateAutoSummaryRow()
+	if p.settings.AutoSummary.Cadence != "1m" {
+		t.Fatalf("third cycle cadence = %q", p.settings.AutoSummary.Cadence)
+	}
+}
+
+func TestPaletteFormCtrlRTogglesRelaunchFromCommandField(t *testing.T) {
+	p := newPalette(nil, "", "", preset.Set{})
+	p.mode = paletteModeSpawnForm
+	p.form = &spawnProcessForm{}
+	// Type without leaving the command field, then Ctrl-R.
+	for _, b := range []byte("xyz") {
+		_, _, _ = p.handleInput([]byte{b}, 0)
+	}
+	if p.form.field != 0 {
+		t.Fatalf("field jumped to %d", p.form.field)
+	}
+	_, _, _ = p.handleInput([]byte{0x12}, 0)
+	if !p.form.relaunch {
+		t.Fatalf("Ctrl-R didn't toggle relaunch from command field")
+	}
+	// Second press toggles back.
+	_, _, _ = p.handleInput([]byte{0x12}, 0)
+	if p.form.relaunch {
+		t.Fatalf("second Ctrl-R didn't toggle off")
+	}
+}
+
+// -- Phase 6: counter / scroll indicator -----------------------------
+
+func TestPaletteFooterCounter(t *testing.T) {
+	pr := []*preset.Preset{{Name: "a"}, {Name: "b"}, {Name: "c"}}
+	p := newPalette(nil, "", "", preset.Set{Agents: pr})
+	total := p.visibleSelectableCount()
+	if total < 4 { // 3 spawn-agents + terminal + custom + quit
+		t.Fatalf("expected ≥4 selectables; got %d", total)
+	}
+	idx := p.selectableIndex()
+	if idx <= 0 {
+		t.Fatalf("selectable index = %d on freshly-built palette; want ≥1", idx)
+	}
+}

internal/app/session.go

@@ -50,6 +50,11 @@ type Session struct {
 	// JSON file so they can be re-spawned after patterm restarts.
 	// Optional; nil means "no persistence" (used by unit tests).
 	persistStore *persist.Store
+
+	// metrics is the optional performance tracker. nil when --profile
+	// is off. The pump goroutine reads it via atomic Load so installing
+	// metrics post-construction doesn't race with running children.
+	metrics atomic.Pointer[metricsTracker]
 }
 
 // SetPersistStore attaches a process-persistence store. Future Spawn /
@@ -61,6 +66,18 @@ func (s *Session) SetPersistStore(p *persist.Store) {
 	s.mu.Unlock()
 }
 
+// SetMetrics installs the per-session performance tracker. Safe to
+// call with nil to disable (the default). Reads on the hot path go
+// through atomic.Pointer.Load() with no lock; SetMetrics swaps the
+// pointer once at startup.
+func (s *Session) SetMetrics(m *metricsTracker) {
+	s.metrics.Store(m)
+}
+
+func (s *Session) loadMetrics() *metricsTracker {
+	return s.metrics.Load()
+}
+
 // ChildEventListener is implemented by the TUI to react to lifecycle
 // events without polling.
 type ChildEventListener interface {
@@ -392,17 +409,37 @@ func (s *Session) pumpChild(c *Child, runID uint64) {
 			}
 			chunk := buf[:n]
 			if em := c.Emulator(); em != nil {
+				m := s.loadMetrics()
+				wstart := time.Time{}
+				if m != nil {
+					wstart = time.Now()
+				}
 				if _, werr := em.Write(chunk); werr != nil {
 					logf("emulator.Write(child %s): %v", c.ID, werr)
 				}
+				if m != nil {
+					m.recordEmuWrite(time.Since(wstart))
+				}
 				// OSC 0/2 title updates ride on the same byte stream as
 				// the rest of the output. Polling the emulator after each
-				// Write is cheap (one cgo call returning a borrowed
-				// string) and lets the classifier treat title changes as
-				// an activity signal — even when the title isn't visible
-				// in the rendered grid.
-				if t, terr := em.Title(); terr == nil {
-					c.recordTitle(t)
+				// chunk is cheap on its own (one CGO call) but codex/
+				// ratatui sends so many small chunks that the per-chunk
+				// CGO cost becomes measurable. Skip the Title poll when
+				// the chunk doesn't carry an OSC start byte at all; the
+				// title can only change on chunks that include one.
+				if containsOSC(chunk) {
+					tstart := time.Time{}
+					if m != nil {
+						tstart = time.Now()
+					}
+					if t, terr := em.Title(); terr == nil {
+						c.recordTitle(t)
+					}
+					if m != nil {
+						m.recordEmuTitle(time.Since(tstart), false)
+					}
+				} else if m != nil {
+					m.recordEmuTitle(0, true)
 				}
 			}
 			c.recordWrite(chunk)
@@ -433,6 +470,23 @@ func (s *Session) reapChild(c *Child, runID uint64) {
 	if !c.restarting.Load() {
 		c.cleanupOwnedPaths()
 	}
+	// Terminals are ephemeral: unlike command entries (kept around for
+	// restart_process) and agents (which the user clears via close_process
+	// once they're done with the corpse), an exited terminal has nothing
+	// useful left to do. Drop it from the session so it disappears from
+	// the Processes sidebar / switch list immediately.
+	if c.Kind == KindTerminal && !c.restarting.Load() {
+		c.teardownPTY()
+		s.mu.Lock()
+		delete(s.children, c.ID)
+		for i, oid := range s.order {
+			if oid == c.ID {
+				s.order = append(s.order[:i], s.order[i+1:]...)
+				break
+			}
+		}
+		s.mu.Unlock()
+	}
 }
 
 // killDescendantsOf terminates every still-live direct child of
@@ -662,6 +716,24 @@ func (s *Session) Shutdown() {
 	}
 }
 
+// containsOSC reports whether chunk holds a sequence that could begin
+// an OSC. OSC starts as ESC ] (0x1b 0x5d) or the bare C1 ] (0x9d),
+// so a chunk without either cannot have changed the emulator's OSC
+// title state. Used to short-circuit the per-chunk Title() poll from
+// pumpChild, which otherwise pays a CGO call for every chunk even
+// when codex/ratatui is just emitting SGR-styled output.
+func containsOSC(chunk []byte) bool {
+	for i, b := range chunk {
+		if b == 0x9d {
+			return true
+		}
+		if b == 0x1b && i+1 < len(chunk) && chunk[i+1] == ']' {
+			return true
+		}
+	}
+	return false
+}
+
 func logf(format string, args ...any) {
 	if os.Getenv("PATTERM_DEBUG_LOG") == "" {
 		return

internal/app/settings.go

@@ -0,0 +1,150 @@
+package app
+
+import (
+	"encoding/json"
+	"fmt"
+	"os"
+	"path/filepath"
+
+	"github.com/hjbdev/patterm/internal/preset"
+)
+
+const (
+	defaultSummaryProvider = "codex"
+	defaultCodexModel      = "gpt-5.4-mini"
+	defaultOpenCodeModel   = "opencode-go/minimax-m2.7"
+	defaultClaudeModel     = "claude-haiku-4-5"
+)
+
+type settings struct {
+	AutoSummary autoSummarySettings `json:"auto_summary"`
+}
+
+type autoSummarySettings struct {
+	Enabled         bool              `json:"enabled"`
+	Provider        string            `json:"provider"`
+	Models          map[string]string `json:"models"`
+	Cadence         string            `json:"cadence"`
+	QuietWindowMS   int               `json:"quiet_window_ms"`
+	MinInputChars   int               `json:"min_input_chars"`
+	MaxHistoryChars int               `json:"max_history_chars"`
+}
+
+func defaultSettings() settings {
+	return settings{
+		AutoSummary: autoSummarySettings{
+			Enabled:         true,
+			Provider:        defaultSummaryProvider,
+			Models:          defaultSummaryModels(),
+			Cadence:         "1m",
+			QuietWindowMS:   3000,
+			MinInputChars:   4,
+			MaxHistoryChars: 12000,
+		},
+	}
+}
+
+func defaultSummaryModels() map[string]string {
+	return map[string]string{
+		"codex":    defaultCodexModel,
+		"opencode": defaultOpenCodeModel,
+		"claude":   defaultClaudeModel,
+	}
+}
+
+func loadSettings() (settings, string, error) {
+	base, err := preset.ConfigDir()
+	if err != nil {
+		return settings{}, "", err
+	}
+	path := filepath.Join(base, "settings.json")
+	st := defaultSettings()
+	b, err := os.ReadFile(path)
+	if err != nil {
+		if os.IsNotExist(err) {
+			return st, path, nil
+		}
+		return st, path, fmt.Errorf("settings: read %s: %w", path, err)
+	}
+	if err := json.Unmarshal(b, &st); err != nil {
+		return defaultSettings(), path, fmt.Errorf("settings: parse %s: %w", path, err)
+	}
+	st.normalize()
+	return st, path, nil
+}
+
+func saveSettings(path string, st settings) error {
+	if path == "" {
+		return fmt.Errorf("settings: empty path")
+	}
+	st.normalize()
+	if err := os.MkdirAll(filepath.Dir(path), 0o700); err != nil {
+		return err
+	}
+	b, err := json.MarshalIndent(st, "", "  ")
+	if err != nil {
+		return err
+	}
+	b = append(b, '\n')
+	return os.WriteFile(path, b, 0o600)
+}
+
+func (st *settings) normalize() {
+	def := defaultSettings()
+	if st.AutoSummary.Provider == "" {
+		st.AutoSummary.Provider = def.AutoSummary.Provider
+	}
+	switch st.AutoSummary.Provider {
+	case "codex", "opencode", "claude":
+	default:
+		st.AutoSummary.Provider = def.AutoSummary.Provider
+	}
+	if st.AutoSummary.Models == nil {
+		st.AutoSummary.Models = defaultSummaryModels()
+	} else {
+		for k, v := range defaultSummaryModels() {
+			if st.AutoSummary.Models[k] == "" {
+				st.AutoSummary.Models[k] = v
+			}
+		}
+	}
+	if st.AutoSummary.Cadence == "" {
+		st.AutoSummary.Cadence = def.AutoSummary.Cadence
+	}
+	if st.AutoSummary.QuietWindowMS <= 0 {
+		st.AutoSummary.QuietWindowMS = def.AutoSummary.QuietWindowMS
+	}
+	if st.AutoSummary.MinInputChars <= 0 {
+		st.AutoSummary.MinInputChars = def.AutoSummary.MinInputChars
+	}
+	if st.AutoSummary.MaxHistoryChars <= 0 {
+		st.AutoSummary.MaxHistoryChars = def.AutoSummary.MaxHistoryChars
+	}
+}
+
+func (st settings) clone() settings {
+	st.normalize()
+	if st.AutoSummary.Models != nil {
+		models := make(map[string]string, len(st.AutoSummary.Models))
+		for k, v := range st.AutoSummary.Models {
+			models[k] = v
+		}
+		st.AutoSummary.Models = models
+	}
+	return st
+}
+
+func (a autoSummarySettings) clone() autoSummarySettings {
+	st := settings{AutoSummary: a}.clone()
+	return st.AutoSummary
+}
+
+func (a autoSummarySettings) modelFor(provider string) string {
+	if a.Models == nil {
+		return defaultSummaryModels()[provider]
+	}
+	if m := a.Models[provider]; m != "" {
+		return m
+	}
+	return defaultSummaryModels()[provider]
+}

internal/app/settings_test.go

@@ -0,0 +1,72 @@
+package app
+
+import (
+	"os"
+	"path/filepath"
+	"testing"
+)
+
+func TestLoadSettingsDefaults(t *testing.T) {
+	t.Setenv("XDG_CONFIG_HOME", t.TempDir())
+	st, path, err := loadSettings()
+	if err != nil {
+		t.Fatalf("loadSettings: %v", err)
+	}
+	if filepath.Base(path) != "settings.json" {
+		t.Fatalf("settings path = %q", path)
+	}
+	if !st.AutoSummary.Enabled {
+		t.Fatal("auto-summary should default enabled")
+	}
+	if st.AutoSummary.Provider != "codex" {
+		t.Fatalf("provider = %q want codex", st.AutoSummary.Provider)
+	}
+	if st.AutoSummary.Cadence != "1m" {
+		t.Fatalf("cadence = %q want 1m", st.AutoSummary.Cadence)
+	}
+	if got := st.AutoSummary.modelFor("codex"); got != "gpt-5.4-mini" {
+		t.Fatalf("codex model = %q", got)
+	}
+	if got := st.AutoSummary.modelFor("opencode"); got != "opencode-go/minimax-m2.7" {
+		t.Fatalf("opencode model = %q", got)
+	}
+}
+
+func TestSettingsCloneDoesNotShareModelMap(t *testing.T) {
+	st := defaultSettings()
+	cp := st.clone()
+	cp.AutoSummary.Models["codex"] = "changed"
+	if st.AutoSummary.Models["codex"] == "changed" {
+		t.Fatal("clone shared Models map with original")
+	}
+	a := st.AutoSummary.clone()
+	a.Models["opencode"] = "changed"
+	if st.AutoSummary.Models["opencode"] == "changed" {
+		t.Fatal("autoSummarySettings clone shared Models map with original")
+	}
+}
+
+func TestSaveAndLoadSettings(t *testing.T) {
+	dir := t.TempDir()
+	t.Setenv("XDG_CONFIG_HOME", dir)
+	st := defaultSettings()
+	st.AutoSummary.Provider = "opencode"
+	st.AutoSummary.Models["opencode"] = "minimax/test"
+	path := filepath.Join(dir, "patterm", "settings.json")
+	if err := saveSettings(path, st); err != nil {
+		t.Fatalf("saveSettings: %v", err)
+	}
+	if _, err := os.Stat(path); err != nil {
+		t.Fatalf("settings file missing: %v", err)
+	}
+	got, _, err := loadSettings()
+	if err != nil {
+		t.Fatalf("loadSettings: %v", err)
+	}
+	if got.AutoSummary.Provider != "opencode" {
+		t.Fatalf("provider = %q", got.AutoSummary.Provider)
+	}
+	if got.AutoSummary.modelFor("opencode") != "minimax/test" {
+		t.Fatalf("opencode model = %q", got.AutoSummary.modelFor("opencode"))
+	}
+}

internal/app/sidebar.go

@@ -12,6 +12,128 @@ const (
 	statusRows  = 1
 )
 
+// fitName returns name truncated to fit budget visible cells, with a
+// trailing "…" when it overflows. Operates on RAW (unstyled) input;
+// the caller wraps the result in SGR. Returns "" when budget <= 0.
+func fitName(name string, budget int) string {
+	if budget <= 0 {
+		return ""
+	}
+	runes := []rune(name)
+	if len(runes) <= budget {
+		return name
+	}
+	if budget == 1 {
+		return "…"
+	}
+	return string(runes[:budget-1]) + "…"
+}
+
+// marqueeWindow returns the window of name starting at offset, exactly
+// budget cells wide. Pre: caller has decided the name overflows budget
+// and offset is in [0, len([]rune(name))-budget]. Operates on RAW
+// (unstyled) input.
+func marqueeWindow(name string, budget, offset int) string {
+	if budget <= 0 {
+		return ""
+	}
+	runes := []rune(name)
+	if len(runes) <= budget {
+		return name
+	}
+	if offset < 0 {
+		offset = 0
+	}
+	end := offset + budget
+	if end > len(runes) {
+		end = len(runes)
+		offset = end - budget
+		if offset < 0 {
+			offset = 0
+		}
+	}
+	return string(runes[offset:end])
+}
+
+// clampVisible truncates s so that its visible (non-SGR) length is at
+// most width cells, preserving any active style by appending a reset.
+// Used as a defensive net by write() so a row whose decoration was
+// mis-sized still cannot spill past the sidebar band into the PTY area.
+func clampVisible(s string, width int) string {
+	if width <= 0 {
+		return ""
+	}
+	if visibleLen(s) <= width {
+		return s
+	}
+	var b strings.Builder
+	b.Grow(len(s))
+	visible := 0
+	inEsc := false
+	for _, r := range s {
+		if inEsc {
+			b.WriteRune(r)
+			if r == 'm' || r == 'H' {
+				inEsc = false
+			}
+			continue
+		}
+		if r == 0x1b {
+			inEsc = true
+			b.WriteRune(r)
+			continue
+		}
+		if visible >= width {
+			break
+		}
+		b.WriteRune(r)
+		visible++
+	}
+	b.WriteString(styleReset)
+	return b.String()
+}
+
+// chooseSidebarSuffix decides whether to keep or drop the trailing
+// timer indicator from a sidebar row's suffix. When the row's name
+// would have to ellipsise with the timer present, but the budget
+// freed by dropping the timer still leaves at least 6 cells for the
+// name, the timer is dropped. The name is the only identifier the
+// user has for that row; the timer is recoverable from the status
+// line and palette.
+func chooseSidebarSuffix(nameRuneLen, width int, prefix, suffix, timer string) (string, int) {
+	prefixCost := visibleLen(prefix)
+	budget := width - prefixCost - visibleLen(suffix)
+	if nameRuneLen <= budget || timer == "" {
+		return suffix, budget
+	}
+	slim := strings.TrimSuffix(suffix, timer)
+	if slim == suffix {
+		return suffix, budget
+	}
+	slimBudget := width - prefixCost - visibleLen(slim)
+	if slimBudget >= 6 {
+		return slim, slimBudget
+	}
+	return suffix, budget
+}
+
+// rowNameSlot returns the unstyled name cell for a sidebar row.
+// Unfocused (or focused-and-fitting) rows get fitName with a trailing
+// "…" on overflow. The focused row, when its name overflows the
+// budget, gets the current marquee window — exactly budget cells
+// wide so the surrounding row geometry stays put while it animates.
+func (st *uiState) rowNameSlot(id, rawName string, budget int, focused bool) string {
+	if budget <= 0 {
+		return ""
+	}
+	runes := []rune(rawName)
+	if !focused || len(runes) <= budget {
+		return fitName(rawName, budget)
+	}
+	off, _, _ := st.marquee.step(id, len(runes), budget, time.Now())
+	return marqueeWindow(rawName, budget, off)
+}
+
 // formatShortDuration renders a duration as a short, sidebar-friendly
 // suffix: ms under 1s, "12s" under 60s, "3m" otherwise.
 func formatShortDuration(d time.Duration) string {
@@ -38,6 +160,10 @@ func formatShortDuration(d time.Duration) string {
 // computed main viewport, so the sidebar region is outside the child's
 // cursor range. We can redraw freely without fighting the child for cells.
 func (st *uiState) drawSidebar() {
+	var entry time.Time
+	if st.metrics != nil {
+		entry = time.Now()
+	}
 	st.mu.Lock()
 	palOpen := st.palette != nil
 	focus := st.focusedID
@@ -69,6 +195,9 @@ func (st *uiState) drawSidebar() {
 		if row > maxRow {
 			return
 		}
+		if visibleLen(content) > width {
+			content = clampVisible(content, width)
+		}
 		pad := width - visibleLen(content)
 		if pad < 0 {
 			pad = 0
@@ -150,14 +279,19 @@ func (st *uiState) drawSidebar() {
 		if c.AutoRestart() {
 			marker = " " + styleDim + "⟳" + styleReset
 		}
-		var line string
+		timer := timerIndicator(c)
+		var prefix, openStyle string
 		if focused {
-			line = " " + styleAccent + "▎" + styleReset + " " + glyph + " " +
-				styleBold + c.DisplayName() + styleReset + marker + timerIndicator(c)
+			prefix = " " + styleAccent + "▎" + styleReset + " " + glyph + " "
+			openStyle = styleBold
 		} else {
-			line = "   " + glyph + " " + styleHint + c.DisplayName() + styleReset + marker + timerIndicator(c)
+			prefix = "   " + glyph + " "
+			openStyle = styleHint
 		}
-		write(line)
+		raw := c.DisplayName()
+		suffix, budget := chooseSidebarSuffix(len([]rune(raw)), width, prefix, marker+timer, timer)
+		nameCell := st.rowNameSlot(c.ID, raw, budget, focused)
+		write(prefix + openStyle + nameCell + styleReset + suffix)
 	}
 
 	// Agent Tree section — formerly "Session tree". Shows the active
@@ -182,14 +316,29 @@ func (st *uiState) drawSidebar() {
 		}
 		focused := c.ID == focus
 		glyph := statusGlyph(c, focused)
-		var line string
+		timer := timerIndicator(c)
+		var prefix, openStyle string
 		if focused {
-			line = " " + styleAccent + "▎" + styleReset + " " + indent + glyph + " " +
-				styleBold + c.DisplayName() + styleReset + timerIndicator(c)
+			prefix = " " + styleAccent + "▎" + styleReset + " " + indent + glyph + " "
+			openStyle = styleBold
 		} else {
-			line = "   " + indent + glyph + " " + styleHint + c.DisplayName() + styleReset + timerIndicator(c)
+			prefix = "   " + indent + glyph + " "
+			openStyle = styleHint
+		}
+		raw := c.DisplayName()
+		suffix, budget := chooseSidebarSuffix(len([]rune(raw)), width, prefix, timer, timer)
+		nameCell := st.rowNameSlot(c.ID, raw, budget, focused)
+		write(prefix + openStyle + nameCell + styleReset + suffix)
+	}
+
+	if summary := st.activeSummaryText(width - 4); summary != "" && row+2 <= maxRow {
+		write("")
+		for _, line := range wrapSidebarSummary(summary, width-4) {
+			if row > maxRow {
+				break
+			}
+			write("   " + styleDim + line + styleReset)
 		}
-		write(line)
 	}
 
 	// Scratchpads list — names only. The preview pane used to live
@@ -208,14 +357,18 @@ func (st *uiState) drawSidebar() {
 					if row > maxRow {
 						break
 					}
-					var line string
-					if e.Name == focusPad {
-						line = " " + styleAccent + "▎" + styleReset + " " +
-							styleBold + e.Name + styleReset
+					focused := e.Name == focusPad
+					var prefix, openStyle string
+					if focused {
+						prefix = " " + styleAccent + "▎" + styleReset + " "
+						openStyle = styleBold
 					} else {
-						line = "   " + styleHint + e.Name + styleReset
+						prefix = "   "
+						openStyle = styleHint
 					}
-					write(line)
+					budget := width - visibleLen(prefix)
+					nameCell := st.rowNameSlot("pad:"+e.Name, e.Name, budget, focused)
+					write(prefix + openStyle + nameCell + styleReset)
 				}
 			}
 		}
@@ -231,13 +384,58 @@ func (st *uiState) drawSidebar() {
 	st.chromeCacheMu.Lock()
 	if frame == st.sidebarCache {
 		st.chromeCacheMu.Unlock()
+		if st.metrics != nil {
+			st.metrics.recordSidebar(time.Since(entry), true)
+		}
 		return
 	}
 	st.sidebarCache = frame
 	st.chromeCacheMu.Unlock()
+	if st.metrics != nil {
+		defer func() { st.metrics.recordSidebar(time.Since(entry), false) }()
+	}
 
 	st.outMu.Lock()
 	// Save cursor; emit the sidebar; restore.
 	fmt.Fprintf(os.Stdout, "\x1b7%s\x1b8", frame)
 	st.outMu.Unlock()
 }
+
+func wrapSidebarSummary(s string, width int) []string {
+	if width < 1 {
+		width = 1
+	}
+	words := strings.Fields(s)
+	if len(words) == 0 {
+		return nil
+	}
+	var out []string
+	var cur string
+	for _, word := range words {
+		if visibleLen(word) > width {
+			if cur != "" {
+				out = append(out, cur)
+				cur = ""
+			}
+			out = append(out, clipRunes(word, width-1)+"…")
+			continue
+		}
+		if cur == "" {
+			cur = word
+			continue
+		}
+		if visibleLen(cur)+1+visibleLen(word) <= width {
+			cur += " " + word
+			continue
+		}
+		out = append(out, cur)
+		cur = word
+	}
+	if cur != "" {
+		out = append(out, cur)
+	}
+	if len(out) > 3 {
+		out = out[:3]
+	}
+	return out
+}

internal/app/spawn_focus_test.go

@@ -0,0 +1,46 @@
+package app
+
+import (
+	"testing"
+)
+
+// TestOnChildSpawnedAgentChildKeepsFocus verifies that when a child is
+// spawned with a ParentID set (i.e. a patterm-managed agent caused the
+// spawn over MCP), OnChildSpawned does NOT steal viewport focus from
+// the currently focused child.
+func TestOnChildSpawnedAgentChildKeepsFocus(t *testing.T) {
+	sess := NewSession(t.TempDir(), "test")
+	st := &uiState{sess: sess}
+
+	parent := newChildEntry("p_parent", "parent", KindAgent, nil, nil, "", "", "")
+	st.focusedID = parent.ID
+	st.focusedName = parent.Name
+
+	subAgent := newChildEntry("p_sub", "sub", KindAgent, nil, nil, parent.ID, "", "")
+
+	st.OnChildSpawned(subAgent)
+
+	if got := st.focusedID; got != parent.ID {
+		t.Fatalf("agent-initiated spawn should not change focusedID: want %q, got %q", parent.ID, got)
+	}
+	if got := st.focusedName; got != parent.Name {
+		t.Fatalf("focusedName changed: want %q, got %q", parent.Name, got)
+	}
+}
+
+// TestOnChildSpawnedPaletteChildTakesFocus verifies the legacy path is
+// preserved: spawns with an empty ParentID (palette, restore, external
+// MCP caller) still auto-focus the new child.
+func TestOnChildSpawnedPaletteChildTakesFocus(t *testing.T) {
+	sess := NewSession(t.TempDir(), "test")
+	st := &uiState{sess: sess}
+	st.lastExit.Store(-1)
+
+	c := newChildEntry("p_new", "newchild", KindAgent, nil, nil, "", "", "")
+
+	st.OnChildSpawned(c)
+
+	if got := st.focusedID; got != c.ID {
+		t.Fatalf("palette-initiated spawn should auto-focus: want %q, got %q", c.ID, got)
+	}
+}

internal/app/summarizer.go

@@ -0,0 +1,463 @@
+package app
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"fmt"
+	"os/exec"
+	"strings"
+	"sync"
+	"time"
+	"unicode"
+
+	"github.com/hjbdev/patterm/internal/preset"
+)
+
+const (
+	summaryTickInterval = time.Second
+	summaryTimeout      = 90 * time.Second
+	summaryMaxLineCells = 240
+)
+
+type summaryState struct {
+	Text      string
+	State     IdleState
+	UpdatedAt time.Time
+	Error     string
+}
+
+type summaryManager struct {
+	sess       *Session
+	projectDir string
+	presets    preset.Set
+	settings   func() autoSummarySettings
+	onUpdate   func()
+	onResult   func(string, summaryState)
+
+	mu      sync.Mutex
+	tracked map[string]bool
+	entries map[string]*summaryEntry
+}
+
+type summaryEntry struct {
+	armed          bool
+	dirty          bool
+	running        bool
+	lastInputAt    time.Time
+	lastOutputAt   time.Time
+	lastAttemptAt  time.Time
+	lastSummarized int64
+	state          summaryState
+}
+
+type summarizerResponse struct {
+	Summary string `json:"summary"`
+	State   string `json:"state"`
+}
+
+func newSummaryManager(sess *Session, projectDir string, presets preset.Set, settingsFn func() autoSummarySettings, onUpdate func(), onResult func(string, summaryState)) *summaryManager {
+	return &summaryManager{
+		sess:       sess,
+		projectDir: projectDir,
+		presets:    presets,
+		settings:   settingsFn,
+		onUpdate:   onUpdate,
+		onResult:   onResult,
+		tracked:    make(map[string]bool),
+		entries:    make(map[string]*summaryEntry),
+	}
+}
+
+func (m *summaryManager) run(ctx context.Context) {
+	ticker := time.NewTicker(summaryTickInterval)
+	defer ticker.Stop()
+	for {
+		select {
+		case <-ctx.Done():
+			return
+		case <-ticker.C:
+			m.maybeStart(ctx, time.Now())
+		}
+	}
+}
+
+func (m *summaryManager) ObserveHumanInput(childID string, b []byte) {
+	if m == nil || !m.isTracked(childID) {
+		return
+	}
+	cfg := m.settings()
+	if len(strings.TrimSpace(string(b))) < cfg.MinInputChars {
+		return
+	}
+	m.mu.Lock()
+	e := m.entryLocked(childID)
+	e.armed = true
+	e.lastInputAt = time.Now()
+	m.mu.Unlock()
+}
+
+func (m *summaryManager) ObserveOutput(childID string) {
+	if m == nil || !m.isTracked(childID) {
+		return
+	}
+	m.mu.Lock()
+	e := m.entryLocked(childID)
+	if e.armed {
+		e.dirty = true
+		e.lastOutputAt = time.Now()
+	}
+	m.mu.Unlock()
+}
+
+func (m *summaryManager) RegisterChild(c *Child) {
+	if m == nil || c == nil {
+		return
+	}
+	m.mu.Lock()
+	defer m.mu.Unlock()
+	if isTopLevelSummarizedAgent(c) {
+		m.tracked[c.ID] = true
+	} else {
+		delete(m.tracked, c.ID)
+	}
+}
+
+func (m *summaryManager) UnregisterChild(id string) {
+	if m == nil || id == "" {
+		return
+	}
+	m.mu.Lock()
+	defer m.mu.Unlock()
+	delete(m.tracked, id)
+}
+
+func (m *summaryManager) isTracked(id string) bool {
+	m.mu.Lock()
+	defer m.mu.Unlock()
+	return m.tracked[id]
+}
+
+func (m *summaryManager) Summary(childID string) summaryState {
+	if m == nil || childID == "" {
+		return summaryState{}
+	}
+	m.mu.Lock()
+	defer m.mu.Unlock()
+	if e := m.entries[childID]; e != nil {
+		return e.state
+	}
+	return summaryState{}
+}
+
+func (m *summaryManager) RunNow(ctx context.Context, childID string) {
+	if m == nil || childID == "" {
+		return
+	}
+	c := m.sess.FindChild(childID)
+	if !isTopLevelSummarizedAgent(c) {
+		return
+	}
+	m.mu.Lock()
+	e := m.entryLocked(c.ID)
+	if e.running {
+		m.mu.Unlock()
+		return
+	}
+	e.running = true
+	e.lastAttemptAt = time.Now()
+	m.mu.Unlock()
+	go m.runOne(ctx, c.ID, true)
+}
+
+func (m *summaryManager) Test(ctx context.Context) error {
+	cfg := m.settings()
+	return runSummarizerHealth(ctx, cfg, m.projectDir)
+}
+
+func (m *summaryManager) entryLocked(id string) *summaryEntry {
+	e := m.entries[id]
+	if e == nil {
+		e = &summaryEntry{}
+		m.entries[id] = e
+	}
+	return e
+}
+
+func (m *summaryManager) maybeStart(ctx context.Context, now time.Time) {
+	cfg := m.settings()
+	if !cfg.Enabled {
+		return
+	}
+	cadence, err := time.ParseDuration(cfg.Cadence)
+	if err != nil || cadence <= 0 {
+		cadence = time.Minute
+	}
+	quiet := time.Duration(cfg.QuietWindowMS) * time.Millisecond
+	var startID string
+	for _, c := range m.sess.Children() {
+		if !isTopLevelSummarizedAgent(c) {
+			continue
+		}
+		m.mu.Lock()
+		e := m.entryLocked(c.ID)
+		eligible := e.armed && e.dirty && !e.running &&
+			!e.lastOutputAt.IsZero() && now.Sub(e.lastOutputAt) >= quiet &&
+			(e.lastAttemptAt.IsZero() || now.Sub(e.lastAttemptAt) >= cadence) &&
+			c.ScreenVersion() != e.lastSummarized
+		if eligible {
+			e.running = true
+			e.lastAttemptAt = now
+			startID = c.ID
+		}
+		m.mu.Unlock()
+		if startID != "" {
+			go m.runOne(ctx, startID, false)
+			return
+		}
+	}
+}
+
+func (m *summaryManager) runOne(ctx context.Context, childID string, manual bool) {
+	c := m.sess.FindChild(childID)
+	if c == nil {
+		m.finish(childID, summaryState{Error: "process disappeared"}, 0)
+		return
+	}
+	cfg := m.settings()
+	snapshot := buildSummarySnapshot(c, cfg.MaxHistoryChars, m.chromeHintsFor(c.PresetRef))
+	if strings.TrimSpace(snapshot) == "" {
+		m.finish(childID, summaryState{Error: "empty snapshot"}, c.ScreenVersion())
+		return
+	}
+	runCtx, cancel := context.WithTimeout(ctx, summaryTimeout)
+	defer cancel()
+	resp, err := runSummarizer(runCtx, cfg, m.projectDir, snapshot)
+	st := summaryState{UpdatedAt: time.Now()}
+	if err != nil {
+		st.Error = err.Error()
+		m.finish(childID, st, c.ScreenVersion())
+		return
+	}
+	st.Text = strings.TrimSpace(resp.Summary)
+	st.State = summaryIdleState(resp.State)
+	if st.Text == "" {
+		st.Error = "empty summary"
+	}
+	if manual && st.Text != "" && st.State == StateUnknown {
+		st.State = c.IdleState()
+	}
+	m.finish(childID, st, c.ScreenVersion())
+}
+
+func (m *summaryManager) finish(childID string, st summaryState, version int64) {
+	m.mu.Lock()
+	e := m.entryLocked(childID)
+	e.running = false
+	if st.Text != "" || st.Error != "" {
+		if st.Text == "" && e.state.Text != "" {
+			st.Text = e.state.Text
+			st.State = e.state.State
+			st.UpdatedAt = e.state.UpdatedAt
+		}
+		e.state = st
+	}
+	if st.Text != "" {
+		e.armed = false
+		e.dirty = false
+		e.lastSummarized = version
+	}
+	m.mu.Unlock()
+	if m.onUpdate != nil {
+		m.onUpdate()
+	}
+	if m.onResult != nil && (st.Text != "" || st.Error != "") {
+		m.onResult(childID, st)
+	}
+}
+
+func isTopLevelSummarizedAgent(c *Child) bool {
+	return c != nil && c.Kind == KindAgent && c.ParentID == "" && c.Status() == StatusRunning
+}
+
+func (m *summaryManager) chromeHintsFor(presetName string) []string {
+	if presetName == "" {
+		return nil
+	}
+	for _, p := range m.presets.Agents {
+		if p.Name == presetName {
+			return p.ChromeTrimHints
+		}
+	}
+	return nil
+}
+
+func buildSummarySnapshot(c *Child, maxChars int, chromeHints []string) string {
+	if maxChars <= 0 {
+		maxChars = 12000
+	}
+	grid := ""
+	if em := c.Emulator(); em != nil {
+		if txt, err := em.PlainText(); err == nil {
+			grid = compactSummaryText(applyChromeTrim(txt, chromeHints))
+		}
+	}
+	tailBytes := max(maxChars*4, maxChars)
+	b := c.tailBytes(tailBytes)
+	history := compactSummaryText(applyChromeTrim(string(stripANSIBytes(nil, b)), chromeHints))
+	history = tailString(history, maxChars)
+	var out strings.Builder
+	if history != "" {
+		out.WriteString("Recent rendered history:\n")
+		out.WriteString(history)
+		out.WriteString("\n\n")
+	}
+	if grid != "" && !strings.Contains(history, grid) {
+		out.WriteString("Current visible grid:\n")
+		out.WriteString(grid)
+	}
+	return tailString(out.String(), maxChars)
+}
+
+func compactSummaryText(in string) string {
+	in = string(stripANSIBytes(nil, []byte(in)))
+	in = strings.ReplaceAll(in, "\r\n", "\n")
+	in = strings.ReplaceAll(in, "\r", "\n")
+	lines := strings.Split(in, "\n")
+	out := make([]string, 0, len(lines))
+	blank := false
+	for _, line := range lines {
+		line = strings.TrimRightFunc(line, unicode.IsSpace)
+		line = strings.Map(func(r rune) rune {
+			if r == '\t' || r == '\n' {
+				return r
+			}
+			if r < 0x20 || r == 0x7f {
+				return -1
+			}
+			return r
+		}, line)
+		line = truncateSummaryLine(line, summaryMaxLineCells)
+		if strings.TrimSpace(line) == "" {
+			if blank {
+				continue
+			}
+			blank = true
+			out = append(out, "")
+			continue
+		}
+		blank = false
+		out = append(out, line)
+	}
+	return strings.TrimSpace(strings.Join(out, "\n"))
+}
+
+func truncateSummaryLine(s string, max int) string {
+	if max <= 0 || visibleLen(s) <= max {
+		return s
+	}
+	return clipRunes(s, max-1) + "…"
+}
+
+func tailString(s string, max int) string {
+	rs := []rune(s)
+	if len(rs) <= max {
+		return s
+	}
+	return string(rs[len(rs)-max:])
+}
+
+func runSummarizer(ctx context.Context, cfg autoSummarySettings, projectDir, snapshot string) (summarizerResponse, error) {
+	prompt := summaryPrompt(snapshot)
+	out, err := runSummarizerCommand(ctx, cfg, projectDir, prompt)
+	if err != nil {
+		return summarizerResponse{}, err
+	}
+	resp, err := parseSummarizerResponse(out)
+	if err != nil {
+		return summarizerResponse{}, err
+	}
+	if summaryIdleState(resp.State) == StateUnknown {
+		return summarizerResponse{}, fmt.Errorf("invalid summary state %q", resp.State)
+	}
+	return resp, nil
+}
+
+func runSummarizerHealth(ctx context.Context, cfg autoSummarySettings, projectDir string) error {
+	out, err := runSummarizerCommand(ctx, cfg, projectDir, "Reply with exactly: patterm okay")
+	if err != nil {
+		return err
+	}
+	if strings.TrimSpace(out) != "patterm okay" {
+		return fmt.Errorf("health check did not return patterm okay")
+	}
+	return nil
+}
+
+func runSummarizerCommand(ctx context.Context, cfg autoSummarySettings, projectDir, prompt string) (string, error) {
+	provider := cfg.Provider
+	model := cfg.modelFor(provider)
+	var cmd *exec.Cmd
+	switch provider {
+	case "opencode":
+		cmd = exec.CommandContext(ctx, "opencode", "run", "--model", model, "--dir", projectDir, prompt)
+	case "claude":
+		cmd = exec.CommandContext(ctx, "claude", "--print", "--model", model, prompt)
+	default:
+		cmd = exec.CommandContext(ctx, "codex", "exec", "--ephemeral", "--skip-git-repo-check", "--sandbox", "read-only", "--ask-for-approval", "never", "--model", model, "-")
+		cmd.Stdin = strings.NewReader(prompt)
+	}
+	cmd.Dir = projectDir
+	var stderr bytes.Buffer
+	cmd.Stderr = &stderr
+	out, err := cmd.Output()
+	if err != nil {
+		msg := strings.TrimSpace(stderr.String())
+		if msg == "" {
+			msg = err.Error()
+		}
+		return "", fmt.Errorf("%s summarizer: %s", provider, msg)
+	}
+	return string(out), nil
+}
+
+func summaryPrompt(snapshot string) string {
+	return "Summarize this terminal/agent snapshot for a compact UI catch-up aid.\n" +
+		"Return only JSON with keys summary and state. State must be one of IDLE, PERMISSION, THINKING, WORKING, ERROR.\n" +
+		"Keep summary under 180 characters, concrete, and avoid mentioning that you are summarizing.\n\n" +
+		snapshot
+}
+
+func parseSummarizerResponse(out string) (summarizerResponse, error) {
+	var resp summarizerResponse
+	if err := json.Unmarshal([]byte(strings.TrimSpace(out)), &resp); err == nil {
+		return resp, nil
+	}
+	for _, line := range strings.Split(out, "\n") {
+		line = strings.TrimSpace(line)
+		if !strings.HasPrefix(line, "{") || !strings.HasSuffix(line, "}") {
+			continue
+		}
+		if err := json.Unmarshal([]byte(line), &resp); err == nil {
+			return resp, nil
+		}
+	}
+	return resp, fmt.Errorf("summary output was not JSON")
+}
+
+func summaryIdleState(s string) IdleState {
+	switch strings.ToUpper(strings.TrimSpace(s)) {
+	case "IDLE":
+		return StateIdle
+	case "PERMISSION":
+		return StatePermission
+	case "THINKING":
+		return StateThinking
+	case "WORKING":
+		return StateWorking
+	case "ERROR":
+		return StateError
+	default:
+		return StateUnknown
+	}
+}

internal/app/summarizer_test.go

@@ -0,0 +1,85 @@
+package app
+
+import (
+	"strings"
+	"testing"
+
+	"github.com/hjbdev/patterm/internal/preset"
+)
+
+func TestParseSummarizerResponseAllowsWrappedJSON(t *testing.T) {
+	resp, err := parseSummarizerResponse("log\n{\"summary\":\"Waiting for tests\",\"state\":\"WORKING\"}\n")
+	if err != nil {
+		t.Fatalf("parseSummarizerResponse: %v", err)
+	}
+	if resp.Summary != "Waiting for tests" || summaryIdleState(resp.State) != StateWorking {
+		t.Fatalf("response = %+v", resp)
+	}
+}
+
+func TestCompactSummaryTextDropsControlAndRedundantWhitespace(t *testing.T) {
+	got := compactSummaryText("hello\x00 world  \n\n\n\x1b[31mred\x1b[0m\n")
+	if strings.ContainsRune(got, '\x00') {
+		t.Fatalf("control byte survived: %q", got)
+	}
+	if strings.Contains(got, "\n\n\n") {
+		t.Fatalf("redundant blanks survived: %q", got)
+	}
+	if strings.Contains(got, "\x1b") {
+		t.Fatalf("ansi survived: %q", got)
+	}
+}
+
+func TestWrapSidebarSummaryKeepsWordBoundaries(t *testing.T) {
+	got := wrapSidebarSummary("alpha beta gamma delta", 12)
+	want := []string{"alpha beta", "gamma delta"}
+	if len(got) != len(want) {
+		t.Fatalf("lines = %#v", got)
+	}
+	for i := range want {
+		if got[i] != want[i] {
+			t.Fatalf("line %d = %q want %q", i, got[i], want[i])
+		}
+	}
+	long := wrapSidebarSummary("supercalifragilistic short", 8)
+	if len(long) == 0 || !strings.HasSuffix(long[0], "…") {
+		t.Fatalf("long word should clip with ellipsis: %#v", long)
+	}
+}
+
+func TestSummaryManagerArmsOnlyTrackedTopLevelAgents(t *testing.T) {
+	sess := NewSession(t.TempDir(), "test")
+	c := newChildEntry("a1", "agent", KindAgent, []string{"fake"}, nil, "", "", "")
+	running := StatusRunning
+	c.status.Store(&running)
+	sess.children[c.ID] = c
+	sess.order = append(sess.order, c.ID)
+	cfg := defaultSettings().AutoSummary
+	m := newSummaryManager(sess, t.TempDir(), preset.Set{}, func() autoSummarySettings {
+		return cfg.clone()
+	}, nil, nil)
+	m.ObserveHumanInput(c.ID, []byte("please summarize"))
+	if got := m.Summary(c.ID); got.Text != "" {
+		t.Fatalf("untracked agent should not update summary state: %+v", got)
+	}
+	m.RegisterChild(c)
+	m.ObserveHumanInput(c.ID, []byte("please summarize"))
+	m.ObserveOutput(c.ID)
+	m.mu.Lock()
+	e := m.entries[c.ID]
+	m.mu.Unlock()
+	if e == nil || !e.armed || !e.dirty {
+		t.Fatalf("tracked top-level agent not armed/dirty: %+v", e)
+	}
+
+	sub := newChildEntry("a2", "sub", KindAgent, []string{"fake"}, nil, c.ID, "", "")
+	sub.status.Store(&running)
+	m.RegisterChild(sub)
+	m.ObserveHumanInput(sub.ID, []byte("please summarize"))
+	m.mu.Lock()
+	_, ok := m.entries[sub.ID]
+	m.mu.Unlock()
+	if ok {
+		t.Fatal("sub-agent should not get a summary entry")
+	}
+}

internal/app/tabbar.go

@@ -4,12 +4,13 @@ import (
 	"fmt"
 	"os"
 	"strings"
+	"time"
 	"unicode/utf8"
 )
 
-// Two-row tab bar: labels row, underline row. The PTY viewport's top
+// Three-row tab bar: labels row, active-thread summary row, underline row. The PTY viewport's top
 // row is therefore mainTop == tabBarRows + 1.
-const tabBarRows = 2
+const tabBarRows = 3
 
 // drawTabBar renders the top tab strip across the full host width.
 // Tabs share the available width with a flex layout — each visible
@@ -17,9 +18,17 @@ const tabBarRows = 2
 // to the leftmost tabs so the strip fills the screen edge-to-edge.
 // A trailing "+ new" hint sits in the rightmost reserved slot.
 func (st *uiState) drawTabBar() {
+	var entry time.Time
+	if st.metrics != nil {
+		entry = time.Now()
+	}
 	st.mu.Lock()
 	palOpen := st.palette != nil
-	focus := st.focusedID
+	// Highlight the top-level agent tab even when focus has stepped
+	// into a sub-agent (or a Processes pane entry). activeAgentID walks
+	// the parent chain to the root, so the user always sees which tab
+	// their current thread belongs to.
+	focus := st.activeAgentID
 	st.mu.Unlock()
 	if palOpen {
 		return
@@ -130,7 +139,8 @@ func (st *uiState) drawTabBar() {
 	}
 
 	var b strings.Builder
-	// Clear both rows so a stale label from the previous frame can't
+	// Clear all tab-bar rows so stale labels or summaries from the
+	// previous frame can't
 	// bleed through. Use ECH clamped to `width` (= childCols) instead of
 	// `\x1b[2K`: 2K wipes the entire line including the sidebar columns,
 	// and if drawSidebar's chrome cache is fresh it won't repaint to
@@ -138,6 +148,7 @@ func (st *uiState) drawTabBar() {
 	// and content should be.
 	fmt.Fprintf(&b, "\x1b[1;1H\x1b[%dX", width)
 	fmt.Fprintf(&b, "\x1b[2;1H\x1b[%dX", width)
+	fmt.Fprintf(&b, "\x1b[3;1H\x1b[%dX", width)
 
 	for _, t := range tabs {
 		// Row 1: centre-ish label inside the tab cell.
@@ -161,9 +172,9 @@ func (st *uiState) drawTabBar() {
 		b.WriteString(strings.Repeat(" ", rightPad))
 		b.WriteString(styleReset)
 
-		// Row 2: underline. Thick accent for the active tab, faint
+		// Row 3: underline. Thick accent for the active tab, faint
 		// border for the rest.
-		fmt.Fprintf(&b, "\x1b[2;%dH", t.startCol)
+		fmt.Fprintf(&b, "\x1b[3;%dH", t.startCol)
 		if t.active {
 			b.WriteString(styleAccent)
 			b.WriteString(strings.Repeat("━", t.width))
@@ -180,18 +191,28 @@ func (st *uiState) drawTabBar() {
 		fmt.Fprintf(&b, "\x1b[1;%dH %s%s%s ", hintCol, styleDim, newHint, styleReset)
 		// Underline continues faintly under the hint so the strip
 		// reads as one bar.
-		fmt.Fprintf(&b, "\x1b[2;%dH%s%s%s",
+		fmt.Fprintf(&b, "\x1b[3;%dH%s%s%s",
 			hintCol, styleBorder, strings.Repeat("─", newHintW), styleReset)
 	}
 
+	if summary := st.activeSummaryText(width - 2); summary != "" {
+		fmt.Fprintf(&b, "\x1b[2;1H %s%s%s", styleDim, summary, styleReset)
+	}
+
 	frame := b.String()
 	st.chromeCacheMu.Lock()
 	if frame == st.tabBarCache {
 		st.chromeCacheMu.Unlock()
+		if st.metrics != nil {
+			st.metrics.recordTabbar(time.Since(entry), true)
+		}
 		return
 	}
 	st.tabBarCache = frame
 	st.chromeCacheMu.Unlock()
+	if st.metrics != nil {
+		defer func() { st.metrics.recordTabbar(time.Since(entry), false) }()
+	}
 
 	st.outMu.Lock()
 	defer st.outMu.Unlock()

internal/app/tree.go

@@ -96,17 +96,24 @@ func firstRunningAgentID(children []*Child) string {
 }
 
 // processList returns every top-level command/terminal entry in spawn
-// order, regardless of running state. The Processes sidebar section
-// keeps showing exited entries so the user can see what just died (and
-// because Session retains KindCommand entries for restart).
+// order. Exited KindCommand entries remain visible so the user can see
+// what just died and reach restart_process; exited KindTerminal entries
+// are filtered out because terminals are ephemeral and have no restart
+// path (Session also drops them in reapChild — this filter is defensive
+// for any window between exit and deletion).
 func processList(children []*Child) []*Child {
 	out := make([]*Child, 0, len(children))
 	for _, c := range children {
 		if c.ParentID != "" {
 			continue
 		}
-		if c.Kind == KindCommand || c.Kind == KindTerminal {
+		switch c.Kind {
+		case KindCommand:
 			out = append(out, c)
+		case KindTerminal:
+			if c.Status() == StatusRunning {
+				out = append(out, c)
+			}
 		}
 	}
 	return out

internal/app/viewport_renderer.go

@@ -33,6 +33,14 @@ type viewportRenderer struct {
 	// cache so the next drawSidebar repaints over the clobber.
 	scrolled bool
 
+	// childOnAlt tracks whether the focused child has entered its
+	// alternate screen (via ?47 / ?1047 / ?1049). Used to gate mouse-
+	// tracking-mode forwarding to the host: filter on primary so
+	// patterm's wheel-scrollback stays armed, forward on alt so codex
+	// (which disables mouse) lets the user select text and vim (which
+	// enables it) still gets mouse events.
+	childOnAlt bool
+
 	// skipUTF8 is set when the current multi-byte UTF-8 character started
 	// past the viewport's right edge. The starter byte was dropped, so
 	// the remaining continuation bytes must be dropped too instead of
@@ -65,6 +73,16 @@ func newViewportRenderer(l terminalLayout) *viewportRenderer {
 	return vr
 }
 
+// SetChildOnAlt seeds the renderer's view of the focused child's screen
+// side. Used when a new renderer is constructed for an already-running
+// child whose alt-screen transition we missed, so subsequent mouse-mode
+// toggles are filtered/forwarded according to the right side.
+func (vr *viewportRenderer) SetChildOnAlt(onAlt bool) {
+	vr.mu.Lock()
+	defer vr.mu.Unlock()
+	vr.childOnAlt = onAlt
+}
+
 func (vr *viewportRenderer) SetLayout(l terminalLayout) {
 	vr.mu.Lock()
 	defer vr.mu.Unlock()
@@ -236,15 +254,36 @@ func (vr *viewportRenderer) emitCSI() {
 			return
 		}
 		if isAltScreenMode(params) {
+			// Track the child's screen side so we know whether to filter
+			// or forward subsequent mouse-mode toggles. Entering alt
+			// disables host mouse reporting by default so codex (and
+			// any other alt-screen TUI that doesn't request mouse)
+			// allows the user to click-drag to select text. Alt-screen
+			// TUIs that want mouse (vim, less with -X) re-enable it
+			// via ?1000h after switching to alt — the forwarder below
+			// passes that through. Leaving alt re-arms host mouse for
+			// primary-screen wheel-scrollback.
+			wasAlt := vr.childOnAlt
+			vr.childOnAlt = final == 'h'
+			if !wasAlt && vr.childOnAlt {
+				vr.pending.WriteString("\x1b[?1000l\x1b[?1006l")
+			}
+			if wasAlt && !vr.childOnAlt {
+				vr.pending.WriteString("\x1b[?1000h\x1b[?1006h")
+			}
 			return
 		}
 		if isMouseTrackingMode(params) {
-			// Patterm owns mouse reporting on the host so wheel events keep
-			// flowing for scroll-viewport. The child's own emulator still
-			// observes the mode set/reset (it processes the same bytes we
-			// hand to ghostty_terminal_vt_write), so we know whether the
-			// child wants mouse input — we just don't let it disarm our
-			// host listener.
+			// On the child's primary screen patterm owns mouse reporting so
+			// wheel events keep flowing for in-pane scrollback — drop the
+			// child's toggle. On the alt screen the child should be free
+			// to enable mouse (vim, less) or disable it (codex); we forward
+			// the toggle to the host so click-and-drag selection works for
+			// alt-screen TUIs that don't want mouse, and mouse-aware ones
+			// still see the events they need.
+			if vr.childOnAlt {
+				vr.pending.Write(vr.buf)
+			}
 			return
 		}
 	}

internal/app/viewport_renderer_test.go

@@ -16,7 +16,7 @@ func bytesRepeat(b byte, n int) []byte {
 func TestViewportRendererShiftsCursor(t *testing.T) {
 	vr := newViewportRenderer(newTerminalLayout(120, 40))
 	got := string(vr.Render([]byte("\x1b[H")))
-	if got != "\x1b[3;1H" {
+	if got != "\x1b[4;1H" {
 		t.Fatalf("CUP home: got %q", got)
 	}
 }
@@ -24,8 +24,36 @@ func TestViewportRendererShiftsCursor(t *testing.T) {
 func TestViewportRendererSwallowsAltScreenToggles(t *testing.T) {
 	vr := newViewportRenderer(newTerminalLayout(120, 40))
 	got := string(vr.Render([]byte("a\x1b[?1049hb\x1b[?1049lc")))
+	// The ?1049h/l toggles themselves must not reach the host (patterm
+	// owns its own alt screen). On the transition we re-sync host mouse
+	// reporting so codex (which doesn't request mouse) lets the user
+	// drag-select; leaving alt re-arms it for primary-screen wheel
+	// scrollback.
+	want := "a\x1b[?1000l\x1b[?1006lb\x1b[?1000h\x1b[?1006hc"
+	if got != want {
+		t.Fatalf("alt-screen toggles: got %q want %q", got, want)
+	}
+}
+
+func TestViewportRendererMouseTrackingFilteredOnPrimary(t *testing.T) {
+	vr := newViewportRenderer(newTerminalLayout(120, 40))
+	got := string(vr.Render([]byte("a\x1b[?1000lb\x1b[?1000hc")))
 	if got != "abc" {
-		t.Fatalf("alt-screen toggles: got %q", got)
+		t.Fatalf("mouse mode on primary should be filtered: got %q", got)
+	}
+}
+
+func TestViewportRendererMouseTrackingForwardedOnAlt(t *testing.T) {
+	vr := newViewportRenderer(newTerminalLayout(120, 40))
+	// Enter alt; subsequent mouse-mode toggles should reach the host so
+	// alt-screen TUIs (vim, less) can run with mouse on, and selection-
+	// using ones (codex) stay with mouse off.
+	got := string(vr.Render([]byte("\x1b[?1049h\x1b[?1000lx\x1b[?1000hy")))
+	if !strings.Contains(got, "\x1b[?1000l") {
+		t.Fatalf("alt-screen mouse disable should reach host: %q", got)
+	}
+	if !strings.Contains(got, "\x1b[?1000h") {
+		t.Fatalf("alt-screen mouse enable should reach host: %q", got)
 	}
 }
 
@@ -38,7 +66,7 @@ func TestViewportRendererSwallowsOriginModeToggles(t *testing.T) {
 	if !strings.Contains(got, "a") || !strings.Contains(got, "b") || !strings.Contains(got, "c") {
 		t.Fatalf("origin-mode toggles should not drop surrounding text: got %q", got)
 	}
-	if strings.Count(got, "\x1b[3;1H") != 2 {
+	if strings.Count(got, "\x1b[4;1H") != 2 {
 		t.Fatalf("origin-mode set/reset should home inside the viewport twice: got %q", got)
 	}
 }
@@ -60,23 +88,23 @@ func TestViewportRendererOriginModeCUPUsesScrollTop(t *testing.T) {
 	if strings.Contains(got, "\x1b[?6h") {
 		t.Fatalf("origin-mode set leaked to host: %q", got)
 	}
-	if !strings.Contains(got, "\x1b[7;1H") {
-		t.Fatalf("CUP row 1 in origin mode should land at scrollTop row 5 shifted to host row 7: got %q", got)
+	if !strings.Contains(got, "\x1b[8;1H") {
+		t.Fatalf("CUP row 1 in origin mode should land at scrollTop row 5 shifted to host row 8: got %q", got)
 	}
 }
 
 func TestViewportRendererClearScreenIsViewportOnly(t *testing.T) {
-	// hostRows=7 leaves four viewport rows after the 2-row tab bar and
+	// hostRows=7 leaves three viewport rows after the 3-row tab bar and
 	// 1-row status reservation.
 	vr := newViewportRenderer(newTerminalLayout(20, 7))
 	got := string(vr.Render([]byte("\x1b[2J")))
 	if strings.Contains(got, "\x1b[2J") {
 		t.Fatalf("host clear-screen leaked through: %q", got)
 	}
-	if strings.Count(got, "\x1b[20X") != 4 {
+	if strings.Count(got, "\x1b[20X") != 3 {
 		t.Fatalf("clear rows: got %q", got)
 	}
-	if !strings.Contains(got, "\x1b[3;1H") || !strings.Contains(got, "\x1b[6;1H") {
+	if !strings.Contains(got, "\x1b[4;1H") || !strings.Contains(got, "\x1b[6;1H") {
 		t.Fatalf("clear did not target viewport rows: %q", got)
 	}
 }
@@ -112,13 +140,12 @@ func TestViewportRendererClearToEndIsViewportOnly(t *testing.T) {
 		t.Fatalf("host clear-to-end leaked through: %q", got)
 	}
 	// childCols == 19 (40 cols - 28 sidebar - 1 gap - 0-index fudge).
-	// Each of the 4 viewport rows should get a 19-cell erase.
 	// childCols == 11 with hostCols=40 (28 sidebar + 1 gap reserved).
-	// 4 viewport rows, but the cursor row uses ECH at cursor (col 1),
-	// so we expect 4 erases of 11 cells each.
+	// 3 viewport rows, but the cursor row uses ECH at cursor (col 1),
+	// so we expect 3 erases of 11 cells each.
 	count := strings.Count(got, "\x1b[11X")
-	if count != 4 {
-		t.Fatalf("expected 4 ECH-11 sequences, got %d in %q", count, got)
+	if count != 3 {
+		t.Fatalf("expected 3 ECH-11 sequences, got %d in %q", count, got)
 	}
 }
 
@@ -154,7 +181,7 @@ func TestViewportRendererClampsCUPColumn(t *testing.T) {
 	// column so the host cursor never lands in the sidebar.
 	vr := newViewportRenderer(newTerminalLayout(120, 40))
 	got := string(vr.Render([]byte("\x1b[5;95H")))
-	if !strings.Contains(got, "\x1b[7;91H") {
+	if !strings.Contains(got, "\x1b[8;91H") {
 		t.Fatalf("CUP col 95 should clamp to 91 (childCols): got %q", got)
 	}
 }
@@ -249,7 +276,7 @@ func TestViewportRendererFlagsScrollVerbs(t *testing.T) {
 
 func TestViewportRendererFlagsLineFeedAtViewportBottomAsScrolling(t *testing.T) {
 	vr := newViewportRenderer(newTerminalLayout(120, 40))
-	_ = vr.Render([]byte("\x1b[37;1H\n"))
+	_ = vr.Render([]byte("\x1b[36;1H\n"))
 	if !vr.TookScrollAction() {
 		t.Fatalf("LF at viewport bottom should flag scroll")
 	}
@@ -257,7 +284,7 @@ func TestViewportRendererFlagsLineFeedAtViewportBottomAsScrolling(t *testing.T)
 
 func TestViewportRendererDoesNotFlagLineFeedBeforeViewportBottom(t *testing.T) {
 	vr := newViewportRenderer(newTerminalLayout(120, 40))
-	_ = vr.Render([]byte("\x1b[36;1H\n"))
+	_ = vr.Render([]byte("\x1b[35;1H\n"))
 	if vr.TookScrollAction() {
 		t.Fatalf("LF before viewport bottom should not flag scroll")
 	}
@@ -284,7 +311,7 @@ func TestViewportRendererClampsCUUAtViewportTop(t *testing.T) {
 	vr := newViewportRenderer(newTerminalLayout(120, 40))
 	// CUP to viewport row 1 then CUU by 50.
 	got := string(vr.Render([]byte("\x1b[1;1H\x1b[50ACLOBBER")))
-	if !strings.Contains(got, "\x1b[3;1H") {
+	if !strings.Contains(got, "\x1b[4;1H") {
 		t.Fatalf("expected CUP shifted to mainTop: got %q", got)
 	}
 	// The CUU should have been swallowed (n clamped to 0 from row 1).
@@ -311,10 +338,10 @@ func TestViewportRendererClampsCUUPartial(t *testing.T) {
 }
 
 func TestViewportRendererClampsCUDAtViewportBottom(t *testing.T) {
-	// childRows=37 for layout(120, 40). Park cursor at row 37, ask for
+	// childRows=36 for layout(120, 40). Park cursor at row 36, ask for
 	// 10 down → safe step is 0.
 	vr := newViewportRenderer(newTerminalLayout(120, 40))
-	got := string(vr.Render([]byte("\x1b[37;1H\x1b[10B")))
+	got := string(vr.Render([]byte("\x1b[36;1H\x1b[10B")))
 	if strings.Contains(got, "\x1b[10B") {
 		t.Fatalf("CUD past viewport bottom should be dropped: got %q", got)
 	}
@@ -335,10 +362,10 @@ func TestViewportRendererClampsCPLAndHomesColumn(t *testing.T) {
 
 func TestViewportRendererClampsCNL(t *testing.T) {
 	vr := newViewportRenderer(newTerminalLayout(120, 40))
-	// CUP to row 35 then CNL by 50 → safe step is 2 (childRows-35).
-	got := string(vr.Render([]byte("\x1b[35;10H\x1b[50E")))
+	// CUP to row 34 then CNL by 50 → safe step is 2 (childRows-34).
+	got := string(vr.Render([]byte("\x1b[34;10H\x1b[50E")))
 	if !strings.Contains(got, "\x1b[2E") {
-		t.Fatalf("CNL 50 from row 35 should clamp to 2: got %q", got)
+		t.Fatalf("CNL 50 from row 34 should clamp to 2: got %q", got)
 	}
 }
 

internal/harness/scenarios/error_flash_preserves_focused_pane.json

@@ -0,0 +1,37 @@
+{
+  "name": "error_flash_preserves_focused_pane",
+  "presets": {
+    "processes": [
+      {
+        "name": "steady",
+        "argv": ["sh", "-lc", "printf 'STEADY READY\\n'; sleep 5"]
+      }
+    ]
+  },
+  "trust": ["steady"],
+  "steps": [
+    {
+      "type": "mcp_call",
+      "method": "spawn_process",
+      "params": {"kind": "command", "preset": "steady", "name": "steady"},
+      "save_as": "proc"
+    },
+    { "type": "wait_text", "contains": "STEADY READY", "timeout_ms": 5000 },
+    { "type": "send_chord", "chord": "ctrl-k" },
+    { "type": "send_text", "text": "Open Settings" },
+    { "type": "send_chord", "chord": "enter" },
+    { "type": "send_chord", "chord": "enter" },
+    { "type": "send_chord", "chord": "ctrl-n" },
+    { "type": "send_chord", "chord": "ctrl-n" },
+    { "type": "send_chord", "chord": "ctrl-n" },
+    { "type": "send_chord", "chord": "ctrl-n" },
+    { "type": "send_chord", "chord": "ctrl-n" },
+    { "type": "send_chord", "chord": "ctrl-n" },
+    { "type": "send_chord", "chord": "ctrl-n" },
+    { "type": "send_chord", "chord": "enter" },
+    { "type": "wait_text", "contains": "no active top-level agent to summarize", "timeout_ms": 5000 },
+    { "type": "wait_text", "contains": "STEADY READY", "timeout_ms": 5000 },
+    { "type": "assert_contains", "contains": "STEADY READY" },
+    { "type": "assert_not_contains", "contains": "Press Ctrl-K to spawn an agent or process" }
+  ]
+}

internal/harness/scenarios/rename_process_via_palette.json

@@ -16,7 +16,7 @@
     { "type": "send_chord", "chord": "ctrl-k" },
     { "type": "send_text", "text": "Rename process" },
     { "type": "send_chord", "chord": "enter" },
-    { "type": "wait_text", "contains": "Rename process", "timeout_ms": 3000 },
+    { "type": "wait_text", "contains": "process: original", "timeout_ms": 3000 },
     { "type": "send_chord", "chord": "ctrl-u" },
     { "type": "send_text", "text": "renamed-pane" },
     { "type": "send_chord", "chord": "enter" },

internal/harness/scenarios/sidebar_survives_linefeed_scroll.json

@@ -5,7 +5,7 @@
   "scripts": [
     {
       "name": "linefeed-scroll",
-      "body": "#!/bin/sh\n# Plain LF at the bottom of the child viewport scrolls the host's\n# DECSTBM region. Because that region spans every column, enough LFs\n# drag the sidebar border and section labels out of the visible region\n# unless patterm invalidates and repaints the sidebar cache.\ni=0\nwhile [ $i -lt 12 ]; do\n  printf 'warmup %02d\\n' \"$i\"\n  i=$((i + 1))\n  sleep 0.05\ndone\nprintf 'LINEFEED READY\\n'\nIFS= read -r _\nprintf '\\033[1;37r'\nprintf '\\033[37;1H'\ni=0\nwhile [ $i -lt 45 ]; do\n  printf 'scroll line %02d\\n' \"$i\"\n  i=$((i + 1))\ndone\nprintf 'LINEFEED DONE\\n'\nsleep 5\n"
+      "body": "#!/bin/sh\n# Plain LF at the bottom of the child viewport scrolls the host's\n# DECSTBM region. Because that region spans every column, enough LFs\n# drag the sidebar border and section labels out of the visible region\n# unless patterm invalidates and repaints the sidebar cache.\ni=0\nwhile [ $i -lt 12 ]; do\n  printf 'warmup %02d\\n' \"$i\"\n  i=$((i + 1))\n  sleep 0.05\ndone\nprintf 'LINEFEED READY\\n'\nIFS= read -r _\nprintf '\\033[1;36r'\nprintf '\\033[36;1H'\ni=0\nwhile [ $i -lt 45 ]; do\n  printf 'scroll line %02d\\n' \"$i\"\n  i=$((i + 1))\ndone\nprintf 'LINEFEED DONE\\n'\nsleep 5\n"
     }
   ],
   "steps": [
@@ -19,13 +19,13 @@
     { "type": "mark_raw", "save_as": "before_scroll" },
     { "type": "send_chord", "chord": "enter" },
     { "type": "wait_text", "contains": "LINEFEED DONE", "timeout_ms": 5000 },
+    { "type": "wait_stable", "timeout_ms": 2000 },
     {
       "type": "assert_raw_since_regex",
       "from": "before_scroll",
-      "regex": "Agent Tree",
+      "regex": "LINEFEED DONE",
       "timeout_ms": 2000
     },
-    { "type": "wait_stable", "timeout_ms": 2000 },
     { "type": "assert_contains", "contains": "Processes" },
     { "type": "assert_contains", "contains": "Agent Tree" },
     { "type": "assert_contains", "contains": "Scratchpads" },

internal/mcp/protocol.go

@@ -27,6 +27,24 @@ var serverInfo = map[string]any{
 	"version": "0.1.0",
 }
 
+// serverInstructions is returned in the MCP `initialize` response. MCP
+// clients show this to the underlying LLM as context for how to use
+// the server. Failure modes we've seen and want to head off:
+//   - The agent assumes patterm is something it has to launch (running
+//     `patterm` or `patterm mcp-stdio` from its own shell). It's
+//     already attached — it just calls the tools.
+//   - The agent reaches for shell tools (perl / nc / socat / curl) to
+//     poke patterm's Unix socket directly. That socket connection
+//     carries no caller identity, so any sub-agent the agent spawns
+//     that way ends up as a stray top-level tab instead of a child
+//     under the spawning agent. Always go through the MCP tools.
+//   - The agent shells out to `claude` / `codex` / `opencode` to start
+//     a peer instead of calling `spawn_agent`. Those peers won't show
+//     up as sub-agents and won't be tied into the patterm lifecycle.
+//
+// Keep this short — clients vary in how much they surface to the LLM.
+const serverInstructions = "You are already running INSIDE patterm; the `patterm` MCP server is connected over the same stdio MCP transport you use for any other MCP server. Use the MCP tools you see in tools/list — do NOT (a) try to launch `patterm` or `patterm mcp-stdio` yourself, (b) poke the Unix socket through perl / nc / socat / curl, or (c) shell out to `claude` / `codex` / `opencode` to start a peer. Any of those bypasses caller-identity and the new agent will land as a stray top-level tab instead of a child under you. Start with `whoami` for your role and the full tool list, then `help('topics')` for orientation. `spawn_agent` is the only correct way to start a sub-agent; `spawn_process` is for non-LLM commands; `list_processes` / `get_process_output` inspect them; `send_input` / `send_message` drive them. Whatever you spawn is yours to `close_process` when done. When you `send_message` a sub-agent, its reply comes back into YOUR pane as `[sub-agent:<name>] …`, not into the sub-agent's output — to wait for it, use `timer_fire_when_idle_any([sub_agent])` and then read your own pane; do NOT `wait_for_pattern` on the sub-agent, that will deadlock until timeout."
+
 // toolDescriptor is the shape returned by `tools/list`. inputSchema is
 // a JSON Schema object — we provide a minimal `{type: "object"}` schema
 // for each tool, which lets MCP clients accept arbitrary arguments and
@@ -88,7 +106,7 @@ func toolCatalog() []toolDescriptor {
 	return []toolDescriptor{
 		{
 			Name:        "spawn_agent",
-			Description: "Spawn a sub-agent from an agent preset and optionally seed it with initial instructions. Caller owns lifecycle: when the sub-agent's work is done (it reports back via send_message, or you no longer need it), call close_process on its process_id to free the pane and tear down the PTY. See help('lifecycle').",
+			Description: "Spawn a sub-agent from an agent preset and optionally seed it with initial instructions. This is the ONLY correct way to start a sub-agent under you — do not shell out to `claude` / `codex` / `opencode` and do not poke patterm's Unix socket via perl / nc / socat. Either bypasses caller identity and the new agent lands as a stray top-level tab instead of your child. Caller owns lifecycle: when the sub-agent's work is done (it reports back via send_message, or you no longer need it), call close_process on its process_id to free the pane and tear down the PTY. See help('spawning') and help('lifecycle').",
 			InputSchema: objectSchema(map[string]any{
 				"agent":              stringProp("Preset name (e.g. \"claude\", \"codex\")."),
 				"agent_instructions": stringProp("Initial prompt typed into the agent after it's ready."),
@@ -201,7 +219,7 @@ func toolCatalog() []toolDescriptor {
 		},
 		{
 			Name:        "wait_for_pattern",
-			Description: "Block until pattern appears in process output or timeout elapses.",
+			Description: "Block until pattern appears in the TARGET process's own output, or timeout elapses. Use this for waiting on text the target itself will emit (a shell prompt, a build's \"tests passed\" line, etc.). Anti-pattern: do NOT use this to wait for a sub-agent's reply to send_message — replies are routed into the CALLER's pane tagged `[sub-agent:<name>]`, not into the sub-agent's output, so this call will spin to timeout. For sub-agent coordination use `timer_fire_when_idle_any` and then read your own pane.",
 			InputSchema: objectSchema(map[string]any{
 				"process_id":      stringProp("Target process id."),
 				"pattern":         stringProp("Regex pattern."),
@@ -231,7 +249,7 @@ func toolCatalog() []toolDescriptor {
 		},
 		{
 			Name:        "send_message",
-			Description: "Deliver a text message to another process as orchestrator-owned input.",
+			Description: "Deliver a text message to another process as orchestrator-owned input. Fire-and-forget: returns immediately, without waiting for the recipient to read or act. If the recipient replies via send_message, that reply arrives in YOUR pane tagged `[sub-agent:<name>]` (child→parent) or `[orchestrator]` (parent→child) — NOT in the recipient's output. To wait for a sub-agent's reply, schedule `timer_fire_when_idle_any([sub_agent_id], body=…)` and then read your own pane when the timer fires. Do not `wait_for_pattern` on the recipient for a reply; it will deadlock.",
 			InputSchema: objectSchema(map[string]any{
 				"target_process_id": stringProp("Recipient process id."),
 				"message":           stringProp("Message body."),
@@ -265,7 +283,7 @@ func toolCatalog() []toolDescriptor {
 		},
 		{
 			Name:        "timer_fire_when_idle_any",
-			Description: "Schedule a timer that fires when any watched process enters idle (already-idle entries excluded), or when max_wait_seconds elapses.",
+			Description: "Canonical way to wait for a sub-agent to finish working: send_message the sub-agent, then schedule this with watched=[sub_agent_id]; when it fires, the reply is already sitting in your own pane tagged `[sub-agent:<name>]`. Schedules a timer that fires when any watched process enters idle (already-idle entries excluded), or when max_wait_seconds elapses.",
 			InputSchema: objectSchema(map[string]any{
 				"watched":          arrayOfStringsProp("Process ids to watch."),
 				"body":             stringProp("Message delivered verbatim to the owning agent when the timer fires."),
@@ -276,7 +294,7 @@ func toolCatalog() []toolDescriptor {
 		},
 		{
 			Name:        "timer_fire_when_idle_all",
-			Description: "Schedule a timer that fires when all watched processes are idle (already-idle entries count as satisfied), or when max_wait_seconds elapses.",
+			Description: "Canonical way to wait for several sub-agents to finish working in parallel: send_message each one, then schedule this with watched=[…ids]; when it fires, each reply is in your own pane tagged `[sub-agent:<name>]`. Schedules a timer that fires when all watched processes are idle (already-idle entries count as satisfied), or when max_wait_seconds elapses.",
 			InputSchema: objectSchema(map[string]any{
 				"watched":          arrayOfStringsProp("Process ids to watch."),
 				"body":             stringProp("Message delivered verbatim to the owning agent when the timer fires."),
@@ -377,7 +395,8 @@ func (s *Server) handleProtocolMethod(callerID, method string, params json.RawMe
 			"capabilities": map[string]any{
 				"tools": map[string]any{"listChanged": false},
 			},
-			"serverInfo": serverInfo,
+			"serverInfo":   serverInfo,
+			"instructions": serverInstructions,
 		}
 		return result, true, 0, "", nil
 

internal/mcp/protocol_test.go

@@ -36,6 +36,13 @@ func TestInitializeReturnsCapabilities(t *testing.T) {
 	if caps["tools"] == nil {
 		t.Fatalf("tools capability missing: %+v", caps)
 	}
+	// patterm-specific orientation: clients show this to the underlying
+	// LLM, so it's our primary hook for steering vendor TUIs (codex in
+	// particular) toward the MCP tool surface instead of shell-ing out.
+	instructions, ok := parsed.Result["instructions"].(string)
+	if !ok || instructions == "" {
+		t.Fatalf("instructions missing or wrong type: %+v", parsed.Result)
+	}
 }
 
 func TestInitializedNotificationSuppressesResponse(t *testing.T) {

internal/preset/preset.go

@@ -300,14 +300,6 @@ func ensureDefaults(base string) error {
     "^\\s*>_"
   ]
 }
-`,
-		},
-		{
-			"presets/processes/shell.json",
-			`{
-  "name": "shell",
-  "argv": ["__SHELL__"]
-}
 `,
 		},
 	}
@@ -319,15 +311,7 @@ func ensureDefaults(base string) error {
 		if err := os.MkdirAll(filepath.Dir(full), 0o700); err != nil {
 			return err
 		}
-		body := d.body
-		if strings.Contains(body, "__SHELL__") {
-			shell := os.Getenv("SHELL")
-			if shell == "" {
-				shell = "/bin/sh"
-			}
-			body = strings.ReplaceAll(body, "__SHELL__", shell)
-		}
-		if err := os.WriteFile(full, []byte(body), 0o600); err != nil {
+		if err := os.WriteFile(full, []byte(d.body), 0o600); err != nil {
 			return err
 		}
 	}