feat(desktop): redesign the Huddle experience - #4281

Requested by Tyler (buzz-tui df272ea1, round 2): pull main into PR #4347
as a new commit on top, no rebase, no force. Main freight is one commit,
the Huddle redesign (#4281), which restructured AppShell.tsx (extracted
AppHuddleShell / AppShellChannelSurface / LazySettingsScreen) and moved
the initial-window helpers from lib.rs into initial_window.rs. Two
conflicts, both resolved in favor of main's new structure with the
terminal lane re-applied on top:

- desktop/src-tauri/src/lib.rs: took main's side (helpers now live in
  initial_window.rs; the lane's copies were byte-equivalent modulo
  blank lines and one comment).
- desktop/src/app/AppShell.tsx: took main's restructured file, then
  re-applied the lane's two changes: useTerminalContext replaces the
  bare activeChannel memo, and TerminalBootstrap mounts behind the app
  surface via a new optional `terminal` slot on AppHuddleShell (not
  mounted in the dedicated Huddle room window).

Co-authored-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Signed-off-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>

…er-snapshots

* origin/main:
  feat(mobile): sync per-group channel sorting (#4231)
  feat(mobile): add channel scroll navigation (#4239)
  feat(desktop): redesign the Huddle experience (#4281)
  feat(mobile): bring channel menus to desktop parity (#3940)

Signed-off-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

# Conflicts:
#	desktop/src-tauri/src/lib.rs

…onfig

* origin/main:
  feat(mobile): sync per-group channel sorting (#4231)
  feat(mobile): add channel scroll navigation (#4239)
  feat(desktop): redesign the Huddle experience (#4281)
  feat(mobile): bring channel menus to desktop parity (#3940)
  feat(agents): model-tuning parity in global Agent Defaults editor (#4578)

Signed-off-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>
Co-authored-by: Taylor Ho <[email protected]>
Signed-off-by: Taylor Ho <[email protected]>

…theme-config

* origin/main: (81 commits)
  feat(mobile): sync per-group channel sorting (#4231)
  feat(mobile): add channel scroll navigation (#4239)
  feat(desktop): redesign the Huddle experience (#4281)
  feat(mobile): bring channel menus to desktop parity (#3940)
  feat(agents): model-tuning parity in global Agent Defaults editor (#4578)
  Polish Share Compute settings (#3735)
  fix(reactions): wrap long popover names (#3834)
  fix(desktop): clarify inherited agent parallelism (#4010)
  feat(desktop): make onboarding model defaults skippable (#3968)
  ci: add guarded desktop release cache prewarm (#4575)
  fix(mobile): recover stale relay sessions (#4372)
  chore(release): release Buzz Desktop version 0.5.4 (#4562)
  test(mobile): assert follow boundary semantics (#4559)
  docs(release): align desktop handoff instructions (#3988)
  fix: report agent usage per provider round, not once per turn (#4545)
  fix(desktop): harden Windows installs against Defender block and orphaned Node (#4382)
  feat(desktop): improve channel template discovery (#4549)
  fix(desktop): save key backups to authorized path (#4022)
  Add channel activity hover menu (#3935)
  feat(desktop): show saved Run on settings when editing an agent (#4539)
  ...

Signed-off-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>
Co-authored-by: Taylor Ho <[email protected]>
Signed-off-by: Taylor Ho <[email protected]>

Resolves two conflicts against main:

AppShell.tsx: HEAD had const { activeCommunity, reinitKey } = communitiesHook
for the composite workspace key; origin/main added useHuddlePresentation()
destructuring from the Huddle redesign (#4281). Resolution keeps both: the
composite key is required for useManagedAgentRuntimeReconciliation, and the
Huddle hooks are needed for the new Huddle UI.

MeshComputeSettingsCard.tsx: HEAD had the Stop using shared compute affordance
plus the legacy inline model section; origin/main (#3735) replaced the inline
model section with the MeshModelPicker component. Resolution keeps the Stop
button block and adopts the MeshModelPicker layout, discarding the replaced
inline model controls.

Also corrects the false comment at runtime_commands_tests.rs:342-345 that
claimed compensate_drain is covered by the desktop integration test suite.
The compensation round-trip requires an AppHandle; the codebase has no
tauri::test harness and no AppHandle mock. The honest coverage statement is:
drain-prefix contract proven by unit test, restart path integration-only.

Co-authored-by: Will Pfleger <[email protected]>
Signed-off-by: Will Pfleger <[email protected]>

…-style

* origin/main: (25 commits)
  Polish mobile inbox and media flows (block#4512)
  feat: ship Buzz Term (block#4347)
  feat(mobile): sync per-group channel sorting (block#4231)
  feat(mobile): add channel scroll navigation (block#4239)
  feat(desktop): redesign the Huddle experience (block#4281)
  feat(mobile): bring channel menus to desktop parity (block#3940)
  feat(agents): model-tuning parity in global Agent Defaults editor (block#4578)
  Polish Share Compute settings (block#3735)
  fix(reactions): wrap long popover names (block#3834)
  fix(desktop): clarify inherited agent parallelism (block#4010)
  feat(desktop): make onboarding model defaults skippable (block#3968)
  ci: add guarded desktop release cache prewarm (block#4575)
  fix(mobile): recover stale relay sessions (block#4372)
  chore(release): release Buzz Desktop version 0.5.4 (block#4562)
  test(mobile): assert follow boundary semantics (block#4559)
  docs(release): align desktop handoff instructions (block#3988)
  fix: report agent usage per provider round, not once per turn (block#4545)
  fix(desktop): harden Windows installs against Defender block and orphaned Node (block#4382)
  feat(desktop): improve channel template discovery (block#4549)
  fix(desktop): save key backups to authorized path (block#4022)
  ...

Amp-Thread-ID: https://ampcode.com/threads/T-019fccee-fa22-7145-95bc-a3aa0c10a174
Co-authored-by: Amp <[email protected]>
Signed-off-by: Joah Gerstenberg <[email protected]>

# Conflicts:
#	desktop/package.json
#	desktop/src-tauri/src/commands/agent_config.rs
#	desktop/src/app/AppShell.tsx
#	desktop/src/features/channels/hooks.ts
#	desktop/src/features/channels/unreadReadMarker.test.mjs
#	desktop/src/features/channels/useUnreadChannels.ts
#	desktop/src/features/sidebar/ui/CommunityRail.tsx
#	desktop/src/main.tsx
#	pnpm-lock.yaml

…-phase2-integration

* origin/main: (23 commits)
  Refine community invite links (#4734)
  feat(desktop): persist sidebar observed-unread across webview reload (#3976)
  feat(desktop): surface config diff in restart-required badge (#3637)
  Polish sidebar unread hierarchy (#4573)
  fix(desktop): show cached display names on startup (#3317)
  docs(acp): explain per-channel session model in base prompt (#4729)
  docs(nip-am): normative amendment — cache SHOULD/MUST + pricingIdentity + consumer cost guidance (#4632)
  Remove blur from Welcome composer guidance (#4691)
  Refine desktop timeline activity presentation (#4582)
  Defer desktop media uploads until send (#4522)
  fix(desktop): stop clipping focus ring on channel intro action cards (#2392) (#4374)
  Polish mobile inbox and media flows (#4512)
  feat: ship Buzz Term (#4347)
  feat(mobile): sync per-group channel sorting (#4231)
  feat(mobile): add channel scroll navigation (#4239)
  feat(desktop): redesign the Huddle experience (#4281)
  feat(mobile): bring channel menus to desktop parity (#3940)
  feat(agents): model-tuning parity in global Agent Defaults editor (#4578)
  Polish Share Compute settings (#3735)
  fix(reactions): wrap long popover names (#3834)
  ...

* fix(mobile): recover stale relay sessions (#4372)

### Summary

Fixes [this
issue](buzz://message?channel=e62570dd-33ad-42c5-b92b-75f2689f9694&id=b726c366abfe62429ee3cdcd34d0c0fb98c33c3ea053480585bed71745412b56):
> I often don’t see my bot responses until after I post. they’re usually
time stamped correctly so I think it’s just a refresh issue?

### What changed?

Buzz Mobile now reconnects relay sessions after the app has remained
backgrounded beyond the existing 5-second grace period, even when the
session still reports a stale `connected` state. This makes resume
recovery independent of whether iOS runs the grace timer before or after
delivering `resumed`.

Reconnection is now based on elapsed background time rather than a
direct socket-health probe.
- If the app was backgrounded for at least the 5-second grace period,
the socket is presumed dead and the session reconnects regardless of
reported status.
- If it was backgrounded for less than that, a reported `connected`
status is still trusted.

In the sub-5-second window the socket is either genuinely alive, which
is the common case for a momentary background, or it is dead and the
client ping detects it within the two-interval worst case described
below. That is now a degraded-latency path, not a silent-forever path.

The mobile relay socket now uses `IOWebSocketChannel.connect` with a
30-second `pingInterval`. An unanswered ping closes the Dart socket
through the existing disconnect and reconnect path.

Detection takes up to two ping intervals, so about 60 seconds worst
case, not 30. One interval of idleness elapses and a ping is sent, then
a second interval elapses with no pong and the socket closes. Any
inbound pong restarts the first stage, so the clock measures idleness
rather than running on a fixed cadence.

### Why?

Buzz iOS can sometimes stop showing new bot or agent responses after a
phone has been locked for 5 to 10 minutes. When the user later posts a
message, the missing responses can appear all at once. iOS may suspend
Buzz before the short delayed cleanup that would normally close its
connection has a chance to run. Before this change, Buzz trusted the
resulting stale healthy status on resume and skipped reconnecting, so
the missing responses stayed hidden until a later post exposed the dead
connection.

A state-machine test with a stubbed connection reproduced this reported
pattern and showed that it matches this failure mode: the failed post
triggered a reconnect that fetched the missing messages. The same test
also checked the other candidate explanation, the bug tracked in
[#3053](https://github.com/block/buzz/pull/3053), where the relay has
closed the app's subscription. That state does not produce the pattern.
Posting succeeds and the user's own message appears, but nothing looks
for the missed messages, so they stay hidden. The test confirmed that
the missed messages were still available to fetch in that state, so the
missing step was a trigger to fetch them. This was not an end-to-end
reproduction on an iOS device or a live relay.

The new resume check covers the normal lock and unlock path. If the app
was backgrounded for less than the 5-second grace period, it still
trusts a connection marked as healthy. A dead connection in that window
is instead detected by the ping check, which can take up to about 60
seconds but prevents the app from remaining silently stuck. The ping
only runs while iOS is running the app, so it does not detect a
connection that died during suspension; the resume check owns the lock
and unlock path.

A pre-existing path also runs the same resume handling when network
connectivity returns while the app is already in the foreground. Because
the app was not backgrounded, this change does not alter that path,
which still trusts a connection marked as healthy and relies on the
slower ping check.

Recovery from a subscription that the relay explicitly closes remains in
[#3053](https://github.com/block/buzz/pull/3053), and the two changes
overlap in one file. Changes to how missed messages are backfilled or
replayed are out of scope.

### How is it tested?

Full mobile suite at base and head. Both runs have the same known
macOS-host-only failure in `ChannelDetailPage keeps follow mode off
while a tall newest message stays visible` at line 1053:

- Base: 1,021 passed, 1 skipped, 1 failed
- Head: 1,025 passed, 1 skipped, 1 failed

Added tests:

-
[`relay_session_test.dart`](https://github.com/block/buzz/tree/main/mobile/test/shared/relay/relay_session_test.dart):
long-background resume reconnect and within-grace control
-
[`relay_socket_liveness_test.dart`](https://github.com/block/buzz/tree/main/mobile/test/shared/relay/relay_socket_liveness_test.dart):
silent-peer disconnect and idle-but-healthy control

Mutation checks confirm that removing elapsed-background resume recovery
fails with one socket instead of two, and removing `pingInterval` leaves
the silent peer connected. Restored production code passes both
mutations' regression tests and the healthy idle control.

Signed-off-by: Tom Brow <[email protected]>
Co-authored-by: npub1tquskdu6yc4h8l7xxtceculxw600grekeq0xg2ukqfrwl7vrzg3quz3gmp <58390b379a262b73ffc632f19c73e6769ef40f36c81e642b960246eff9831222@buzz.block.builderlab.xyz>

* ci: add guarded desktop release cache prewarm (#4575)

## Summary

Gate 1 only for desktop release caching:

- replaces canary `rust-cache` use with explicit exact-key
`actions/cache/restore` + `save`
- computes keys after `cargo update --workspace`, including platform,
target, Rust toolchain, Cargo manifests/locks, profile/features, and
native-toolchain inputs
- normalizes only the desktop package version so a trusted `main` canary
can warm an otherwise identical release tag
- excludes Tauri bundle directories, so installers and signed artifacts
are never cached
- adds a restore-only `cache-proof-*` tag workflow that fails unless tag
scope sees the exact default-branch cache
- adds contract tests that enforce no release-workflow cache change in
Gate 1

`release.yml` is intentionally unchanged. A cache miss remains the
current cold canary build; the release path cannot be affected by
merging this PR.

## Validation

- `scripts/test-desktop-release-cache-key.sh`
- `scripts/test-desktop-release-cache-workflow.sh`
- `scripts/test-release-ref-contract.sh`
- Ruby YAML parse of all four changed workflows
- `git diff --check`
- pre-push `branch-skew`

## Post-merge proof plan

1. Run each canary cold on trusted `main`, recording cache size/save
time and fresh artifact inventory.
2. Run each canary warm, requiring the exact-key hit and recording
restore/build time.
3. Create a disposable `cache-proof-*` tag at that same trusted `main`
SHA and dispatch **Desktop release cache tag-scope proof** from the tag.
4. Do not begin Gate 2 or modify `release.yml` unless the exact
tag-scope restore succeeds and cache transfer economics are favorable.

---------

Signed-off-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

* feat(desktop): make onboarding model defaults skippable (#3968)

**Category:** improvement
**User Impact:** Users can skip default model configuration during
onboarding and finish it later in Settings → Agents.

**Problem:** Requiring model defaults during onboarding can block users
who are not ready to choose a harness, provider, or model. Skipping also
needs to leave existing configuration untouched rather than persisting
partial selections.

**Solution:** Stage onboarding edits locally and persist them only when
users choose Next or Back. A delayed Skip action advances without any
configuration write, while a footer hint points users to the settings
location for completing setup later.

<details>
<summary>File changes</summary>

**desktop/src/features/onboarding/ui/DefaultConfigStep.tsx**
Adds the skip action and future-settings hint, and makes model
configuration transactional so Skip discards staged changes while Next
and Back preserve the intended save behavior.

**desktop/src/testing/e2eBridge.ts**
Exposes model-config setter call counts so tests can distinguish a true
zero-write skip from a write-and-rollback implementation.

**desktop/tests/e2e/onboarding-agent-defaults.spec.ts**
Covers skipping during loading and after staged edits, verifies zero
persistence calls, and confirms Next and Back still commit changes.

</details>

## Reproduction steps

1. Start fresh onboarding and continue through harness setup to
**Configure your default model settings**.
2. Change the selected harness or model, then choose **Skip for now**.
3. Confirm onboarding advances to **Join or create a community** and the
prior global model configuration remains unchanged.
4. Return through onboarding and confirm **Next** saves the staged
selection; confirm **Back** also preserves staged changes before
returning.
5. Confirm the footer says model defaults can be configured later in
**Settings → Agents**.

---------

Signed-off-by: Taylor Ho <[email protected]>
Signed-off-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

* fix(desktop): clarify inherited agent parallelism (#4010)

## Summary

- show an unambiguous `App default (10)` inherited state for parallelism
in create and edit forms
- explain that blank inherits the app default and suppress create-form
number steppers that could silently set `1`
- align the E2E mint fallback with production while preserving explicit
input → definition → app-default precedence

## Why

The forms displayed `1` even though an untouched field is omitted and
desktop minting materializes `10`. The create-form spinner could also
turn blank/inherited into an explicit `1` with one click while leaving
the field looking nearly unchanged.

## Testing

- `pnpm test` (desktop: 3,886 passed)
- `pnpm typecheck` (desktop)
- `pnpm check` (desktop)
- pre-push `desktop-check` and `desktop-test`

---------

Signed-off-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

* fix(reactions): wrap long popover names (#3834)

**Category:** fix
**User Impact:** Long custom emoji names now stay contained inside
reaction popovers and remain fully readable.

**Problem:** An unbroken custom emoji name could force a reaction
popover beyond its intended maximum width and overflow the message view.

**Solution:** Give the reaction popover a definite 288px width and allow
the complete emoji name to wrap within it without truncation or
ellipsis. Short names retain the same content and interaction behavior.

<details>
<summary>File changes</summary>

**desktop/src/features/messages/ui/MessageReactions.tsx**
Bounds the reaction popover width and allows long names to break across
lines while preserving the full shortcode.

**desktop/tests/e2e/reaction-names.spec.ts**
Covers fixed width, full text preservation, and wrapping for the maximum
supported colon-wrapped reaction name, with deterministic seeded Picsum
visual fixtures and explicit image-load waits.

</details>

## Reproduction Steps

1. Open a message with a custom emoji reaction whose name is 64
characters.
2. Hover or focus the reaction pill to open its details popover.
3. Confirm the popover remains 288px wide and the complete name wraps
within it without ellipsis.
4. Open a short-name reaction and confirm its popover remains readable
and unchanged in behavior.

## Screenshots

| Before | After |
| --- | --- |
| ![Maximum-length name
before](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3834/max-length-before-picsum.png)
| ![Maximum-length name
after](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3834/max-length-after-picsum.png)
|

**Short-name regression check**

![Short reaction
name](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3834/short-name-after-picsum.png)

## Verification

- `pnpm test` in `desktop`: 3,858 passed
- Focused reaction-name E2E with seeded Picsum captures: 2 passed
- Desktop checks and commit hooks passed

Originating Buzz channel: `f2ec9671-d78e-4cde-894c-9f4c458c7f1f`

---------

Signed-off-by: Taylor Ho <[email protected]>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

* Polish Share Compute settings (#3735)

## Summary
- Refresh Share Compute with the shared agent-style model controls.
- Reveal sharing details and advanced options only while sharing.
- Remove the preview-only mesh API path.

## Validation
- `pnpm check`
- `pnpm test`
- `pnpm exec playwright test tests/e2e/mesh-compute.spec.ts`

Snapshots are attached in a follow-up comment.

---------

Signed-off-by: kenny lopez <[email protected]>

* feat(agents): model-tuning parity in global Agent Defaults editor (#4578)

## Overview

The global Agent Defaults surface (Settings card, defaults modal,
onboarding) exposed structured controls for Effort but left Max Output
Tokens, Context Limit, and Max Rounds as raw env vars. Per-agent dialogs
had structured numeric fields but only for `isBuzzAgentRuntime` —
incorrectly excluding Goose. This PR unifies numeric-tuning capability
across all surfaces, fixes a pre-existing dual-editor defect, and adds
full test coverage.

## What changed

### Phase 1 — Catalog projection

- Add `max_rounds_env_var` to `KnownAcpRuntime` in `runtime_metadata.rs`
(`Some("BUZZ_AGENT_MAX_ROUNDS")` for buzz-agent, `None` elsewhere).
- Project all three numeric env-var fields (`max_tokens_env_var`,
`context_limit_env_var`, `max_rounds_env_var`) end-to-end:
`AcpRuntimeCatalogEntry` Rust struct, TS `types.ts`,
`RawAcpRuntimeCatalogEntry` + `fromRawAcpRuntimeCatalogEntry` in
`tauri.ts`, and the e2e mock bridge (`withMockRuntimeConfigMetadata`).

### Phase 2 — Field model

- `deriveAgentConfigFieldModel` now derives `maxOutputTokens` /
`contextLimit` / `maxRounds` descriptors from catalog-projected fields.
- `structuredEnvKeys(descriptors)` — exported helper that takes the
**rendered** descriptor set (not the whole model). Hidden keys follow
what is actually rendered per surface: global hides effort + all three
numeric keys for buzz-agent / two for Goose; per-agent buzz-agent hides
effort + three numeric keys; per-agent Goose hides only its two numeric
keys. `BUZZ_AGENT_THINKING_EFFORT` stays a visible generic env row
per-agent because no effort control renders there.

### Phase 3 — UI

- Extract `NumericTuningFields` from `buzzAgentModelTuningFields.tsx` as
a shared descriptor-driven component (`descriptors`, `envVars`,
`inheritedEnvVars`, `onEnvVarChange`). Kind-specific minima:
`NUMERIC_KIND_MIN` map (`maxOutputTokens`/`contextLimit`: 1,
`maxRounds`: 0) applied to `<input min>`.
- **Global surface** (`AgentConfigFields.tsx`): deduplicate the
previously duplicated Advanced env-editor block; render
`NumericTuningFields` below the env editor when descriptors exist;
`hiddenKeys` and `bakedGenericRows` exclusions use `structuredEnvKeys`
so structured keys are never double-rendered. Under 1000 lines.
- **Per-agent surfaces** (`EditAgentAdvancedFields`,
`PersonaAdvancedFields`): replace `isBuzzAgentRuntime` as the
numeric-field gate with `deriveNumericDescriptors(selectedRuntime)` from
`agentConfigCore`; hidden keys come from
`structuredEnvKeys(numericDescriptors)` — the same rendered descriptor
set, no local rebuilding (fixes pre-existing dual-editor defect).
Catalog status carried as `RuntimeCatalogStatus` (`loading | ready |
error`); both error and loading withhold structured controls and leave
saved values visible as generic rows, making error distinguishable from
"runtime not capable" (`ready` + no runtime).
- **Dialogs** (`AgentDefinitionDialog`, `AgentInstanceEditDialog`,
callers): `AgentDefinitionDialog` accepts `runtimeCatalogStatus?:
"loading" | "ready" | "error"` (replaces separate
`runtimesLoading`/`runtimesError` booleans); all call sites —
`AgentManagementDialogs`, `AgentsView`, `RequestedAgentCreateDialogs`,
`UserProfilePersonaDialogs` — compute and pass the status.

### Phase 4 — Tests

- `buildRecord` exported from `EnvVarsEditor.tsx` as a pure `(nextRows,
value, requiredKeys, hiddenKeys) => Record<string, string>` helper for
isolation testing.
- **17 new node tests** in `agentConfigCore.test.mjs`:
`deriveNumericDescriptors` (all three fields, partial, undefined
runtime, matches field-model subset); `structuredEnvKeys` per surface
including discriminating Goose per-agent effort-key invariant;
`NUMERIC_KIND_MIN` values.
- **4 new node tests** in `EnvVarsEditor.test.mjs`: hidden tuning key
preserved through generic row edits; runtime-switch then generic edit
(derives both descriptor sets, asserts new-runtime hidden key survives
`buildRecord` via `hiddenKeys` and old-runtime key survives via generic
rows); baked numeric key excluded via `filterBakedGenericRows` with
`numericTuningPlaceholder` assertion; clearing a structured override —
`numericTuningPlaceholder` verifies placeholder text.
- **5 new Playwright tests** in `agent-numeric-tuning.spec.ts` (added to
smoke project `testMatch`): global numeric fields visible for
buzz-agent; global: non-capable runtime hides numeric controls; Goose
per-agent shows `Inherit (16384)` after saving global value through the
UI; delayed catalog: saved values visible as generic rows while loading
then structured controls appear after settle; failed catalog: saved
values remain visible as generic rows (never the "unsupported" empty
state).

## Result

- buzz-agent global defaults: Max output tokens, Context limit, Max
rounds as structured inputs with `Inherit (N)` placeholders from baked
env.
- Goose global defaults: Max output tokens, Context limit as structured
inputs.
- A Goose global value surfaces as `Inherit (<value>)` in the per-agent
Goose edit dialog.
- No structured key is editable in two places on any surface; no
persisted key has zero editors.
- No `runtime.id === "buzz-agent"` comparison decides numeric-field
visibility anywhere — capability flows catalog →
`AcpRuntimeCatalogEntry` → field model → UI.

Signed-off-by: Will Pfleger <[email protected]>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>

* feat(mobile): bring channel menus to desktop parity (#3940)

## Overview

**Category:** improvement  
**User Impact:** Mobile users can now access consistent channel and DM
actions from both the channel list and conversation header.
**Problem:** Mobile channel menus exposed a narrower, inconsistent set
of actions than desktop, and the available actions differed by entry
point.
**Solution:** This change introduces one reusable action sheet with a
clear quick-action hierarchy, role-aware lifecycle controls,
confirmations for consequential actions, and a deliberately narrower DM
menu.

## Changes

<details>
<summary>File changes</summary>

**mobile/lib/features/channels/channel_actions_sheet.dart**  
Adds the shared channel and DM action-sheet experience used by both
entry points, including Star/Unstar and Read/Unread quick actions for
channels, section movement, mute, management, inline copy actions,
guarded lifecycle actions, confirmations, and a compact DM menu without
quick actions.

**mobile/lib/features/channels/channel_detail_page.dart**  
Routes the header ellipsis through the shared action sheet so the
in-channel menu matches the channel-list experience, including for DMs.

**mobile/lib/features/channels/channel_management_provider.dart**  
Adds archive and delete operations using the desktop-compatible relay
event kinds and refreshes channel state after completion.

**mobile/lib/features/channels/channels_page.dart**  
Makes the shared channel action-sheet entry point available to the
channel-list implementation.

**mobile/lib/features/channels/channels_page/channel_tile.dart**  
Replaces the tile-specific long-press menu with the reusable action
sheet while preserving read state and section context.

**mobile/test/features/channels/channel_actions_sheet_test.dart**  
Covers action hierarchy, owner/admin/member capability guards, loading
and failure states, DM narrowing with no quick-action row, and inline
copy actions.

**mobile/test/features/channels/channel_detail_page_test.dart**  
Updates channel-header flows to exercise management through the new
shared action sheet.

**mobile/test/features/channels/channel_management_provider_test.dart**
Verifies archive and delete event tags stay compatible with desktop
behavior.

</details>

## Reproduction Steps

1. Run the mobile app and open a populated channel list.
2. Long-press a regular channel and verify the Star/Unstar and
Read/Unread quick actions appear above Move to section…, Mute, Manage,
Copy channel name, and Copy channel ID.
3. Choose either copy action and verify it copies the expected value.
4. Open a channel, tap the header ellipsis, and verify the same action
sheet appears.
5. As an admin or owner, verify Archive appears; as an owner, verify
Delete also appears. Confirm that lifecycle actions require
confirmation.
6. Long-press or open the header menu for a DM and verify it has no
quick-action row and starts with Mute, followed by Copy channel name and
Copy channel ID.

## Screenshots

### Channel menu

| Regular channel — Mark Unread | DM — no quick actions | Archive
confirmation |
|---|---|---|
| ![Regular channel actions with Mark
Unread](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3940/final-regular-channel-mark-unread.png)
| ![DM actions without quick
actions](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3940/final-dm-no-quick-actions.png)
| ![Archive
confirmation](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3940/final-archive-confirmation.png)
|

---------

Signed-off-by: Taylor Ho <[email protected]>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

* feat(desktop): redesign the Huddle experience (#4281)

## Summary

- open Huddles in a focused companion window with a clean handoff back
to the in-app drawer and backing channel
- redesign the participant film strip, sidebar control, transcript
surface, and themed shell treatment
- preserve microphone and device control across windows, start agent
voice on the first reply, and show agent speaking activity in the film
strip
- give each agent a distinct session voice, beginning with the
configured default, plus compact per-agent text-to-speech and voice
controls
- enroll only agents explicitly mentioned or deliberately added through
an agent panel into the live Huddle roster
- keep temporary Huddle channels out of the sidebar unless the user
explicitly brings one into the main app
- remove Huddle-only avatar policy badges and filter short silence or
noise segments before speech-to-text posts

## Why

The previous flow exposed the temporary channel as product UI, obscured
who was present or speaking, and split transcript and audio state
between the main and companion windows. This keeps backing channels as
implementation details unless a user explicitly brings a Huddle into the
app, while sharing the live conversation and audio lifecycle across both
surfaces. Agent participants now join only after an explicit invitation,
distinct voices make multi-agent Huddles easier to follow, and short
microphone noise no longer becomes stray transcript messages.

## Validation

- `pnpm check`
- `pnpm build:e2e`
- `pnpm exec playwright test tests/e2e/huddle-transcription.spec.ts
--project=smoke` (13 passed)
- Huddle sidebar visibility unit coverage (4 passed)
- focused managed-agent and persona-mention E2E coverage (2 passed)
- `pnpm test` (3,910 passed)
- `cargo clippy --manifest-path desktop/src-tauri/Cargo.toml
--all-targets -- -D warnings`
- `cargo test --manifest-path desktop/src-tauri/Cargo.toml` (2,093
passed, 14 ignored; 3 diagnostics passed)

---------

Signed-off-by: kenny lopez <[email protected]>
Signed-off-by: Wes <[email protected]>
Co-authored-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

* feat(mobile): add channel scroll navigation (#4239)

**Category:** improvement
**User Impact:** Mobile readers can jump directly to their oldest unread
message and return to the latest message with compact directional
controls.
**Problem:** Opening an active channel at its newest message makes it
easy to miss where unread conversation began, while moving back through
history lacks a lightweight route to the live edge.
**Solution:** Capture the channel's unread boundary when it opens, offer
an accessible up-chevron beneath the app bar to reach that stable
target, then reveal the inverse down-chevron at the bottom whenever the
reader is away from latest. Deep links retain precedence, and
live-follow, pagination, composer resizing, and explicit scroll
ownership continue to use the existing timeline behavior.

<details>
<summary>File changes</summary>

**mobile/lib/features/channels/channel_detail_page.dart**
Captures the channel's read state at open time and passes a stable
unread snapshot into the timeline before the normal deferred read update
advances it.

**mobile/lib/features/channels/channel_detail_page/message_list.dart**
Adds mutually exclusive oldest-unread and latest navigation, with
accessible icon controls positioned at opposite edges of the message
surface while preserving existing follow and deep-link behavior.

**mobile/test/features/channels/channel_detail_page_test.dart**
Covers the unread target, compact inverse controls, accessible tooltips,
and placement beneath the frosted app bar.

</details>

## Reproduction steps

1. Open a Flutter mobile channel that has unread messages without
entering through a message or thread deep link.
2. Confirm an up-chevron appears directly below the channel app bar
while the timeline remains at latest.
3. Tap the up-chevron and confirm the timeline scrolls to the oldest
message that was unread when the channel opened.
4. Confirm the unread control is replaced by a down-chevron at the
bottom of the timeline.
5. Tap the down-chevron and confirm the timeline returns to latest and
resumes following new messages.

## Screenshots

| At latest — up-chevron to oldest unread | Away from latest —
down-chevron to latest |
|---|---|
| ![Up-chevron beneath the mobile channel app
bar](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/4239/buzz-mobile-scroll-to-oldest-unread.png)
| ![Down-chevron above the mobile channel
composer](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/4239/buzz-mobile-scroll-to-latest.png)
|

_Real iPhone 17 Pro Simulator captures from the neutral
`buzz-mobile-scroll-to` channel._

Originating Buzz thread:
`buzz://message?channel=5b16c478-22d8-4ddd-951a-6036e19b81ff&id=6a78af32d7ac6f531b182c4e70dd5a04c503a2dab2ce2c0c74b2c6baa5921741&thread=6a78af32d7ac6f531b182c4e70dd5a04c503a2dab2ce2c0c74b2c6baa5921741`

---------

Signed-off-by: Taylor Ho <[email protected]>
Signed-off-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

* feat(mobile): sync per-group channel sorting (#4231)

**Category:** improvement
**User Impact:** Mobile users can sort each channel group by recent
activity or A–Z, with their choices synchronized with desktop.
**Problem:** Desktop supports persistent per-group channel sorting, but
mobile shows the same groups without equivalent controls or shared
preferences. The earlier mobile attempt coupled sorting to unsafe
dirty-state behavior that could overwrite newer cross-client changes.
**Solution:** Add mobile sorting controls and encrypted NIP-78
synchronization using the existing desktop `channel-sort` contract,
while retaining ordinary whole-blob last-write-wins behavior. Local
state is scoped by identity and normalized relay, startup closes
fetch/subscription gaps, and both clients use the same deterministic
ordering rules.

<details>
<summary>File changes</summary>

**desktop/src/features/sidebar/lib/channelSortPreference.test.mjs**
Updates ordering coverage for the deterministic, cross-client A–Z
comparison rule.

**desktop/src/features/sidebar/lib/channelSortPreference.ts**
Aligns desktop channel-name collation with mobile so synchronized
preferences produce the same visible order.

**mobile/lib/features/channels/channel_sort/channel_sort_manager.dart**
Adds encrypted relay synchronization with safe startup gap handling,
clock checks, and ordinary last-write-wins conflicts.

**mobile/lib/features/channels/channel_sort/channel_sort_provider.dart**
Scopes sort state to the active identity and community lifecycle.

**mobile/lib/features/channels/channel_sort/channel_sort_storage.dart**
Defines the desktop-compatible payload, relay-scoped cache and
migration, cleanup, and shared ordering behavior.

**mobile/lib/features/channels/channels_page.dart**
Connects sort state to the channel page.

**mobile/lib/features/channels/channels_page/body.dart**
Applies each selected order to Starred, custom groups, Channels, and
DMs.

**mobile/lib/features/channels/channels_page/sections.dart**
Adds checked Recent and A–Z actions using the existing anchored-popover
UI.


**mobile/test/features/channels/channel_sort/channel_sort_manager_test.dart**
Covers payload adoption, encrypted publication, conflicts, timestamps,
retries, and cleanup.


**mobile/test/features/channels/channel_sort/channel_sort_storage_test.dart**
Covers parsing, relay isolation, migration, cleanup, and ordering modes.

**mobile/test/features/channels/channels_page_test.dart**
Verifies the group controls expose both choices.

</details>

### Reproduction steps

1. Open the mobile channel list with populated built-in and custom
groups.
2. Open a group menu and choose **Sort: Recent**; confirm active
channels move to the top.
3. Choose **Sort: A–Z**; confirm deterministic alphabetical ordering
returns.
4. Repeat for Starred, a custom group, Channels, and DMs.
5. Open desktop with the same identity and community and confirm each
synchronized preference.
6. Switch communities and confirm cached preferences do not bleed across
relays.

### Screenshots

Approved `live` custom-section flow with `research` kept offscreen.

| Recent selected | A–Z result | A–Z selected |
|---|---|---|
| ![live custom section with Recent
selected](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/4231/live-recent-selected.png)
| ![live custom section sorted
A–Z](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/4231/live-az-result.png)
| ![live custom section with A–Z
selected](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/4231/live-az-selected.png)
|

### Validation

- Mobile `flutter analyze` — clean
- Focused mobile sort and channel-page suites — 37/37 passed
- Desktop full suite — 3906/3906 passed
- Mobile full suite — 1034 passed, 1 skipped, 1 unrelated baseline
failure reproduced at `ac4fa13b8`

<!-- Originating Buzz channel: 2a16a2bb-6fd3-4d69-8182-2afcb21b2d14 -->

---------

Signed-off-by: Taylor Ho <[email protected]>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

* feat: ship Buzz Term (#4347)

## Summary

- ship **Buzz Term** end to end: the terminal engine/runtime, mounted
desktop substrate, and user-visible naming
- add Quinn's tape-deck-inspired banner: a beveled chassis filled by the
`buzz term` wordmark, surrounded by a complete-hex field
- derive the wordmark's three-stop sweep from each theme's terminal
palette so primary, secondary, and accent roles remain visibly distinct
across all 62 shipped themes, including light themes
- paint the banner once on its own pointer-transparent canvas; PTY
rendering beneath it remains unchanged

## Banner behavior

- uses the renderer's shared `8.4 × 17` cell metrics and production
aspect ratio `2.0238`
- regenerates only for viewport/theme changes; palette switches repaint
correctly while the banner is visible
- dismisses on non-empty output from the active terminal session; empty
output and inactive sessions do not dismiss it
- fails closed below **70 columns** rather than squeezing or clipping
the wordmark
- adds **8 lines** to `terminalRenderer.ts` for shared cell metrics and
**zero lines inside `paint()`**

## Screenshots

| Buzz (light) | Buzz Dark |
|---|---|
| ![Buzz Term — Buzz
light](https://buzz.block.builderlab.xyz/media/fe4c50c1cd03645bff8f3cff588353618fd3004f320ab360165b6cb48f041eb6.png)
| ![Buzz Term — Buzz
Dark](https://buzz.block.builderlab.xyz/media/e5731f40fe020070c0b327287b4b0b6f0103adde852d0f6451ad66a0ff9d14bb.png)
|

| Kanagawa Lotus (light) | Red |
|---|---|
| ![Buzz Term — Kanagawa
Lotus](https://buzz.block.builderlab.xyz/media/260b211fe7bd232eb7faa4ec34c99baee5fed380556427c64495ceed3396a8c5.png)
| ![Buzz Term —
Red](https://buzz.block.builderlab.xyz/media/42a90830824683a0054e24649c58564134af831607736d823992181b792087d3.png)
|

Additional production-aspect finals:
[Vesper](https://buzz.block.builderlab.xyz/media/9ca6514b63f8cfb2107a85ca46f16a940c0883848e6fbc718e411af94aa13100.png),
[Min
Dark](https://buzz.block.builderlab.xyz/media/f67bd2970e5d64ffb07b1ae78ab58c847e6ebc23e7e7a48e067eb024dba64ec8.png),
and [Dark
Plus](https://buzz.block.builderlab.xyz/media/290fee08924f37d064abc687ecf3e9526ab05b87e8e56d610f23048949793dbe.png).

The screenshot harness was checked against the shipped painter at this
exact head: all **2,541 draw calls** matched on color, glyph, x, and y;
four deliberate divergence controls fired.

## Verification at `98ebc8f9048bd5f0ceb7e843b67874d642f0b7fd`

- desktop tests: **3,946 / 3,946**
- TypeScript: clean
- checks: pass (two pre-existing informational `useTemplate` notices
only)
- integration/e2e: PASS (independent exact-SHA lane; artifacts recorded
in the originating Buzz thread)
- artifact/dead-path sweep: clean
- redteam G1–G7: PASS
  - all six named banner emitter-deletion mutants die
- independent handwritten five-row full-wordmark fixture kills Quinn's
seven-mutant battery, including a one-pixel glyph change
- real `112 × 46` canvas-rect dismissal tests separately cover active
non-empty, active empty, and inactive non-empty output
- layer-drop and zero-draw painter mutants die; z-order and
pointer-events verified
  - CI's `tsc && vite build` includes all three banner modules
- performance at DPR 2 (worst-case measured envelope):
- one-time content paint: **~0.7–0.8 ms**, paid only when the banner is
built or its palette changes
- busy compositor, CSS `1277 × 697`, backing `2554 × 1394`: **470–497
µs/frame** for the full banner (**2.82–2.98%** of a 60 Hz frame)
- busy compositor, CSS `1920 × 1080`, backing `3840 × 2160`:
**1,139–1,212 µs/frame** (**6.83–7.27%**)
- empty, one-glyph, and full-banner controls converge: compositor cost
follows backing-layer area and DPR rather than painted-cell count
- in the actual idle welcome state, cost is below both vsync-clamped
rigs' resolution; it is not claimed as zero
- **Pane cross-rig spread: resolved at matched loop rate.** Two
independent rigs initially differed 2.3× (58–68 vs 136 µs/Mpx of backing
store; pane, CSS 1277×697 / backing 2554×1394, DPR 2). The cause of
*that* spread is rAF loop rate: the higher figure came from a
free-running loop at ~1600fps. Throttled to ~200–236fps, both rigs read
58–68 µs/Mpx (1.25–1.44% of a 60Hz frame). The busy-composite figures
quoted above remain the **unthrottled worst case** and are conservative
by ~2.3× at the pane. Not established: the mechanism and sign of
free-running distortion (one rig under-charges ~15%, the other
over-charges 2.3×), and the 1080p figure has not been re-measured
throttled.
- the layer paints only on generation/theme/resize and dismisses on
first non-empty active-session output, so the measurable busy cost is a
short-lived worst case rather than a persistent PTY paint-path tax

## Follow-ups in this PR

These are intentionally subsequent commits after the certified
static-banner head, not claims about `98ebc8f90`:

1. close the compositor metrology: remeasure the 1080p point throttled
and characterize the opposite-sign free-running rAF distortion, with
each measurement regime stated
2. add Tyler's animated honeycomb color waves, gated by
`prefers-reduced-motion`, a full 62-theme phase-sweep contrast check,
and DPR-2 per-tick performance certification
3. land the already-proven mounted theme-switch regression probe from
`RESEARCH/BUZZ_TERM_G3A_PROBE/`
4. bound the slow/hang-shaped G1-c mutant `waitFor`
5. optionally trim the generator to its ink bounding box, reducing the
minimum viewport from 70 to 62 columns

---------

Signed-off-by: tlongwell-block <[email protected]>
Signed-off-by: npub1mprnacetjua2xx3p5eddmhxyk6wv929ymm5py8kd2xfxurxahspqqlgyta <d8473ee32b973aa31a21a65adddcc4b69cc2a8a4dee8121ecd51926e0cddbc02@buzz.block.builderlab.xyz>
Signed-off-by: npub1jh9wn95s0472h86ahapupaf7m6kx4v9sx2n0atj2hltcfer8k06s5n3pyf <95cae996907d7cab9f5dbf43c0f53edeac6ab0b032a6feae4abfd784e467b3f5@buzz.block.builderlab.xyz>
Signed-off-by: npub17jjz49l9jjmhhk7cac63j8yt9z555n9cw8vk7v5jz4vzw4ppld5qgj57cc <f4a42a97e594b77bdbd8ee35191c8b28a94a4cb871d96f32921558275421fb68@buzz.block.builderlab.xyz>
Signed-off-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Co-authored-by: Dawn (sprout agent) <c6237ef84fa537c78dcee78efd2d4e59f728859c7f194da42ac51ededfa0be05@sprout-oss.stage.blox.sqprod.co>
Co-authored-by: npub1t2tgm7d8f995uqvmnm8h88sg3wnpp9a5xysjf6dg3tjmgt3ltulqdp8ehr <5a968df9a7494b4e019b9ecf739e088ba61097b4312124e9a88ae5b42e3f5f3e@buzz.block.builderlab.xyz>
Co-authored-by: npub1cc3ha7z055mu0rwwu7806t2wt8mj3pvu0uv5mfp2c50dahaqhczshdalg6 <c6237ef84fa537c78dcee78efd2d4e59f728859c7f194da42ac51ededfa0be05@buzz.block.builderlab.xyz>
Co-authored-by: npub17jjz49l9jjmhhk7cac63j8yt9z555n9cw8vk7v5jz4vzw4ppld5qgj57cc <f4a42a97e594b77bdbd8ee35191c8b28a94a4cb871d96f32921558275421fb68@buzz.block.builderlab.xyz>
Co-authored-by: npub1jh9wn95s0472h86ahapupaf7m6kx4v9sx2n0atj2hltcfer8k06s5n3pyf <95cae996907d7cab9f5dbf43c0f53edeac6ab0b032a6feae4abfd784e467b3f5@buzz.block.builderlab.xyz>
Co-authored-by: npub1mprnacetjua2xx3p5eddmhxyk6wv929ymm5py8kd2xfxurxahspqqlgyta <d8473ee32b973aa31a21a65adddcc4b69cc2a8a4dee8121ecd51926e0cddbc02@buzz.block.builderlab.xyz>
Co-authored-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>

* Polish mobile inbox and media flows (#4512)

## Summary

- make mobile unread state visible with bold channel names, an animated
Inbox badge, and swipe-to-toggle Inbox rows
- add directional transitions for top-level mobile navigation
- let mobile send while media uploads, with cancellable progress UI
- normalize iOS and Android video uploads, attach poster frames, and
improve native video playback

## Validation

- `just mobile-check`
- `just mobile-test`
- `cargo test -p buzz-media`
- Pixel smoke test
- iPhone smoke test

Desktop background uploads moved to #4522 so the two platforms can be
reviewed independently.

---------

Signed-off-by: kenny lopez <[email protected]>
Signed-off-by: Tom Brow <[email protected]>
Co-authored-by: leader <71e9f2c44a6932b6772caaaccda1911d010463c3e2c6c40410b8329956046801@buzz.block.builderlab.xyz>
Co-authored-by: Tom Brow <[email protected]>

* fix(desktop): stop clipping focus ring on channel intro action cards (#2392) (#4374)

## What

Fixes #2392 — the action cards in the empty-channel intro ("Create
agent", "Add people") had their `focus-visible` ring clipped by the
surrounding scroll container.

## Root cause

The cards sit in a `flex ... overflow-x-auto pb-1` row. Setting
`overflow-x` (without `overflow-y`) makes the browser compute
`overflow-y: auto` as well, so the container clips anything painted
outside its padding box — including the cards' `focus-visible:ring-2`
box-shadow. With only `pb-1` padding, the top/left/right of the ring
were cut off when Tabbing to a card.

## Change

`desktop/src/features/messages/ui/ChannelIntroBlock.tsx` — `pb-1` →
`p-1` on the action-cards scroll container, reserving 4px on all four
sides so the focus ring renders fully inside the scroll container's
padding box.

- 1 file, 1 line. No behavior change for mouse users or layout.

## Verification

- `pnpm typecheck` — clean
- `pnpm exec biome check src/features/messages/ui/ChannelIntroBlock.tsx`
— clean
- `pnpm check:file-sizes` — clean
- Desktop unit suite — **3906/3906 pass**

Signed-off-by: Sarthak Singh <[email protected]>

Signed-off-by: Sarthak Singh <[email protected]>

* Defer desktop media uploads until send (#4522)

## Summary

- send desktop messages immediately while media uploads continue in
background state across channel navigation
- show immediate progress above the composer and keep Jump to latest
above it
- report the real media stages as Preparing, Processing, Converting,
Uploading, and Finishing
- use Buzz's shared spinner during local media work, then switch to the
real percentage when byte transfer begins
- animate phase-label and status-suffix changes without overlap or
layout jumps
- keep cancel, progress fill, message publication, and community-reset
behavior coordinated with the background task
- use raw Tauri IPC for large browser files so renderer-side byte
serialization does not block initial feedback

## Why

Desktop previously blocked sending while attachments uploaded in the
composer. Large videos could also pause the renderer before progress
appeared, and the progress pill said Uploading while native media
processing was still underway. This makes the initial response immediate
and describes the work actually happening.

## Validation

- `cd desktop && pnpm check`
- `cd desktop && pnpm typecheck`
- `cd desktop && pnpm test` (3,931 passed)
- `cd desktop && pnpm exec vite build --mode e2e`
- `cd desktop && pnpm exec playwright test
tests/e2e/file-attachment.spec.ts --project=smoke` (11 passed)
- focused native media tests (80 passed)
- native Clippy with all targets and features
- pre-push native suite (2,107 passed, 14 ignored; 3 diagnostics passed)

Updated phase snapshots are included in the PR comments.

Split from #4512 so the desktop and mobile changes can be reviewed
independently.

---------

Signed-off-by: kenny lopez <[email protected]>
Signed-off-by: Wes <[email protected]>
Co-authored-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

* Refine desktop timeline activity presentation (#4582)

## Summary

- Make channel join/leave activity use the selected inline avatar-stack
treatment.
- Group related membership activity for one hour and preserve
profile/overflow-name interactions.
- Restore the virtualized day-divider handoff and align the sticky date
behavior with the message timeline.

## Validation

- `pnpm check`
- `pnpm test`
- `cargo test --manifest-path desktop/src-tauri/Cargo.toml`
- Visual desktop screenshot captured with seeded membership activity

---------

Signed-off-by: kenny lopez <[email protected]>

* Remove blur from Welcome composer guidance (#4691)

## Summary
- Keep the Welcome composer prompt above the dock blur so it stays
readable.
- Remove blur from the prompt and persona-motion paths.
- Cover the crisp, correctly layered banner in the onboarding browser
test.

## Validation
- `pnpm -C desktop exec biome check
src/features/channels/ui/WelcomeComposerBanner.tsx
tests/e2e/onboarding.spec.ts`
- `pnpm -C desktop build:e2e`
- `pnpm -C desktop exec playwright test tests/e2e/onboarding.spec.ts
--grep "finishing onboarding creates starter channels and focuses
welcome-everyone for a new member" --project=integration`

---------

Signed-off-by: kenny lopez <[email protected]>

* docs(nip-am): normative amendment — cache SHOULD/MUST + pricingIdentity + consumer cost guidance (#4632)

Amends `docs/nips/NIP-AM.md` with three normative publisher-behavior
changes per the cleared Usage v2 plan (plan v3, D4 + D2').

## Changes

### 1. Cache emission semantics (D4)

Replaces the unconditional `MAY` with qualified obligations:

- Publishers SHOULD emit `cacheReadTokens` / `cacheWriteTokens` when the
provider exposes a cache component.
- Publishers MUST preserve an explicit zero when the provider reports
zero.
- Publishers MUST omit the field (never null or fabricated zero) when
that component is unavailable to the publisher — including when the
provider supports it but the harness does not surface it.

An explicit carve-out in both the JSON comment block and the
Numeric-validity prose exempts these fields from the payload-wide null
guidance. Omission is the only valid representation for an unavailable
cache component.

### 2. Optional `pricingIdentity` field (D2')

Adds an optional, non-nullable `pricingIdentity` object (`authority`,
`model`, `cacheClass`), defined as billing authority — distinct from the
transport `Provider` enum.

- `authority` is a registered billing-namespace identifier: exact
lowercase hostname, no scheme, no path, no trailing slash. Registered
values: `api.anthropic.com`, `api.openai.com`, `openrouter.ai`. The set
extends only by NIP amendment. Pricing lookup is an exact string match
on `(authority, model)`.
- Present only when the publisher can prove applicability: direct
official-endpoint connections prove via the actually-requested resolved
model; other routes MUST receive response-supplied authoritative billing
identity.
- MUST omit for custom/overridden base URLs, gateways (unless the
gateway is the named billing authority), unresolved aliases, and turns
where usage contributions carry more than one billing identity
(including identity-bearing mixed with unresolved).
- `cacheClass` is omitted (not null) when not applicable.
- `pricingIdentity` is optional but not nullable — omission is the only
absence representation.
- The existing `model` field retains its non-billing semantics
(configured/session model) and is never overloaded.
- Consumers MUST treat omission as "price unknown" and MUST NOT infer a
price from the session `model` field.

### 3. Consumer cost guidance (D4)

- Consumers MAY recompute cost estimates using the billing identity and
a pricing manifest.
- Consumers MUST retain the provenance of any cost value (e.g.
`manifest-estimated`, `wire-reported`).
- Consumers MUST NOT merge manifest-estimated and wire-reported costs
into an unlabeled total.

Manifest-vs-wire display preference is application policy and
deliberately excluded from this NIP.

## Scope

Doc-only. Single file: `docs/nips/NIP-AM.md`.

---------

Signed-off-by: Will Pfleger <[email protected]>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>

* docs(acp): explain per-channel session model in base prompt (#4729)

## Overview

Agents running in Buzz have no built-in awareness that each channel is
an isolated conversation context. When a human mentions work "you" are
doing in another channel, the current session can misread this as its
own active context and try to coordinate, re-plan, or take ownership of
it — causing confusion and wasted turns.

## What changed

Added a `## Session Model` section to
`crates/buzz-acp/src/base_prompt.md`, inserted immediately after the
opening paragraph and before `## Buzz CLI`. The section explains:

- Each channel is a separate session; multiple sessions of the same
agent identity may be active simultaneously.
- Sessions share core memory, workspace, and relay — but not
conversation context or in-flight reasoning.
- Cross-channel work belongs to the owning session by default; the
current session may take it over only when the human explicitly requests
it.

No runtime code changes. Base prompt only.

---------

Signed-off-by: Will Pfleger <[email protected]>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>

* fix(desktop): show cached display names on startup (#3317)

## Why

Buzz restores cached channels and messages before profile lookups
complete. On launch, that briefly exposes pubkey-derived labels in place
of familiar display names.

## What

- Persist a bounded, relay-scoped cache of last-known display names,
NIP-01 names, and NIP-05 handles
- Seed batch profile queries from those labels immediately, while
keeping them stale so the existing relay request revalidates them
- Keep cached data presentation-only: avatars and ownership metadata are
not persisted or used to seed profile-detail caches
- Remove cleared or missing profiles, purge a relay's labels when its
community is removed, and include the cache in local-storage quota
recovery
- Add unit coverage for parsing, bounds, eviction, malformed data, and
cleared profiles
- Add an E2E regression that delays the relay profile response and
verifies the cached name is rendered first

## Risk Assessment

Low. The cache is disposable, capped at 1,000 entries per relay, scoped
by normalized relay URL, and always revalidated. It contains only public
label fields and does not restore avatars, agent ownership, or
authorization state.

## Verification

- `just ci`
- `pnpm typecheck`
- `pnpm test` — 3,727 passed
- `pnpm exec playwright test tests/e2e/channels.spec.ts --grep "cached
profile labels"` — passed

Generated with Codex

* Polish sidebar unread hierarchy (#4573)

## Summary

- Keep selected sidebar rows regular by default; manually unread rows
become bold immediately.
- Apply a clearer dark-mode hierarchy: standard inactive rows at 75%,
muted rows at 45%, and unread rows at full emphasis.
- Keep hover text color stable while retaining the selected-row and
unread cues.

## Validation

- `pnpm typecheck`
- `pnpm build:e2e`
- Playwright: sidebar badge and channel-mute coverage

## Screenshots

Posted in the PR comments.

---------

Signed-off-by: kenny lopez <[email protected]>

* feat(desktop): surface config diff in restart-required badge (#3637)

The "Restart required" badge reports that an agent's running config has
drifted from its spawn-time config, but never says what changed. This
ships the full feature: a typed Rust diff engine and a TS/UI layer that
renders it at every badge site.

## Rust core (spawn-snapshot diff engine)

Replaces the lossy `u64` `spawn_config_hash` with a typed
`SpawnConfigSnapshot`. The snapshot is stamped from the already-resolved
command/env/config values immediately before `spawn()`, closing the race
window where a mid-spawn config edit would suppress the badge.

`SpawnConfigSnapshot::canonical()` is the single JSON projection shared
by the badge and the diff. Drift is `to_value(stamped) !=
to_value(current)`; the diff is a generic leaf walk over those same two
values, so badge-on and diff-non-empty are structurally guaranteed.
Adding a snapshot field reaches the UI with no code change to the diff
engine — `mutation_table_covers_every_serialized_field` fails CI if a
new field arrives without a mutation row.

`eligible_restart_diff(persona_orphaned, Option<TrackedSpawnState>)`
returns the final vector — snapshot walk entries plus a synthetic
`adapter_availability` entry. It returns empty for an orphaned instance
(spawning one would fail) and for agents with no tracked spawn state
(never stamped, can never have drifted). `needs_restart =
!restart_diff.is_empty()` derives from that vector and nothing else.

Redaction policy (`policy_for(path)`) is shared by the wire diff and the
snapshot's manual `Debug` via `is_safe_to_reveal()` from
`managed_agents::env_vars` as the single authority for env-key masking:

| Policy | Paths | Rendering |
|---|---|---|
| `Text` | `system_prompt`, `team_instructions` | character counts only
|
| `MaskedBare` | `args`, `relay_url` | `••••`, no suffix |
| `MaskedSuffix` | non-allowlisted `env.*` | `••••` + last 4 chars when
longer than 8 |
| `Plain` | allowlisted `env.*` (`BUZZ_AGENT_THINKING_EFFORT`,
`BUZZ_AGENT_PROVIDER`, `BUZZ_AGENT_MODEL`, `DATABRICKS_HOST/MODEL`) and
everything else | verbatim |

Default-deny: every env key not in the explicit allowlist stays masked.
`is_safe_to_reveal()` is the single allowlist authority for both the
baked-env display and the diff.

`restart_diff` is omitted from the wire when empty
(`skip_serializing_if`).

## TypeScript / UI layer

New `restartDiff.ts` module defines `RestartDiffEntry`, `RestartChange`,
`JsonValue`; `tauri.ts` and `types.ts` re-export and add `restart_diff`
/ `restartDiff` fields (Rust omission → `restartDiff: []`).

**`RestartDiffBadge`** — hover tooltip capped at 6 entries + "and N
more", `asChild` span trigger (never inside a `<button>`), auto-restart
blurb below the diff list (on/off variant from `autoRestartEnabled`
prop; same `AUTO_RESTART_ON_BLURB` / `AUTO_RESTART_OFF_BLURB` constants
shared with the Runtime-tab banner). **`RestartDiffList`** renders the
full uncapped list for the Runtime-tab banner with `tooltip`/`inline`
presentation variants for correct foreground in both surfaces.

**`ManagedAgentRow` B4 fix** — badge moved to a sibling `div` of the row
expansion button; tooltip trigger has no `button` ancestor.

**`UnifiedAgentsSection`** — both badge sites render
`<RestartDiffBadge>` instead of a raw `<Badge>`, with
`autoRestartEnabled` threaded from `agent.autoRestartOnConfigChange`.

**Side-panel fix** — `RestartDiffBadge` rendered tab-independently in
the `ProfileSummaryView` hero area (was Runtime-tab only — root cause of
the ~50% inconsistency Will reported). Hero badge is `self-center` in
the flex column. `ProfileRuntimeTabContent` early-return checks
`needsRestart` so the banner is never dropped when all other content is
empty. Auto-restart blurb in the Runtime-tab banner uses the shared
constants.

## Wire shape

```jsonc
"restart_diff": [
  { "field": "model",              "change": { "kind": "value",  "before": "gpt-5", "after": "claude-4" } },
  { "field": "system_prompt",      "change": { "kind": "text",   "before_chars": 1234, "after_chars": 1410 } },
  { "field": "env.OPENAI_API_KEY", "change": { "kind": "masked", "before": "••••bc12", "after": "••••xyz9" } },
  { "field": "env.BUZZ_AGENT_THINKING_EFFORT", "change": { "kind": "value", "before": "medium", "after": "high" } }
]
```

`added`/`removed` occur only for dynamic-map keys; nullable struct
fields always serialize as `null`; arrays are atomic leaves (`args`,
never `args.0`).

## Tests

**Rust** — 1902 passing: snapshot mutation coverage, diff entry
serialization, allowlist-aware env masking
(`allowlisted_env_key_shows_plain_value`,
`allowlisted_env_key_is_case_insensitive`,
`non_allowlisted_env_key_stays_masked`),
`unstamped_agent_yields_no_badge_and_no_entries` (both orphan values),
`summary_without_drift_omits_restart_diff_from_the_wire`,
`unstamped_availability_is_not_drift`. Clippy clean, fmt clean.

**TypeScript** — `needs-restart-screenshots.spec.ts`: 11 E2E cases
registered in the smoke project — all three badge sites, tooltip +
keyboard focus, DOM no-button-ancestor assertion, 6+1 truncation,
uncapped Runtime list, unknown field humanisation, side-panel badge on
default Info tab, inactive/friendly-error Runtime opening path.

Consolidates [#3652](https://github.com/block/buzz/pull/3652)

Signed-off-by: Will Pfleger <[email protected]>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>

* feat(desktop): persist sidebar observed-unread across webview reload (#3976)

## Problem

`Command+R` (webview reload) wipes the two in-memory refs driving
sidebar channel unread badges: `observedUnreadEventsByChannelRef` and
`latestByChannelRef`. The boot catch-up REQ can only fetch events newer
than each channel's NIP-RS frontier, so thread replies that arrived
before the frontier was passively advanced (the common case) are never
re-discovered.

Inbox is unaffected because it rebuilds candidates from a relay feed
query and checks fine-grained `thread:`/`msg:` markers. The sidebar
badge path lacks an equivalent recovery mechanism.

## Solution

Persist the sidebar's per-event candidate set to localStorage as a
disposable, versioned projection cache
(`buzz-observed-unread.v1:<relay>:<pubkey>`) and hydrate it on boot
before the catch-up REQ runs.

### New files

**`observedUnreadStorage.ts`** — storage module for the cache:
- Keyed
`buzz-observed-unread.v1:<normalizedRelayUrl>:<normalizedPubkey>`
(relay-scoped to prevent cross-community leakage, matching
`threadActivityStorage`)
- Stores validated per-event `ObservedUnreadEvent` rows;
`latestByChannel` is derived at hydration — no divergent dual aggregate
- Age pruning (7d = `READ_STATE_HORIZON_SECONDS`), per-channel cap
(1000), global cap (5000) across all channels in a scope bucket
- Payload `updatedAt` for LRU ordering; registered in
`PURE_CACHE_KEY_PREFIXES` for 2 MiB eviction budget
- Field-level validation on decode; write failure is non-fatal
(session-only degradation)
- Snapshot-owning timers: `scheduleObservedUnreadWrite` deep-clones the
events map at schedule time — a late A-scope timer can never read B's
mutable refs or write under B's key

**`useObservedUnreadPersistence.ts`** — hook that owns all persistence
lifecycle:
- Scope fence: `normalized pubkey + normalized relay` identity;
`isScopeLoaded()` callback guards both projection (`rawUnread`) and
every **observed-cache mutation** (`recordUnreadEvent`, `removeChannel`,
`clearAll`) before touching refs or storage. Note: stale-scope calls to
`markChannelRead`/`markAllChannelsRead` can still affect
`forcedUnreadRef` and NIP-RS markers, which are pre-existing on `main`
and deferred to the NIP-RS arc (see Deferred below).
- Synchronous `pagehide` flush closes the Cmd+R timing gap
(`useReloadShortcut.ts` reloads within 500ms of teardown, before the
1-second debounce fires)
- Identity-reset effect: flushes old scope, resets refs, hydrates from
storage, stamps loaded scope — all atomic; cleanup flushes on unmount
- `clearAll` cancels the pending timer, resets both in-memory refs, and
clears storage in a single transactional operation; `removeChannel`
deletes the channel from both refs and replaces any pending snapshot
with the current full map — never cancel-without-replacement, preserving
sibling-channel events on reload
- Marker-prune effect on `readStateVersion`: evaluates each retained
event with `observedUnreadEventReadAt()` (the same evaluator used by the
projection memo) and removes covered events, rederiving per-channel
latest — never clears a whole channel for a single thread/msg marker
- Returns a stable `useMemo`-wrapped API object keyed on actual deps so
unrelated re-renders do not restart the catch-up REQ
- `isScopeLoaded` is a `useCallback` (not a memoized boolean) — always
reads the ref at call time, never stale

### Modified files

**`useUnreadChannels.ts`** — hook integration:
- Calls `useObservedUnreadPersistence` with all persistence wired
through the returned API
- `rawUnread`: `isScopeLoaded()` guard suppresses A-scope refs from
projecting under B
- `recordUnreadEvent`: `isScopeLoaded()` fence before touching refs;
schedules a debounced write on each successful record
- `markChannelRead` clearObserved path: calls `removeChannel` so the
cleared state survives reload
- `markAllChannelsRead`: delegates to the owner's fenced `clearAll` —
the parent does not reset the observed refs directly; `clearAll` owns
the transactional clear of both refs and storage, preventing a stale
scope-A callback from corrupting scope B

**`localStorageQuota.ts`** — registers `buzz-observed-unread.v1:` in
`PURE_CACHE_KEY_PREFIXES`

## Design constraints

The cache is a **disposable projection**: versioned key, read-through
only, safe to delete wholesale. It does not touch `ReadStateManager`,
marker semantics, or `forcedUnreadStore`. Zero overlap with the NIP-RS
manual mark-read/unread protocol work in progress in another channel;
migration path when that lands is "stop reading the key."

## Test coverage

**`observedUnreadStorage.test.mjs`** covers storage primitives:
- Key normalization, relay-scoped isolation, round-trip correctness
- Age-prune and per-channel cap on read and write; global cap across
channels
- `deriveLatestByChannel` correctness
- Thread-marker prune leaves sibling thread events persisted and lit
- Scope-isolation state machine: A rows visible in A, absent in B,
restored on A again; late A-scope write does not overwrite B's bucket
- Malformed structures/fields, relay/pubkey isolation, quota failure
degradation

**`useObservedUnreadPersistence.test.mjs`** exercises the real hook via
`createRoot` + `act`:
- pagehide flush: event recorded within debounce window survives reload
(headline regression)
- Unmount with pending write flushes before teardown
- `clearAll` cancels pending debounce so no resurrection after reload
- `removeChannel` replaces pending snapshot so sibling channel B
survives reload (two-channel repro)
- Marker prune: thread and channel markers prune covered events; sibling
channels survive
- `isScopeLoaded` returns false before identity-reset effect commits,
true after
- A→B scope switch: pending A-timer is cancelled by flush, A data
persisted synchronously (hydration round-trip)
- Stale `clearAll` from scope A rejects after scope B loads
(observed-cache scope fence)
- Stale `removeChannel` from scope A rejects after scope B loads
(observed-cache scope fence)
- API object identity stable across unrelated re-renders (catch-up
stability)

**`useUnreadChannels.test.mjs`** exercises the full parent-to-owner seam
with real hook mounts:
- Stale `markChannelRead` from scope A does not corrupt B's observed
bucket after flush
- Stale `markAllChannelsRead` from scope A does not overwrite B's bucket
after flush

## Deferred

Issues deferred to the NIP-RS arc (`#unread-messages-ux`) or future
hardening — not regressions introduced by this PR:

- **Stale-scope `forcedUnreadRef` / `markContextRead` exposure**: a
stale scope-A `markChannelRead` or `markAllChannelsRead` still deletes
B's `forcedUnreadRef` entries and advances B's NIP-RS markers via
`markContextRead` before the observed-cache fence rejects. This is
pre-existing on `origin/main` (identical shape at lines 316/330). Fix
requires touching `forcedUnreadStore` and marker paths — out of scope
for Fix A. Deferred to the NIP-RS work.
- **`isScopeLoaded` empty-scope hardening**: `isScopeLoaded()` returns
`true` when `pubkey` and `relay` are empty strings (no active session).
A guard could assert non-empty identity before stamping scope-loaded.
Low risk in practice since the hook is only mounted after auth, but
could be tightened.
- **Catch-up batch scheduling**: `handleChannelMessage` and the catch-up
loop each clone the full events map per event via
`scheduleObservedUnreadWrite`. For channels with large backlogs this
produces O(n) snapshot clones per catch-up batch. A batch-schedule API
(single snapshot at end of batch) would reduce allocations. Not
observable in normal use; deferred as a performance optimization.

---------

Signed-off-by: Will Pfleger <[email protected]>
Signed-off-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>

* Refine community invite links (#4734)

## Summary

- Separate direct invites from link sharing with a labeled divider.
- Show the generated invite URL inline with truncation and a copy
control.
- Use shared loading feedback and a restrained copy-status resize.

## Validation

- `pnpm -C desktop exec playwright test
tests/e2e/invite-link-copy.spec.ts
tests/e2e/invites-settings-screenshots.spec.ts` (7 passed)

---------

Signed-off-by: kenny lopez <[email protected]>

* fix(agents): canonicalize stale p…

* fix(mobile): recover stale relay sessions (#4372)

### Summary

Fixes [this
issue](buzz://message?channel=e62570dd-33ad-42c5-b92b-75f2689f9694&id=b726c366abfe62429ee3cdcd34d0c0fb98c33c3ea053480585bed71745412b56):
> I often don’t see my bot responses until after I post. they’re usually
time stamped correctly so I think it’s just a refresh issue?

### What changed?

Buzz Mobile now reconnects relay sessions after the app has remained
backgrounded beyond the existing 5-second grace period, even when the
session still reports a stale `connected` state. This makes resume
recovery independent of whether iOS runs the grace timer before or after
delivering `resumed`.

Reconnection is now based on elapsed background time rather than a
direct socket-health probe.
- If the app was backgrounded for at least the 5-second grace period,
the socket is presumed dead and the session reconnects regardless of
reported status.
- If it was backgrounded for less than that, a reported `connected`
status is still trusted.

In the sub-5-second window the socket is either genuinely alive, which
is the common case for a momentary background, or it is dead and the
client ping detects it within the two-interval worst case described
below. That is now a degraded-latency path, not a silent-forever path.

The mobile relay socket now uses `IOWebSocketChannel.connect` with a
30-second `pingInterval`. An unanswered ping closes the Dart socket
through the existing disconnect and reconnect path.

Detection takes up to two ping intervals, so about 60 seconds worst
case, not 30. One interval of idleness elapses and a ping is sent, then
a second interval elapses with no pong and the socket closes. Any
inbound pong restarts the first stage, so the clock measures idleness
rather than running on a fixed cadence.

### Why?

Buzz iOS can sometimes stop showing new bot or agent responses after a
phone has been locked for 5 to 10 minutes. When the user later posts a
message, the missing responses can appear all at once. iOS may suspend
Buzz before the short delayed cleanup that would normally close its
connection has a chance to run. Before this change, Buzz trusted the
resulting stale healthy status on resume and skipped reconnecting, so
the missing responses stayed hidden until a later post exposed the dead
connection.

A state-machine test with a stubbed connection reproduced this reported
pattern and showed that it matches this failure mode: the failed post
triggered a reconnect that fetched the missing messages. The same test
also checked the other candidate explanation, the bug tracked in
[#3053](https://github.com/block/buzz/pull/3053), where the relay has
closed the app's subscription. That state does not produce the pattern.
Posting succeeds and the user's own message appears, but nothing looks
for the missed messages, so they stay hidden. The test confirmed that
the missed messages were still available to fetch in that state, so the
missing step was a trigger to fetch them. This was not an end-to-end
reproduction on an iOS device or a live relay.

The new resume check covers the normal lock and unlock path. If the app
was backgrounded for less than the 5-second grace period, it still
trusts a connection marked as healthy. A dead connection in that window
is instead detected by the ping check, which can take up to about 60
seconds but prevents the app from remaining silently stuck. The ping
only runs while iOS is running the app, so it does not detect a
connection that died during suspension; the resume check owns the lock
and unlock path.

A pre-existing path also runs the same resume handling when network
connectivity returns while the app is already in the foreground. Because
the app was not backgrounded, this change does not alter that path,
which still trusts a connection marked as healthy and relies on the
slower ping check.

Recovery from a subscription that the relay explicitly closes remains in
[#3053](https://github.com/block/buzz/pull/3053), and the two changes
overlap in one file. Changes to how missed messages are backfilled or
replayed are out of scope.

### How is it tested?

Full mobile suite at base and head. Both runs have the same known
macOS-host-only failure in `ChannelDetailPage keeps follow mode off
while a tall newest message stays visible` at line 1053:

- Base: 1,021 passed, 1 skipped, 1 failed
- Head: 1,025 passed, 1 skipped, 1 failed

Added tests:

-
[`relay_session_test.dart`](https://github.com/block/buzz/tree/main/mobile/test/shared/relay/relay_session_test.dart):
long-background resume reconnect and within-grace control
-
[`relay_socket_liveness_test.dart`](https://github.com/block/buzz/tree/main/mobile/test/shared/relay/relay_socket_liveness_test.dart):
silent-peer disconnect and idle-but-healthy control

Mutation checks confirm that removing elapsed-background resume recovery
fails with one socket instead of two, and removing `pingInterval` leaves
the silent peer connected. Restored production code passes both
mutations' regression tests and the healthy idle control.

Signed-off-by: Tom Brow <[email protected]>
Co-authored-by: npub1tquskdu6yc4h8l7xxtceculxw600grekeq0xg2ukqfrwl7vrzg3quz3gmp <58390b379a262b73ffc632f19c73e6769ef40f36c81e642b960246eff9831222@buzz.block.builderlab.xyz>

* ci: add guarded desktop release cache prewarm (#4575)

## Summary

Gate 1 only for desktop release caching:

- replaces canary `rust-cache` use with explicit exact-key
`actions/cache/restore` + `save`
- computes keys after `cargo update --workspace`, including platform,
target, Rust toolchain, Cargo manifests/locks, profile/features, and
native-toolchain inputs
- normalizes only the desktop package version so a trusted `main` canary
can warm an otherwise identical release tag
- excludes Tauri bundle directories, so installers and signed artifacts
are never cached
- adds a restore-only `cache-proof-*` tag workflow that fails unless tag
scope sees the exact default-branch cache
- adds contract tests that enforce no release-workflow cache change in
Gate 1

`release.yml` is intentionally unchanged. A cache miss remains the
current cold canary build; the release path cannot be affected by
merging this PR.

## Validation

- `scripts/test-desktop-release-cache-key.sh`
- `scripts/test-desktop-release-cache-workflow.sh`
- `scripts/test-release-ref-contract.sh`
- Ruby YAML parse of all four changed workflows
- `git diff --check`
- pre-push `branch-skew`

## Post-merge proof plan

1. Run each canary cold on trusted `main`, recording cache size/save
time and fresh artifact inventory.
2. Run each canary warm, requiring the exact-key hit and recording
restore/build time.
3. Create a disposable `cache-proof-*` tag at that same trusted `main`
SHA and dispatch **Desktop release cache tag-scope proof** from the tag.
4. Do not begin Gate 2 or modify `release.yml` unless the exact
tag-scope restore succeeds and cache transfer economics are favorable.

---------

Signed-off-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

* feat(desktop): make onboarding model defaults skippable (#3968)

**Category:** improvement
**User Impact:** Users can skip default model configuration during
onboarding and finish it later in Settings → Agents.

**Problem:** Requiring model defaults during onboarding can block users
who are not ready to choose a harness, provider, or model. Skipping also
needs to leave existing configuration untouched rather than persisting
partial selections.

**Solution:** Stage onboarding edits locally and persist them only when
users choose Next or Back. A delayed Skip action advances without any
configuration write, while a footer hint points users to the settings
location for completing setup later.

<details>
<summary>File changes</summary>

**desktop/src/features/onboarding/ui/DefaultConfigStep.tsx**
Adds the skip action and future-settings hint, and makes model
configuration transactional so Skip discards staged changes while Next
and Back preserve the intended save behavior.

**desktop/src/testing/e2eBridge.ts**
Exposes model-config setter call counts so tests can distinguish a true
zero-write skip from a write-and-rollback implementation.

**desktop/tests/e2e/onboarding-agent-defaults.spec.ts**
Covers skipping during loading and after staged edits, verifies zero
persistence calls, and confirms Next and Back still commit changes.

</details>

## Reproduction steps

1. Start fresh onboarding and continue through harness setup to
**Configure your default model settings**.
2. Change the selected harness or model, then choose **Skip for now**.
3. Confirm onboarding advances to **Join or create a community** and the
prior global model configuration remains unchanged.
4. Return through onboarding and confirm **Next** saves the staged
selection; confirm **Back** also preserves staged changes before
returning.
5. Confirm the footer says model defaults can be configured later in
**Settings → Agents**.

---------

Signed-off-by: Taylor Ho <[email protected]>
Signed-off-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

* fix(desktop): clarify inherited agent parallelism (#4010)

## Summary

- show an unambiguous `App default (10)` inherited state for parallelism
in create and edit forms
- explain that blank inherits the app default and suppress create-form
number steppers that could silently set `1`
- align the E2E mint fallback with production while preserving explicit
input → definition → app-default precedence

## Why

The forms displayed `1` even though an untouched field is omitted and
desktop minting materializes `10`. The create-form spinner could also
turn blank/inherited into an explicit `1` with one click while leaving
the field looking nearly unchanged.

## Testing

- `pnpm test` (desktop: 3,886 passed)
- `pnpm typecheck` (desktop)
- `pnpm check` (desktop)
- pre-push `desktop-check` and `desktop-test`

---------

Signed-off-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

* fix(reactions): wrap long popover names (#3834)

**Category:** fix
**User Impact:** Long custom emoji names now stay contained inside
reaction popovers and remain fully readable.

**Problem:** An unbroken custom emoji name could force a reaction
popover beyond its intended maximum width and overflow the message view.

**Solution:** Give the reaction popover a definite 288px width and allow
the complete emoji name to wrap within it without truncation or
ellipsis. Short names retain the same content and interaction behavior.

<details>
<summary>File changes</summary>

**desktop/src/features/messages/ui/MessageReactions.tsx**
Bounds the reaction popover width and allows long names to break across
lines while preserving the full shortcode.

**desktop/tests/e2e/reaction-names.spec.ts**
Covers fixed width, full text preservation, and wrapping for the maximum
supported colon-wrapped reaction name, with deterministic seeded Picsum
visual fixtures and explicit image-load waits.

</details>

## Reproduction Steps

1. Open a message with a custom emoji reaction whose name is 64
characters.
2. Hover or focus the reaction pill to open its details popover.
3. Confirm the popover remains 288px wide and the complete name wraps
within it without ellipsis.
4. Open a short-name reaction and confirm its popover remains readable
and unchanged in behavior.

## Screenshots

| Before | After |
| --- | --- |
| ![Maximum-length name
before](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3834/max-length-before-picsum.png)
| ![Maximum-length name
after](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3834/max-length-after-picsum.png)
|

**Short-name regression check**

![Short reaction
name](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3834/short-name-after-picsum.png)

## Verification

- `pnpm test` in `desktop`: 3,858 passed
- Focused reaction-name E2E with seeded Picsum captures: 2 passed
- Desktop checks and commit hooks passed

Originating Buzz channel: `f2ec9671-d78e-4cde-894c-9f4c458c7f1f`

---------

Signed-off-by: Taylor Ho <[email protected]>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

* Polish Share Compute settings (#3735)

## Summary
- Refresh Share Compute with the shared agent-style model controls.
- Reveal sharing details and advanced options only while sharing.
- Remove the preview-only mesh API path.

## Validation
- `pnpm check`
- `pnpm test`
- `pnpm exec playwright test tests/e2e/mesh-compute.spec.ts`

Snapshots are attached in a follow-up comment.

---------

Signed-off-by: kenny lopez <[email protected]>

* feat(agents): model-tuning parity in global Agent Defaults editor (#4578)

## Overview

The global Agent Defaults surface (Settings card, defaults modal,
onboarding) exposed structured controls for Effort but left Max Output
Tokens, Context Limit, and Max Rounds as raw env vars. Per-agent dialogs
had structured numeric fields but only for `isBuzzAgentRuntime` —
incorrectly excluding Goose. This PR unifies numeric-tuning capability
across all surfaces, fixes a pre-existing dual-editor defect, and adds
full test coverage.

## What changed

### Phase 1 — Catalog projection

- Add `max_rounds_env_var` to `KnownAcpRuntime` in `runtime_metadata.rs`
(`Some("BUZZ_AGENT_MAX_ROUNDS")` for buzz-agent, `None` elsewhere).
- Project all three numeric env-var fields (`max_tokens_env_var`,
`context_limit_env_var`, `max_rounds_env_var`) end-to-end:
`AcpRuntimeCatalogEntry` Rust struct, TS `types.ts`,
`RawAcpRuntimeCatalogEntry` + `fromRawAcpRuntimeCatalogEntry` in
`tauri.ts`, and the e2e mock bridge (`withMockRuntimeConfigMetadata`).

### Phase 2 — Field model

- `deriveAgentConfigFieldModel` now derives `maxOutputTokens` /
`contextLimit` / `maxRounds` descriptors from catalog-projected fields.
- `structuredEnvKeys(descriptors)` — exported helper that takes the
**rendered** descriptor set (not the whole model). Hidden keys follow
what is actually rendered per surface: global hides effort + all three
numeric keys for buzz-agent / two for Goose; per-agent buzz-agent hides
effort + three numeric keys; per-agent Goose hides only its two numeric
keys. `BUZZ_AGENT_THINKING_EFFORT` stays a visible generic env row
per-agent because no effort control renders there.

### Phase 3 — UI

- Extract `NumericTuningFields` from `buzzAgentModelTuningFields.tsx` as
a shared descriptor-driven component (`descriptors`, `envVars`,
`inheritedEnvVars`, `onEnvVarChange`). Kind-specific minima:
`NUMERIC_KIND_MIN` map (`maxOutputTokens`/`contextLimit`: 1,
`maxRounds`: 0) applied to `<input min>`.
- **Global surface** (`AgentConfigFields.tsx`): deduplicate the
previously duplicated Advanced env-editor block; render
`NumericTuningFields` below the env editor when descriptors exist;
`hiddenKeys` and `bakedGenericRows` exclusions use `structuredEnvKeys`
so structured keys are never double-rendered. Under 1000 lines.
- **Per-agent surfaces** (`EditAgentAdvancedFields`,
`PersonaAdvancedFields`): replace `isBuzzAgentRuntime` as the
numeric-field gate with `deriveNumericDescriptors(selectedRuntime)` from
`agentConfigCore`; hidden keys come from
`structuredEnvKeys(numericDescriptors)` — the same rendered descriptor
set, no local rebuilding (fixes pre-existing dual-editor defect).
Catalog status carried as `RuntimeCatalogStatus` (`loading | ready |
error`); both error and loading withhold structured controls and leave
saved values visible as generic rows, making error distinguishable from
"runtime not capable" (`ready` + no runtime).
- **Dialogs** (`AgentDefinitionDialog`, `AgentInstanceEditDialog`,
callers): `AgentDefinitionDialog` accepts `runtimeCatalogStatus?:
"loading" | "ready" | "error"` (replaces separate
`runtimesLoading`/`runtimesError` booleans); all call sites —
`AgentManagementDialogs`, `AgentsView`, `RequestedAgentCreateDialogs`,
`UserProfilePersonaDialogs` — compute and pass the status.

### Phase 4 — Tests

- `buildRecord` exported from `EnvVarsEditor.tsx` as a pure `(nextRows,
value, requiredKeys, hiddenKeys) => Record<string, string>` helper for
isolation testing.
- **17 new node tests** in `agentConfigCore.test.mjs`:
`deriveNumericDescriptors` (all three fields, partial, undefined
runtime, matches field-model subset); `structuredEnvKeys` per surface
including discriminating Goose per-agent effort-key invariant;
`NUMERIC_KIND_MIN` values.
- **4 new node tests** in `EnvVarsEditor.test.mjs`: hidden tuning key
preserved through generic row edits; runtime-switch then generic edit
(derives both descriptor sets, asserts new-runtime hidden key survives
`buildRecord` via `hiddenKeys` and old-runtime key survives via generic
rows); baked numeric key excluded via `filterBakedGenericRows` with
`numericTuningPlaceholder` assertion; clearing a structured override —
`numericTuningPlaceholder` verifies placeholder text.
- **5 new Playwright tests** in `agent-numeric-tuning.spec.ts` (added to
smoke project `testMatch`): global numeric fields visible for
buzz-agent; global: non-capable runtime hides numeric controls; Goose
per-agent shows `Inherit (16384)` after saving global value through the
UI; delayed catalog: saved values visible as generic rows while loading
then structured controls appear after settle; failed catalog: saved
values remain visible as generic rows (never the "unsupported" empty
state).

## Result

- buzz-agent global defaults: Max output tokens, Context limit, Max
rounds as structured inputs with `Inherit (N)` placeholders from baked
env.
- Goose global defaults: Max output tokens, Context limit as structured
inputs.
- A Goose global value surfaces as `Inherit (<value>)` in the per-agent
Goose edit dialog.
- No structured key is editable in two places on any surface; no
persisted key has zero editors.
- No `runtime.id === "buzz-agent"` comparison decides numeric-field
visibility anywhere — capability flows catalog →
`AcpRuntimeCatalogEntry` → field model → UI.

Signed-off-by: Will Pfleger <[email protected]>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>

* feat(mobile): bring channel menus to desktop parity (#3940)

## Overview

**Category:** improvement  
**User Impact:** Mobile users can now access consistent channel and DM
actions from both the channel list and conversation header.
**Problem:** Mobile channel menus exposed a narrower, inconsistent set
of actions than desktop, and the available actions differed by entry
point.
**Solution:** This change introduces one reusable action sheet with a
clear quick-action hierarchy, role-aware lifecycle controls,
confirmations for consequential actions, and a deliberately narrower DM
menu.

## Changes

<details>
<summary>File changes</summary>

**mobile/lib/features/channels/channel_actions_sheet.dart**  
Adds the shared channel and DM action-sheet experience used by both
entry points, including Star/Unstar and Read/Unread quick actions for
channels, section movement, mute, management, inline copy actions,
guarded lifecycle actions, confirmations, and a compact DM menu without
quick actions.

**mobile/lib/features/channels/channel_detail_page.dart**  
Routes the header ellipsis through the shared action sheet so the
in-channel menu matches the channel-list experience, including for DMs.

**mobile/lib/features/channels/channel_management_provider.dart**  
Adds archive and delete operations using the desktop-compatible relay
event kinds and refreshes channel state after completion.

**mobile/lib/features/channels/channels_page.dart**  
Makes the shared channel action-sheet entry point available to the
channel-list implementation.

**mobile/lib/features/channels/channels_page/channel_tile.dart**  
Replaces the tile-specific long-press menu with the reusable action
sheet while preserving read state and section context.

**mobile/test/features/channels/channel_actions_sheet_test.dart**  
Covers action hierarchy, owner/admin/member capability guards, loading
and failure states, DM narrowing with no quick-action row, and inline
copy actions.

**mobile/test/features/channels/channel_detail_page_test.dart**  
Updates channel-header flows to exercise management through the new
shared action sheet.

**mobile/test/features/channels/channel_management_provider_test.dart**
Verifies archive and delete event tags stay compatible with desktop
behavior.

</details>

## Reproduction Steps

1. Run the mobile app and open a populated channel list.
2. Long-press a regular channel and verify the Star/Unstar and
Read/Unread quick actions appear above Move to section…, Mute, Manage,
Copy channel name, and Copy channel ID.
3. Choose either copy action and verify it copies the expected value.
4. Open a channel, tap the header ellipsis, and verify the same action
sheet appears.
5. As an admin or owner, verify Archive appears; as an owner, verify
Delete also appears. Confirm that lifecycle actions require
confirmation.
6. Long-press or open the header menu for a DM and verify it has no
quick-action row and starts with Mute, followed by Copy channel name and
Copy channel ID.

## Screenshots

### Channel menu

| Regular channel — Mark Unread | DM — no quick actions | Archive
confirmation |
|---|---|---|
| ![Regular channel actions with Mark
Unread](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3940/final-regular-channel-mark-unread.png)
| ![DM actions without quick
actions](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3940/final-dm-no-quick-actions.png)
| ![Archive
confirmation](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/3940/final-archive-confirmation.png)
|

---------

Signed-off-by: Taylor Ho <[email protected]>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

* feat(desktop): redesign the Huddle experience (#4281)

## Summary

- open Huddles in a focused companion window with a clean handoff back
to the in-app drawer and backing channel
- redesign the participant film strip, sidebar control, transcript
surface, and themed shell treatment
- preserve microphone and device control across windows, start agent
voice on the first reply, and show agent speaking activity in the film
strip
- give each agent a distinct session voice, beginning with the
configured default, plus compact per-agent text-to-speech and voice
controls
- enroll only agents explicitly mentioned or deliberately added through
an agent panel into the live Huddle roster
- keep temporary Huddle channels out of the sidebar unless the user
explicitly brings one into the main app
- remove Huddle-only avatar policy badges and filter short silence or
noise segments before speech-to-text posts

## Why

The previous flow exposed the temporary channel as product UI, obscured
who was present or speaking, and split transcript and audio state
between the main and companion windows. This keeps backing channels as
implementation details unless a user explicitly brings a Huddle into the
app, while sharing the live conversation and audio lifecycle across both
surfaces. Agent participants now join only after an explicit invitation,
distinct voices make multi-agent Huddles easier to follow, and short
microphone noise no longer becomes stray transcript messages.

## Validation

- `pnpm check`
- `pnpm build:e2e`
- `pnpm exec playwright test tests/e2e/huddle-transcription.spec.ts
--project=smoke` (13 passed)
- Huddle sidebar visibility unit coverage (4 passed)
- focused managed-agent and persona-mention E2E coverage (2 passed)
- `pnpm test` (3,910 passed)
- `cargo clippy --manifest-path desktop/src-tauri/Cargo.toml
--all-targets -- -D warnings`
- `cargo test --manifest-path desktop/src-tauri/Cargo.toml` (2,093
passed, 14 ignored; 3 diagnostics passed)

---------

Signed-off-by: kenny lopez <[email protected]>
Signed-off-by: Wes <[email protected]>
Co-authored-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

* feat(mobile): add channel scroll navigation (#4239)

**Category:** improvement
**User Impact:** Mobile readers can jump directly to their oldest unread
message and return to the latest message with compact directional
controls.
**Problem:** Opening an active channel at its newest message makes it
easy to miss where unread conversation began, while moving back through
history lacks a lightweight route to the live edge.
**Solution:** Capture the channel's unread boundary when it opens, offer
an accessible up-chevron beneath the app bar to reach that stable
target, then reveal the inverse down-chevron at the bottom whenever the
reader is away from latest. Deep links retain precedence, and
live-follow, pagination, composer resizing, and explicit scroll
ownership continue to use the existing timeline behavior.

<details>
<summary>File changes</summary>

**mobile/lib/features/channels/channel_detail_page.dart**
Captures the channel's read state at open time and passes a stable
unread snapshot into the timeline before the normal deferred read update
advances it.

**mobile/lib/features/channels/channel_detail_page/message_list.dart**
Adds mutually exclusive oldest-unread and latest navigation, with
accessible icon controls positioned at opposite edges of the message
surface while preserving existing follow and deep-link behavior.

**mobile/test/features/channels/channel_detail_page_test.dart**
Covers the unread target, compact inverse controls, accessible tooltips,
and placement beneath the frosted app bar.

</details>

## Reproduction steps

1. Open a Flutter mobile channel that has unread messages without
entering through a message or thread deep link.
2. Confirm an up-chevron appears directly below the channel app bar
while the timeline remains at latest.
3. Tap the up-chevron and confirm the timeline scrolls to the oldest
message that was unread when the channel opened.
4. Confirm the unread control is replaced by a down-chevron at the
bottom of the timeline.
5. Tap the down-chevron and confirm the timeline returns to latest and
resumes following new messages.

## Screenshots

| At latest — up-chevron to oldest unread | Away from latest —
down-chevron to latest |
|---|---|
| ![Up-chevron beneath the mobile channel app
bar](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/4239/buzz-mobile-scroll-to-oldest-unread.png)
| ![Down-chevron above the mobile channel
composer](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/4239/buzz-mobile-scroll-to-latest.png)
|

_Real iPhone 17 Pro Simulator captures from the neutral
`buzz-mobile-scroll-to` channel._

Originating Buzz thread:
`buzz://message?channel=5b16c478-22d8-4ddd-951a-6036e19b81ff&id=6a78af32d7ac6f531b182c4e70dd5a04c503a2dab2ce2c0c74b2c6baa5921741&thread=6a78af32d7ac6f531b182c4e70dd5a04c503a2dab2ce2c0c74b2c6baa5921741`

---------

Signed-off-by: Taylor Ho <[email protected]>
Signed-off-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

* feat(mobile): sync per-group channel sorting (#4231)

**Category:** improvement
**User Impact:** Mobile users can sort each channel group by recent
activity or A–Z, with their choices synchronized with desktop.
**Problem:** Desktop supports persistent per-group channel sorting, but
mobile shows the same groups without equivalent controls or shared
preferences. The earlier mobile attempt coupled sorting to unsafe
dirty-state behavior that could overwrite newer cross-client changes.
**Solution:** Add mobile sorting controls and encrypted NIP-78
synchronization using the existing desktop `channel-sort` contract,
while retaining ordinary whole-blob last-write-wins behavior. Local
state is scoped by identity and normalized relay, startup closes
fetch/subscription gaps, and both clients use the same deterministic
ordering rules.

<details>
<summary>File changes</summary>

**desktop/src/features/sidebar/lib/channelSortPreference.test.mjs**
Updates ordering coverage for the deterministic, cross-client A–Z
comparison rule.

**desktop/src/features/sidebar/lib/channelSortPreference.ts**
Aligns desktop channel-name collation with mobile so synchronized
preferences produce the same visible order.

**mobile/lib/features/channels/channel_sort/channel_sort_manager.dart**
Adds encrypted relay synchronization with safe startup gap handling,
clock checks, and ordinary last-write-wins conflicts.

**mobile/lib/features/channels/channel_sort/channel_sort_provider.dart**
Scopes sort state to the active identity and community lifecycle.

**mobile/lib/features/channels/channel_sort/channel_sort_storage.dart**
Defines the desktop-compatible payload, relay-scoped cache and
migration, cleanup, and shared ordering behavior.

**mobile/lib/features/channels/channels_page.dart**
Connects sort state to the channel page.

**mobile/lib/features/channels/channels_page/body.dart**
Applies each selected order to Starred, custom groups, Channels, and
DMs.

**mobile/lib/features/channels/channels_page/sections.dart**
Adds checked Recent and A–Z actions using the existing anchored-popover
UI.


**mobile/test/features/channels/channel_sort/channel_sort_manager_test.dart**
Covers payload adoption, encrypted publication, conflicts, timestamps,
retries, and cleanup.


**mobile/test/features/channels/channel_sort/channel_sort_storage_test.dart**
Covers parsing, relay isolation, migration, cleanup, and ordering modes.

**mobile/test/features/channels/channels_page_test.dart**
Verifies the group controls expose both choices.

</details>

### Reproduction steps

1. Open the mobile channel list with populated built-in and custom
groups.
2. Open a group menu and choose **Sort: Recent**; confirm active
channels move to the top.
3. Choose **Sort: A–Z**; confirm deterministic alphabetical ordering
returns.
4. Repeat for Starred, a custom group, Channels, and DMs.
5. Open desktop with the same identity and community and confirm each
synchronized preference.
6. Switch communities and confirm cached preferences do not bleed across
relays.

### Screenshots

Approved `live` custom-section flow with `research` kept offscreen.

| Recent selected | A–Z result | A–Z selected |
|---|---|---|
| ![live custom section with Recent
selected](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/4231/live-recent-selected.png)
| ![live custom section sorted
A–Z](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/4231/live-az-result.png)
| ![live custom section with A–Z
selected](https://d24qwcpro867f5.cloudfront.net/repos/buzz/prs/4231/live-az-selected.png)
|

### Validation

- Mobile `flutter analyze` — clean
- Focused mobile sort and channel-page suites — 37/37 passed
- Desktop full suite — 3906/3906 passed
- Mobile full suite — 1034 passed, 1 skipped, 1 unrelated baseline
failure reproduced at `ac4fa13b8`

<!-- Originating Buzz channel: 2a16a2bb-6fd3-4d69-8182-2afcb21b2d14 -->

---------

Signed-off-by: Taylor Ho <[email protected]>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>

* feat: ship Buzz Term (#4347)

## Summary

- ship **Buzz Term** end to end: the terminal engine/runtime, mounted
desktop substrate, and user-visible naming
- add Quinn's tape-deck-inspired banner: a beveled chassis filled by the
`buzz term` wordmark, surrounded by a complete-hex field
- derive the wordmark's three-stop sweep from each theme's terminal
palette so primary, secondary, and accent roles remain visibly distinct
across all 62 shipped themes, including light themes
- paint the banner once on its own pointer-transparent canvas; PTY
rendering beneath it remains unchanged

## Banner behavior

- uses the renderer's shared `8.4 × 17` cell metrics and production
aspect ratio `2.0238`
- regenerates only for viewport/theme changes; palette switches repaint
correctly while the banner is visible
- dismisses on non-empty output from the active terminal session; empty
output and inactive sessions do not dismiss it
- fails closed below **70 columns** rather than squeezing or clipping
the wordmark
- adds **8 lines** to `terminalRenderer.ts` for shared cell metrics and
**zero lines inside `paint()`**

## Screenshots

| Buzz (light) | Buzz Dark |
|---|---|
| ![Buzz Term — Buzz
light](https://buzz.block.builderlab.xyz/media/fe4c50c1cd03645bff8f3cff588353618fd3004f320ab360165b6cb48f041eb6.png)
| ![Buzz Term — Buzz
Dark](https://buzz.block.builderlab.xyz/media/e5731f40fe020070c0b327287b4b0b6f0103adde852d0f6451ad66a0ff9d14bb.png)
|

| Kanagawa Lotus (light) | Red |
|---|---|
| ![Buzz Term — Kanagawa
Lotus](https://buzz.block.builderlab.xyz/media/260b211fe7bd232eb7faa4ec34c99baee5fed380556427c64495ceed3396a8c5.png)
| ![Buzz Term —
Red](https://buzz.block.builderlab.xyz/media/42a90830824683a0054e24649c58564134af831607736d823992181b792087d3.png)
|

Additional production-aspect finals:
[Vesper](https://buzz.block.builderlab.xyz/media/9ca6514b63f8cfb2107a85ca46f16a940c0883848e6fbc718e411af94aa13100.png),
[Min
Dark](https://buzz.block.builderlab.xyz/media/f67bd2970e5d64ffb07b1ae78ab58c847e6ebc23e7e7a48e067eb024dba64ec8.png),
and [Dark
Plus](https://buzz.block.builderlab.xyz/media/290fee08924f37d064abc687ecf3e9526ab05b87e8e56d610f23048949793dbe.png).

The screenshot harness was checked against the shipped painter at this
exact head: all **2,541 draw calls** matched on color, glyph, x, and y;
four deliberate divergence controls fired.

## Verification at `98ebc8f9048bd5f0ceb7e843b67874d642f0b7fd`

- desktop tests: **3,946 / 3,946**
- TypeScript: clean
- checks: pass (two pre-existing informational `useTemplate` notices
only)
- integration/e2e: PASS (independent exact-SHA lane; artifacts recorded
in the originating Buzz thread)
- artifact/dead-path sweep: clean
- redteam G1–G7: PASS
  - all six named banner emitter-deletion mutants die
- independent handwritten five-row full-wordmark fixture kills Quinn's
seven-mutant battery, including a one-pixel glyph change
- real `112 × 46` canvas-rect dismissal tests separately cover active
non-empty, active empty, and inactive non-empty output
- layer-drop and zero-draw painter mutants die; z-order and
pointer-events verified
  - CI's `tsc && vite build` includes all three banner modules
- performance at DPR 2 (worst-case measured envelope):
- one-time content paint: **~0.7–0.8 ms**, paid only when the banner is
built or its palette changes
- busy compositor, CSS `1277 × 697`, backing `2554 × 1394`: **470–497
µs/frame** for the full banner (**2.82–2.98%** of a 60 Hz frame)
- busy compositor, CSS `1920 × 1080`, backing `3840 × 2160`:
**1,139–1,212 µs/frame** (**6.83–7.27%**)
- empty, one-glyph, and full-banner controls converge: compositor cost
follows backing-layer area and DPR rather than painted-cell count
- in the actual idle welcome state, cost is below both vsync-clamped
rigs' resolution; it is not claimed as zero
- **Pane cross-rig spread: resolved at matched loop rate.** Two
independent rigs initially differed 2.3× (58–68 vs 136 µs/Mpx of backing
store; pane, CSS 1277×697 / backing 2554×1394, DPR 2). The cause of
*that* spread is rAF loop rate: the higher figure came from a
free-running loop at ~1600fps. Throttled to ~200–236fps, both rigs read
58–68 µs/Mpx (1.25–1.44% of a 60Hz frame). The busy-composite figures
quoted above remain the **unthrottled worst case** and are conservative
by ~2.3× at the pane. Not established: the mechanism and sign of
free-running distortion (one rig under-charges ~15%, the other
over-charges 2.3×), and the 1080p figure has not been re-measured
throttled.
- the layer paints only on generation/theme/resize and dismisses on
first non-empty active-session output, so the measurable busy cost is a
short-lived worst case rather than a persistent PTY paint-path tax

## Follow-ups in this PR

These are intentionally subsequent commits after the certified
static-banner head, not claims about `98ebc8f90`:

1. close the compositor metrology: remeasure the 1080p point throttled
and characterize the opposite-sign free-running rAF distortion, with
each measurement regime stated
2. add Tyler's animated honeycomb color waves, gated by
`prefers-reduced-motion`, a full 62-theme phase-sweep contrast check,
and DPR-2 per-tick performance certification
3. land the already-proven mounted theme-switch regression probe from
`RESEARCH/BUZZ_TERM_G3A_PROBE/`
4. bound the slow/hang-shaped G1-c mutant `waitFor`
5. optionally trim the generator to its ink bounding box, reducing the
minimum viewport from 70 to 62 columns

---------

Signed-off-by: tlongwell-block <[email protected]>
Signed-off-by: npub1mprnacetjua2xx3p5eddmhxyk6wv929ymm5py8kd2xfxurxahspqqlgyta <d8473ee32b973aa31a21a65adddcc4b69cc2a8a4dee8121ecd51926e0cddbc02@buzz.block.builderlab.xyz>
Signed-off-by: npub1jh9wn95s0472h86ahapupaf7m6kx4v9sx2n0atj2hltcfer8k06s5n3pyf <95cae996907d7cab9f5dbf43c0f53edeac6ab0b032a6feae4abfd784e467b3f5@buzz.block.builderlab.xyz>
Signed-off-by: npub17jjz49l9jjmhhk7cac63j8yt9z555n9cw8vk7v5jz4vzw4ppld5qgj57cc <f4a42a97e594b77bdbd8ee35191c8b28a94a4cb871d96f32921558275421fb68@buzz.block.builderlab.xyz>
Signed-off-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Co-authored-by: Dawn (sprout agent) <c6237ef84fa537c78dcee78efd2d4e59f728859c7f194da42ac51ededfa0be05@sprout-oss.stage.blox.sqprod.co>
Co-authored-by: npub1t2tgm7d8f995uqvmnm8h88sg3wnpp9a5xysjf6dg3tjmgt3ltulqdp8ehr <5a968df9a7494b4e019b9ecf739e088ba61097b4312124e9a88ae5b42e3f5f3e@buzz.block.builderlab.xyz>
Co-authored-by: npub1cc3ha7z055mu0rwwu7806t2wt8mj3pvu0uv5mfp2c50dahaqhczshdalg6 <c6237ef84fa537c78dcee78efd2d4e59f728859c7f194da42ac51ededfa0be05@buzz.block.builderlab.xyz>
Co-authored-by: npub17jjz49l9jjmhhk7cac63j8yt9z555n9cw8vk7v5jz4vzw4ppld5qgj57cc <f4a42a97e594b77bdbd8ee35191c8b28a94a4cb871d96f32921558275421fb68@buzz.block.builderlab.xyz>
Co-authored-by: npub1jh9wn95s0472h86ahapupaf7m6kx4v9sx2n0atj2hltcfer8k06s5n3pyf <95cae996907d7cab9f5dbf43c0f53edeac6ab0b032a6feae4abfd784e467b3f5@buzz.block.builderlab.xyz>
Co-authored-by: npub1mprnacetjua2xx3p5eddmhxyk6wv929ymm5py8kd2xfxurxahspqqlgyta <d8473ee32b973aa31a21a65adddcc4b69cc2a8a4dee8121ecd51926e0cddbc02@buzz.block.builderlab.xyz>
Co-authored-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>

* Polish mobile inbox and media flows (#4512)

## Summary

- make mobile unread state visible with bold channel names, an animated
Inbox badge, and swipe-to-toggle Inbox rows
- add directional transitions for top-level mobile navigation
- let mobile send while media uploads, with cancellable progress UI
- normalize iOS and Android video uploads, attach poster frames, and
improve native video playback

## Validation

- `just mobile-check`
- `just mobile-test`
- `cargo test -p buzz-media`
- Pixel smoke test
- iPhone smoke test

Desktop background uploads moved to #4522 so the two platforms can be
reviewed independently.

---------

Signed-off-by: kenny lopez <[email protected]>
Signed-off-by: Tom Brow <[email protected]>
Co-authored-by: leader <71e9f2c44a6932b6772caaaccda1911d010463c3e2c6c40410b8329956046801@buzz.block.builderlab.xyz>
Co-authored-by: Tom Brow <[email protected]>

* fix(desktop): stop clipping focus ring on channel intro action cards (#2392) (#4374)

## What

Fixes #2392 — the action cards in the empty-channel intro ("Create
agent", "Add people") had their `focus-visible` ring clipped by the
surrounding scroll container.

## Root cause

The cards sit in a `flex ... overflow-x-auto pb-1` row. Setting
`overflow-x` (without `overflow-y`) makes the browser compute
`overflow-y: auto` as well, so the container clips anything painted
outside its padding box — including the cards' `focus-visible:ring-2`
box-shadow. With only `pb-1` padding, the top/left/right of the ring
were cut off when Tabbing to a card.

## Change

`desktop/src/features/messages/ui/ChannelIntroBlock.tsx` — `pb-1` →
`p-1` on the action-cards scroll container, reserving 4px on all four
sides so the focus ring renders fully inside the scroll container's
padding box.

- 1 file, 1 line. No behavior change for mouse users or layout.

## Verification

- `pnpm typecheck` — clean
- `pnpm exec biome check src/features/messages/ui/ChannelIntroBlock.tsx`
— clean
- `pnpm check:file-sizes` — clean
- Desktop unit suite — **3906/3906 pass**

Signed-off-by: Sarthak Singh <[email protected]>

Signed-off-by: Sarthak Singh <[email protected]>

* Defer desktop media uploads until send (#4522)

## Summary

- send desktop messages immediately while media uploads continue in
background state across channel navigation
- show immediate progress above the composer and keep Jump to latest
above it
- report the real media stages as Preparing, Processing, Converting,
Uploading, and Finishing
- use Buzz's shared spinner during local media work, then switch to the
real percentage when byte transfer begins
- animate phase-label and status-suffix changes without overlap or
layout jumps
- keep cancel, progress fill, message publication, and community-reset
behavior coordinated with the background task
- use raw Tauri IPC for large browser files so renderer-side byte
serialization does not block initial feedback

## Why

Desktop previously blocked sending while attachments uploaded in the
composer. Large videos could also pause the renderer before progress
appeared, and the progress pill said Uploading while native media
processing was still underway. This makes the initial response immediate
and describes the work actually happening.

## Validation

- `cd desktop && pnpm check`
- `cd desktop && pnpm typecheck`
- `cd desktop && pnpm test` (3,931 passed)
- `cd desktop && pnpm exec vite build --mode e2e`
- `cd desktop && pnpm exec playwright test
tests/e2e/file-attachment.spec.ts --project=smoke` (11 passed)
- focused native media tests (80 passed)
- native Clippy with all targets and features
- pre-push native suite (2,107 passed, 14 ignored; 3 diagnostics passed)

Updated phase snapshots are included in the PR comments.

Split from #4512 so the desktop and mobile changes can be reviewed
independently.

---------

Signed-off-by: kenny lopez <[email protected]>
Signed-off-by: Wes <[email protected]>
Co-authored-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

* Refine desktop timeline activity presentation (#4582)

## Summary

- Make channel join/leave activity use the selected inline avatar-stack
treatment.
- Group related membership activity for one hour and preserve
profile/overflow-name interactions.
- Restore the virtualized day-divider handoff and align the sticky date
behavior with the message timeline.

## Validation

- `pnpm check`
- `pnpm test`
- `cargo test --manifest-path desktop/src-tauri/Cargo.toml`
- Visual desktop screenshot captured with seeded membership activity

---------

Signed-off-by: kenny lopez <[email protected]>

* Remove blur from Welcome composer guidance (#4691)

## Summary
- Keep the Welcome composer prompt above the dock blur so it stays
readable.
- Remove blur from the prompt and persona-motion paths.
- Cover the crisp, correctly layered banner in the onboarding browser
test.

## Validation
- `pnpm -C desktop exec biome check
src/features/channels/ui/WelcomeComposerBanner.tsx
tests/e2e/onboarding.spec.ts`
- `pnpm -C desktop build:e2e`
- `pnpm -C desktop exec playwright test tests/e2e/onboarding.spec.ts
--grep "finishing onboarding creates starter channels and focuses
welcome-everyone for a new member" --project=integration`

---------

Signed-off-by: kenny lopez <[email protected]>

* docs(nip-am): normative amendment — cache SHOULD/MUST + pricingIdentity + consumer cost guidance (#4632)

Amends `docs/nips/NIP-AM.md` with three normative publisher-behavior
changes per the cleared Usage v2 plan (plan v3, D4 + D2').

## Changes

### 1. Cache emission semantics (D4)

Replaces the unconditional `MAY` with qualified obligations:

- Publishers SHOULD emit `cacheReadTokens` / `cacheWriteTokens` when the
provider exposes a cache component.
- Publishers MUST preserve an explicit zero when the provider reports
zero.
- Publishers MUST omit the field (never null or fabricated zero) when
that component is unavailable to the publisher — including when the
provider supports it but the harness does not surface it.

An explicit carve-out in both the JSON comment block and the
Numeric-validity prose exempts these fields from the payload-wide null
guidance. Omission is the only valid representation for an unavailable
cache component.

### 2. Optional `pricingIdentity` field (D2')

Adds an optional, non-nullable `pricingIdentity` object (`authority`,
`model`, `cacheClass`), defined as billing authority — distinct from the
transport `Provider` enum.

- `authority` is a registered billing-namespace identifier: exact
lowercase hostname, no scheme, no path, no trailing slash. Registered
values: `api.anthropic.com`, `api.openai.com`, `openrouter.ai`. The set
extends only by NIP amendment. Pricing lookup is an exact string match
on `(authority, model)`.
- Present only when the publisher can prove applicability: direct
official-endpoint connections prove via the actually-requested resolved
model; other routes MUST receive response-supplied authoritative billing
identity.
- MUST omit for custom/overridden base URLs, gateways (unless the
gateway is the named billing authority), unresolved aliases, and turns
where usage contributions carry more than one billing identity
(including identity-bearing mixed with unresolved).
- `cacheClass` is omitted (not null) when not applicable.
- `pricingIdentity` is optional but not nullable — omission is the only
absence representation.
- The existing `model` field retains its non-billing semantics
(configured/session model) and is never overloaded.
- Consumers MUST treat omission as "price unknown" and MUST NOT infer a
price from the session `model` field.

### 3. Consumer cost guidance (D4)

- Consumers MAY recompute cost estimates using the billing identity and
a pricing manifest.
- Consumers MUST retain the provenance of any cost value (e.g.
`manifest-estimated`, `wire-reported`).
- Consumers MUST NOT merge manifest-estimated and wire-reported costs
into an unlabeled total.

Manifest-vs-wire display preference is application policy and
deliberately excluded from this NIP.

## Scope

Doc-only. Single file: `docs/nips/NIP-AM.md`.

---------

Signed-off-by: Will Pfleger <[email protected]>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>

* docs(acp): explain per-channel session model in base prompt (#4729)

## Overview

Agents running in Buzz have no built-in awareness that each channel is
an isolated conversation context. When a human mentions work "you" are
doing in another channel, the current session can misread this as its
own active context and try to coordinate, re-plan, or take ownership of
it — causing confusion and wasted turns.

## What changed

Added a `## Session Model` section to
`crates/buzz-acp/src/base_prompt.md`, inserted immediately after the
opening paragraph and before `## Buzz CLI`. The section explains:

- Each channel is a separate session; multiple sessions of the same
agent identity may be active simultaneously.
- Sessions share core memory, workspace, and relay — but not
conversation context or in-flight reasoning.
- Cross-channel work belongs to the owning session by default; the
current session may take it over only when the human explicitly requests
it.

No runtime code changes. Base prompt only.

---------

Signed-off-by: Will Pfleger <[email protected]>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>

* fix(desktop): show cached display names on startup (#3317)

## Why

Buzz restores cached channels and messages before profile lookups
complete. On launch, that briefly exposes pubkey-derived labels in place
of familiar display names.

## What

- Persist a bounded, relay-scoped cache of last-known display names,
NIP-01 names, and NIP-05 handles
- Seed batch profile queries from those labels immediately, while
keeping them stale so the existing relay request revalidates them
- Keep cached data presentation-only: avatars and ownership metadata are
not persisted or used to seed profile-detail caches
- Remove cleared or missing profiles, purge a relay's labels when its
community is removed, and include the cache in local-storage quota
recovery
- Add unit coverage for parsing, bounds, eviction, malformed data, and
cleared profiles
- Add an E2E regression that delays the relay profile response and
verifies the cached name is rendered first

## Risk Assessment

Low. The cache is disposable, capped at 1,000 entries per relay, scoped
by normalized relay URL, and always revalidated. It contains only public
label fields and does not restore avatars, agent ownership, or
authorization state.

## Verification

- `just ci`
- `pnpm typecheck`
- `pnpm test` — 3,727 passed
- `pnpm exec playwright test tests/e2e/channels.spec.ts --grep "cached
profile labels"` — passed

Generated with Codex

* Polish sidebar unread hierarchy (#4573)

## Summary

- Keep selected sidebar rows regular by default; manually unread rows
become bold immediately.
- Apply a clearer dark-mode hierarchy: standard inactive rows at 75%,
muted rows at 45%, and unread rows at full emphasis.
- Keep hover text color stable while retaining the selected-row and
unread cues.

## Validation

- `pnpm typecheck`
- `pnpm build:e2e`
- Playwright: sidebar badge and channel-mute coverage

## Screenshots

Posted in the PR comments.

---------

Signed-off-by: kenny lopez <[email protected]>

* feat(desktop): surface config diff in restart-required badge (#3637)

The "Restart required" badge reports that an agent's running config has
drifted from its spawn-time config, but never says what changed. This
ships the full feature: a typed Rust diff engine and a TS/UI layer that
renders it at every badge site.

## Rust core (spawn-snapshot diff engine)

Replaces the lossy `u64` `spawn_config_hash` with a typed
`SpawnConfigSnapshot`. The snapshot is stamped from the already-resolved
command/env/config values immediately before `spawn()`, closing the race
window where a mid-spawn config edit would suppress the badge.

`SpawnConfigSnapshot::canonical()` is the single JSON projection shared
by the badge and the diff. Drift is `to_value(stamped) !=
to_value(current)`; the diff is a generic leaf walk over those same two
values, so badge-on and diff-non-empty are structurally guaranteed.
Adding a snapshot field reaches the UI with no code change to the diff
engine — `mutation_table_covers_every_serialized_field` fails CI if a
new field arrives without a mutation row.

`eligible_restart_diff(persona_orphaned, Option<TrackedSpawnState>)`
returns the final vector — snapshot walk entries plus a synthetic
`adapter_availability` entry. It returns empty for an orphaned instance
(spawning one would fail) and for agents with no tracked spawn state
(never stamped, can never have drifted). `needs_restart =
!restart_diff.is_empty()` derives from that vector and nothing else.

Redaction policy (`policy_for(path)`) is shared by the wire diff and the
snapshot's manual `Debug` via `is_safe_to_reveal()` from
`managed_agents::env_vars` as the single authority for env-key masking:

| Policy | Paths | Rendering |
|---|---|---|
| `Text` | `system_prompt`, `team_instructions` | character counts only
|
| `MaskedBare` | `args`, `relay_url` | `••••`, no suffix |
| `MaskedSuffix` | non-allowlisted `env.*` | `••••` + last 4 chars when
longer than 8 |
| `Plain` | allowlisted `env.*` (`BUZZ_AGENT_THINKING_EFFORT`,
`BUZZ_AGENT_PROVIDER`, `BUZZ_AGENT_MODEL`, `DATABRICKS_HOST/MODEL`) and
everything else | verbatim |

Default-deny: every env key not in the explicit allowlist stays masked.
`is_safe_to_reveal()` is the single allowlist authority for both the
baked-env display and the diff.

`restart_diff` is omitted from the wire when empty
(`skip_serializing_if`).

## TypeScript / UI layer

New `restartDiff.ts` module defines `RestartDiffEntry`, `RestartChange`,
`JsonValue`; `tauri.ts` and `types.ts` re-export and add `restart_diff`
/ `restartDiff` fields (Rust omission → `restartDiff: []`).

**`RestartDiffBadge`** — hover tooltip capped at 6 entries + "and N
more", `asChild` span trigger (never inside a `<button>`), auto-restart
blurb below the diff list (on/off variant from `autoRestartEnabled`
prop; same `AUTO_RESTART_ON_BLURB` / `AUTO_RESTART_OFF_BLURB` constants
shared with the Runtime-tab banner). **`RestartDiffList`** renders the
full uncapped list for the Runtime-tab banner with `tooltip`/`inline`
presentation variants for correct foreground in both surfaces.

**`ManagedAgentRow` B4 fix** — badge moved to a sibling `div` of the row
expansion button; tooltip trigger has no `button` ancestor.

**`UnifiedAgentsSection`** — both badge sites render
`<RestartDiffBadge>` instead of a raw `<Badge>`, with
`autoRestartEnabled` threaded from `agent.autoRestartOnConfigChange`.

**Side-panel fix** — `RestartDiffBadge` rendered tab-independently in
the `ProfileSummaryView` hero area (was Runtime-tab only — root cause of
the ~50% inconsistency Will reported). Hero badge is `self-center` in
the flex column. `ProfileRuntimeTabContent` early-return checks
`needsRestart` so the banner is never dropped when all other content is
empty. Auto-restart blurb in the Runtime-tab banner uses the shared
constants.

## Wire shape

```jsonc
"restart_diff": [
  { "field": "model",              "change": { "kind": "value",  "before": "gpt-5", "after": "claude-4" } },
  { "field": "system_prompt",      "change": { "kind": "text",   "before_chars": 1234, "after_chars": 1410 } },
  { "field": "env.OPENAI_API_KEY", "change": { "kind": "masked", "before": "••••bc12", "after": "••••xyz9" } },
  { "field": "env.BUZZ_AGENT_THINKING_EFFORT", "change": { "kind": "value", "before": "medium", "after": "high" } }
]
```

`added`/`removed` occur only for dynamic-map keys; nullable struct
fields always serialize as `null`; arrays are atomic leaves (`args`,
never `args.0`).

## Tests

**Rust** — 1902 passing: snapshot mutation coverage, diff entry
serialization, allowlist-aware env masking
(`allowlisted_env_key_shows_plain_value`,
`allowlisted_env_key_is_case_insensitive`,
`non_allowlisted_env_key_stays_masked`),
`unstamped_agent_yields_no_badge_and_no_entries` (both orphan values),
`summary_without_drift_omits_restart_diff_from_the_wire`,
`unstamped_availability_is_not_drift`. Clippy clean, fmt clean.

**TypeScript** — `needs-restart-screenshots.spec.ts`: 11 E2E cases
registered in the smoke project — all three badge sites, tooltip +
keyboard focus, DOM no-button-ancestor assertion, 6+1 truncation,
uncapped Runtime list, unknown field humanisation, side-panel badge on
default Info tab, inactive/friendly-error Runtime opening path.

Consolidates [#3652](https://github.com/block/buzz/pull/3652)

Signed-off-by: Will Pfleger <[email protected]>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>

* feat(desktop): persist sidebar observed-unread across webview reload (#3976)

## Problem

`Command+R` (webview reload) wipes the two in-memory refs driving
sidebar channel unread badges: `observedUnreadEventsByChannelRef` and
`latestByChannelRef`. The boot catch-up REQ can only fetch events newer
than each channel's NIP-RS frontier, so thread replies that arrived
before the frontier was passively advanced (the common case) are never
re-discovered.

Inbox is unaffected because it rebuilds candidates from a relay feed
query and checks fine-grained `thread:`/`msg:` markers. The sidebar
badge path lacks an equivalent recovery mechanism.

## Solution

Persist the sidebar's per-event candidate set to localStorage as a
disposable, versioned projection cache
(`buzz-observed-unread.v1:<relay>:<pubkey>`) and hydrate it on boot
before the catch-up REQ runs.

### New files

**`observedUnreadStorage.ts`** — storage module for the cache:
- Keyed
`buzz-observed-unread.v1:<normalizedRelayUrl>:<normalizedPubkey>`
(relay-scoped to prevent cross-community leakage, matching
`threadActivityStorage`)
- Stores validated per-event `ObservedUnreadEvent` rows;
`latestByChannel` is derived at hydration — no divergent dual aggregate
- Age pruning (7d = `READ_STATE_HORIZON_SECONDS`), per-channel cap
(1000), global cap (5000) across all channels in a scope bucket
- Payload `updatedAt` for LRU ordering; registered in
`PURE_CACHE_KEY_PREFIXES` for 2 MiB eviction budget
- Field-level validation on decode; write failure is non-fatal
(session-only degradation)
- Snapshot-owning timers: `scheduleObservedUnreadWrite` deep-clones the
events map at schedule time — a late A-scope timer can never read B's
mutable refs or write under B's key

**`useObservedUnreadPersistence.ts`** — hook that owns all persistence
lifecycle:
- Scope fence: `normalized pubkey + normalized relay` identity;
`isScopeLoaded()` callback guards both projection (`rawUnread`) and
every **observed-cache mutation** (`recordUnreadEvent`, `removeChannel`,
`clearAll`) before touching refs or storage. Note: stale-scope calls to
`markChannelRead`/`markAllChannelsRead` can still affect
`forcedUnreadRef` and NIP-RS markers, which are pre-existing on `main`
and deferred to the NIP-RS arc (see Deferred below).
- Synchronous `pagehide` flush closes the Cmd+R timing gap
(`useReloadShortcut.ts` reloads within 500ms of teardown, before the
1-second debounce fires)
- Identity-reset effect: flushes old scope, resets refs, hydrates from
storage, stamps loaded scope — all atomic; cleanup flushes on unmount
- `clearAll` cancels the pending timer, resets both in-memory refs, and
clears storage in a single transactional operation; `removeChannel`
deletes the channel from both refs and replaces any pending snapshot
with the current full map — never cancel-without-replacement, preserving
sibling-channel events on reload
- Marker-prune effect on `readStateVersion`: evaluates each retained
event with `observedUnreadEventReadAt()` (the same evaluator used by the
projection memo) and removes covered events, rederiving per-channel
latest — never clears a whole channel for a single thread/msg marker
- Returns a stable `useMemo`-wrapped API object keyed on actual deps so
unrelated re-renders do not restart the catch-up REQ
- `isScopeLoaded` is a `useCallback` (not a memoized boolean) — always
reads the ref at call time, never stale

### Modified files

**`useUnreadChannels.ts`** — hook integration:
- Calls `useObservedUnreadPersistence` with all persistence wired
through the returned API
- `rawUnread`: `isScopeLoaded()` guard suppresses A-scope refs from
projecting under B
- `recordUnreadEvent`: `isScopeLoaded()` fence before touching refs;
schedules a debounced write on each successful record
- `markChannelRead` clearObserved path: calls `removeChannel` so the
cleared state survives reload
- `markAllChannelsRead`: delegates to the owner's fenced `clearAll` —
the parent does not reset the observed refs directly; `clearAll` owns
the transactional clear of both refs and storage, preventing a stale
scope-A callback from corrupting scope B

**`localStorageQuota.ts`** — registers `buzz-observed-unread.v1:` in
`PURE_CACHE_KEY_PREFIXES`

## Design constraints

The cache is a **disposable projection**: versioned key, read-through
only, safe to delete wholesale. It does not touch `ReadStateManager`,
marker semantics, or `forcedUnreadStore`. Zero overlap with the NIP-RS
manual mark-read/unread protocol work in progress in another channel;
migration path when that lands is "stop reading the key."

## Test coverage

**`observedUnreadStorage.test.mjs`** covers storage primitives:
- Key normalization, relay-scoped isolation, round-trip correctness
- Age-prune and per-channel cap on read and write; global cap across
channels
- `deriveLatestByChannel` correctness
- Thread-marker prune leaves sibling thread events persisted and lit
- Scope-isolation state machine: A rows visible in A, absent in B,
restored on A again; late A-scope write does not overwrite B's bucket
- Malformed structures/fields, relay/pubkey isolation, quota failure
degradation

**`useObservedUnreadPersistence.test.mjs`** exercises the real hook via
`createRoot` + `act`:
- pagehide flush: event recorded within debounce window survives reload
(headline regression)
- Unmount with pending write flushes before teardown
- `clearAll` cancels pending debounce so no resurrection after reload
- `removeChannel` replaces pending snapshot so sibling channel B
survives reload (two-channel repro)
- Marker prune: thread and channel markers prune covered events; sibling
channels survive
- `isScopeLoaded` returns false before identity-reset effect commits,
true after
- A→B scope switch: pending A-timer is cancelled by flush, A data
persisted synchronously (hydration round-trip)
- Stale `clearAll` from scope A rejects after scope B loads
(observed-cache scope fence)
- Stale `removeChannel` from scope A rejects after scope B loads
(observed-cache scope fence)
- API object identity stable across unrelated re-renders (catch-up
stability)

**`useUnreadChannels.test.mjs`** exercises the full parent-to-owner seam
with real hook mounts:
- Stale `markChannelRead` from scope A does not corrupt B's observed
bucket after flush
- Stale `markAllChannelsRead` from scope A does not overwrite B's bucket
after flush

## Deferred

Issues deferred to the NIP-RS arc (`#unread-messages-ux`) or future
hardening — not regressions introduced by this PR:

- **Stale-scope `forcedUnreadRef` / `markContextRead` exposure**: a
stale scope-A `markChannelRead` or `markAllChannelsRead` still deletes
B's `forcedUnreadRef` entries and advances B's NIP-RS markers via
`markContextRead` before the observed-cache fence rejects. This is
pre-existing on `origin/main` (identical shape at lines 316/330). Fix
requires touching `forcedUnreadStore` and marker paths — out of scope
for Fix A. Deferred to the NIP-RS work.
- **`isScopeLoaded` empty-scope hardening**: `isScopeLoaded()` returns
`true` when `pubkey` and `relay` are empty strings (no active session).
A guard could assert non-empty identity before stamping scope-loaded.
Low risk in practice since the hook is only mounted after auth, but
could be tightened.
- **Catch-up batch scheduling**: `handleChannelMessage` and the catch-up
loop each clone the full events map per event via
`scheduleObservedUnreadWrite`. For channels with large backlogs this
produces O(n) snapshot clones per catch-up batch. A batch-schedule API
(single snapshot at end of batch) would reduce allocations. Not
observable in normal use; deferred as a performance optimization.

---------

Signed-off-by: Will Pfleger <[email protected]>
Signed-off-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>

* Refine community invite links (#4734)

## Summary

- Separate direct invites from link sharing with a labeled divider.
- Show the generated invite URL inline with truncation and a copy
control.
- Use shared loading feedback and a restrained copy-status resize.

## Validation

- `pnpm -C desktop exec playwright test
tests/e2e/invite-link-copy.spec.ts
tests/e2e/invites-settings-screenshots.spec.ts` (7 passed)

---------

Signed-off-by: kenny lopez <[email protected]>

* fix(agents): canonicalize stale persona harness…

…ng over the community rail (#5947)

## Summary

Collapsing the sidebar left a phantom copy of it painted over the
community/relay rail — opaquely on flat themes (vesper et al., which
made the rail look *removed*), and as ghost fragments (muted search-box
fill, truncated channel-name tails) on the Buzz themes whose chrome is
intentionally transparent for the gradient.

**Cause:** #4281 made the app-sidebar layer `overflow-visible` (the
huddle drawer needs to escape it). That removed the ancestor clipping
the offcanvas collapse relied on: the sidebar slides to `left:
-sidebar-width` but kept painting, exactly over the `z-0` rail (`z-10`
sidebar layer).

**Fix:** the offcanvas-collapsed sidebar container is now `invisible` +
`pointer-events-none`, with `visibility` added to the transition list so
the 200 ms slide-out still animates and the flip happens only at the
transition's end. Theme-independent; no per-theme CSS touched; the
huddle drawer's `overflow-visible` is preserved.

## Before / after

Left 420px of the app with the sidebar collapsed. Before = unpatched
`origin/main` @ 69107dc; after = this branch. Same seeded state, same
build pipeline (`build:e2e` between checkouts).

| theme | before (ghost sidebar over the rail) | after (rail clean: A /
B / + visible) |
|---|---|---|
| vesper |
![before-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-vesper.png)
|
![after-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-vesper.png)
|
| buzz |
![before-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz.png)
|
![after-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz.png)
|
| buzz-dark |
![before-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz-dark.png)
|
![after-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz-dark.png)
|

Before shots: ghost `⌘K` search chip + blue active-item pill painted
over the rail column; on vesper the opaque panel hides the rail buttons
entirely. After: the rail's community buttons (A, B) and `+` are visible
and clickable in all three themes.

Reported by Thomas P in #buzz-bugs:
buzz://message?channel=e62570dd-33ad-42c5-b92b-75f2689f9694&id=9ea401ca1d009f555ca4324e136f8d8d8156db2f8afa3ff89fd038d2c16260f7

cc @klopez4212 — this touches the layout your #4281/#5478 work shaped;
please confirm it doesn't defeat the huddle drawer or glass intentions.
The change deliberately hides only the *offcanvas-collapsed* container,
nothing in the expanded path.

## Test plan

- [x] New Playwright regression spec `sidebar-offcanvas-rail.spec.ts`
(buzz / buzz-dark / vesper): collapsed sidebar must be `visibility:
hidden` + `pointer-events: none`, community rail stays visible and
interactive. **Fails on unpatched build** (verified), passes with the
fix.
- [x] Full desktop unit suite: 4,954 pass / 0 fail
- [x] `pnpm typecheck`, `pnpm check` (biome + file-size ratchet +
px-text + pubkey-truncation) green
- [x] Before/after screenshots above captured via the e2e harness on
both builds

Signed-off-by: Thomas Petersen <[email protected]>
Co-authored-by: Wintermute <165f0c871dd2586bb18b6aa109eeaf57bb2132ff4d27b10120f4368a0f627022@buzz.block.builderlab.xyz>

* Projects v3: unify sharing, discussions, and issue ownership (block#5792)

## Summary

Projects v3 makes repository work shareable, discussion-aware, and
easier to scan in one coherent workspace. People can copy canonical
links, reopen the exact workspace tab, understand issue and pull-request
context at a glance, find related channel conversations, and assign or
unassign issues across Desktop and CLI.

- **Unified workspace** — top-level sections sit above repository
controls in one rounded workspace, with navigation positioned close to
the page heading. README and Files retain branch selection; every
section has a labeled icon header, and Issues and Pull Requests expose
creation from a consistent right-aligned action.
- **Repository management** — the repository selector is always
available, including single-repository projects. Its integrated add flow
lets project owners create a repository manually or select an existing
repository without a separate toolbar button.
- **Readable work-item lists** — issue and pull-request rows use
plain-language context instead of opaque metadata. Files, commits,
issues, pull requests, channels, and contributors share consistent row
density and right-aligned timestamps, while deterministic
fallback-avatar colors keep participants distinct on light backgrounds.
Inbox pull-request metadata wraps between complete phrases and truncates
long channel names instead of compressing copy into narrow columns.
- **Reliable entity links** — projects, repositories, issues, pull
requests, and commits have canonical `buzz://` links, preview cards, OS
deep-link routing, and tab-aware navigation. Reopening the same link
re-applies its destination instead of leaving the user on a locally
selected tab.
- **Related conversations** — repository and work-item views surface
channels discussing the current entity, including participants, channel
navigation, message context, and an explicit notice when discovery
reaches its 500-result cap.
- **Reversible issue ownership** — trusted assignment and unassignment
events work across Desktop, Tauri, `buzz-sdk`, and `buzz issues`.
Assignees appear in project views and the assigned inbox, while
authorized users can remove assignments directly from the assignee row.

Assignment state is derived chronologically from labeled Nostr notes.
Issue authors and repository owners may change any assignee; other users
may only assign or unassign themselves. Shared golden fixtures keep
entity-link grammar and validation aligned across TypeScript and Rust.

The branch also updates `webbrowser` to the patched release for
RUSTSEC-2026-0257.

### Related issue

N/A.

### Testing

- [x] `just ci` — formatting, lint, typechecking, unit tests, and builds
passed
- [x] Full pre-push suite — organization, branch-skew, Desktop checks,
typechecking, and tests passed on the latest push
- [x] `cargo test -p buzz-cli` and focused `buzz-sdk` assignment tests
passed
- [x] Focused Tauri recipient-note and 500-result search-limit tests
passed
- [x] Desktop entity-link and issue-assignment unit tests passed
- [x] Playwright smoke coverage passed for assignment, repeated
entity-link navigation, repository create/select flows, section headers
and actions, timestamp alignment, timeline icons, sentence-style
issue/PR metadata, header spacing, avatar contrast, and Inbox metadata
at stacked and side-rail breakpoints
- [ ] Manual staging pass: link round-trips, Channels tab, assignment
flows, and inbox routing

### Screenshots

Pull requests explain who opened the request, where it lives, and which
branch it comes from; fallback avatars remain visually distinct.

![Pull request list with conversational
metadata](https://raw.githubusercontent.com/block/buzz/2a536de86f7e6f79b349d7bc147b2923ff2b817d/pr-5624--05-pr-list-metadata.png)

Issues use the same sentence-style hierarchy while keeping status and
recency easy to scan.

![Issue list with conversational
metadata](https://raw.githubusercontent.com/block/buzz/2a536de86f7e6f79b349d7bc147b2923ff2b817d/pr-5624--06-issue-list-metadata.png)

The wide Inbox detail keeps author, timestamp, and origin context
readable beside its metadata rail.

![Pull request Inbox detail with readable
metadata](https://raw.githubusercontent.com/block/buzz/e65b433e14b97c45365ed7b68ea402ec01d26615/pr-5624--02-pull-request-detail-wide.png)

[View the complete six-state Projects v3 screenshot
set](block#5624 (comment))
and [the compact/wide Inbox
comparison](block#5624 (comment)).


---

> Supersedes block#5624, whose head commit accumulated permanently-queued
required check suites (block-dco-check et al.) that GitHub never
dispatched. History flattened into a single signed-off commit on latest
main; tree verified byte-identical (`git merge-tree`) to merging the
original branch into main.

---------

Signed-off-by: Thomas Petersen <[email protected]>
Co-authored-by: Wintermute <3f1797424fd9ad6653a83665c660517777cd7f8c228c0d5907f49e01537f3ca5@buzz.block.builderlab.xyz>

* fix(desktop): restore the agent trading-card mint button (block#5900)

## Problem

PR block#5574's profile-panel redesign dropped `ProfileSummaryView`'s
`onCreateCard` prop — the only caller of `setCardMintTarget` — so the
entire Agent Trading Cards feature (block#3278) became unreachable from the
GUI while staying fully wired underneath: mint dialog, background job
store, viewer, gallery, composer chip, and the Rust
`mint_agent_card`/`save_agent_card` commands all survive at main. `git
log -S 'setCardMintTarget('` shows exactly two commits: the feature and
the accidental removal.

## Outcome

The mint trigger returns as a management row in the agent profile's Info
tab, directly under **Export agent**, gated `isBot && canManagePersona`
exactly like Duplicate/Export. Target resolution is byte-for-byte the
original logic: prefer the live instance pubkey, fall back to the
persona/definition id, allow locking only when an instance keypair
exists.

## Shape

- `UserProfileAgentManagementRows`: new optional `onCreateCard` row
(Sparkles icon, `user-profile-create-card-row`), placed after Export.
- Prop threaded `UserProfilePanel` → `ProfileSummaryView` →
`ProfileInfoTabContent` → management rows, mirroring `onExportAgent` at
every layer.
- The mint-target state + open callback move into a `useCardMint` hook
in `UserProfilePersonaDialogs` (beside the `CardMintTarget` type it
manages). This keeps `UserProfilePanel.tsx` at 999 lines — the file sits
at the size-ratchet cap and may not grow.

## Validation

- `pnpm check` green (biome, file-size ratchet, px-text,
pubkey-truncation).
- `pnpm typecheck` green.
- Full desktop unit suite: **4888 passed, 0 failed**.
- Profile e2e spec: **32 passed**, including the updated
management-row-order assertion and a new click → mint-dialog-visible →
Escape → closed exercise of the restored row.

Verified at `bff3110a0aeb3d63683eac9ed3e587829f9436da`, one commit atop
main `01f76ec97`.

Signed-off-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Co-authored-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>

* fix(ci): read Playwright version without nested shell quoting (block#5910)

## Summary

- replace the nested one-line shell quoting used to read the Playwright
package version
- write the resolved version to `GITHUB_OUTPUT` from a multiline shell
step

## Why

The `desktop-v0.5.12` release smoke job failed before executing tests
because Bash received escaped quotes inside command substitution and
parsed the Node expression as shell syntax.

## Validation

- `bash scripts/test-release-ref-contract.sh`
- isolated execution of the new shell fragment with a fixture
`@playwright/test/package.json`, producing `version=1.58.2`
- `git diff --check`

Signed-off-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

* chore(release): release Buzz Desktop version 0.5.13 (block#5912)

## Buzz Desktop release v0.5.13

- **Frozen main:** `09768100ec3420f0aa7cd278bd00fe0baab5de8d`
- **Reviewed candidate:** `a239e0f6793ac6e88ccf92cc231054090a9753cc`
- **Previous desktop release:** `desktop-v0.5.12`
- **Proposed immutable tag:** `desktop-v0.5.13`

This PR may be **squash merged** after the Desktop Release Candidate
check and all protected-branch checks pass. Merging authorizes
publication of the exact reviewed candidate; later or unrelated changes
on `main` cannot alter it.

The checked-in changelog accounts for every non-merge commit in the
release range. The Desktop tag points to the reviewed candidate commit,
not the later squash commit. Publication remains bound to that immutable
candidate tag.

Signed-off-by: Wes <[email protected]>
Co-authored-by: Release Automation <[email protected]>

* ci(release): remove desktop smoke gate (block#5914)

## Summary

- remove the GitHub-hosted desktop smoke job from the desktop release
workflow
- remove the smoke result from manifest assembly dependencies and
promotion conditions
- retain the local smoke tooling for future repair and targeted
validation

The first release execution of this gate spent its full 10-minute
Playwright timeout traversing the 10,000-row fixture, then produced a
987 MB diagnostics upload. All signed platform builds succeeded, but the
smoke prevented manifest publication. This restores the previously
established release boundary while the harness is made suitable for CI
separately.

### Testing

- parsed `.github/workflows/release.yml` with Ruby Psych and asserted
the smoke job/dependencies are absent
- `scripts/test-release-ref-contract.sh`
- exact pushed commit passed the repository pre-push hook

Signed-off-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

* chore(release): release Buzz Desktop version 0.5.14 (block#5917)

## Buzz Desktop release v0.5.14

- **Frozen main:** `1b3dbcaaea882eeea90359c1db02e306d2f4f50a`
- **Reviewed candidate:** `391495e7d347d20b67e39e3c240d17ef63c5c2c0`
- **Previous desktop release:** `desktop-v0.5.13`
- **Proposed immutable tag:** `desktop-v0.5.14`

This PR may be **squash merged** after the Desktop Release Candidate
check and all protected-branch checks pass. Merging authorizes
publication of the exact reviewed candidate; later or unrelated changes
on `main` cannot alter it.

The checked-in changelog accounts for every non-merge commit in the
release range. The Desktop tag points to the reviewed candidate commit,
not the later squash commit. Publication remains bound to that immutable
candidate tag.

Signed-off-by: Wes <[email protected]>
Co-authored-by: Release Automation <[email protected]>

* Polish mobile message threads and composer (block#5645)

## Summary

- refine mobile message metadata, search spacing, and Activity filter
semantics
- add channel-parity Latest navigation and stable tail following to
threads
- synchronize Android composer/keyboard geometry and keep Latest spacing
stable across IME transitions

## Validation

- `bin/just mobile-check`
- `bin/just mobile-test` (1,276 tests)
- Pixel 10 install/launch and channel/thread keyboard, Latest, tail, and
back-navigation review
- signed iPhone install/launch workflow

## Snapshots

See the review snapshots below.

---------

Signed-off-by: kenny lopez <[email protected]>
Signed-off-by: Kenny Lopez <[email protected]>
Signed-off-by: Princess Donut <b238ea756dee4d98afa5883fc7f1de61eeabe65bf700e3a5a5a80db5e42e2c2b@buzz.block.builderlab.xyz>
Co-authored-by: Fast Fizz <2df81cb51f05a9d5387ef24d7b9ecb8fcdfcd1c70ffabc67061c9596e1b5b1c4@buzz.block.builderlab.xyz>
Co-authored-by: Princess Donut <b238ea756dee4d98afa5883fc7f1de61eeabe65bf700e3a5a5a80db5e42e2c2b@buzz.block.builderlab.xyz>

* fix(desktop): hide the offcanvas-collapsed sidebar so it stops painting over the community rail (block#5947)

## Summary

Collapsing the sidebar left a phantom copy of it painted over the
community/relay rail — opaquely on flat themes (vesper et al., which
made the rail look *removed*), and as ghost fragments (muted search-box
fill, truncated channel-name tails) on the Buzz themes whose chrome is
intentionally transparent for the gradient.

**Cause:** block#4281 made the app-sidebar layer `overflow-visible` (the
huddle drawer needs to escape it). That removed the ancestor clipping
the offcanvas collapse relied on: the sidebar slides to `left:
-sidebar-width` but kept painting, exactly over the `z-0` rail (`z-10`
sidebar layer).

**Fix:** the offcanvas-collapsed sidebar container is now `invisible` +
`pointer-events-none`, with `visibility` added to the transition list so
the 200 ms slide-out still animates and the flip happens only at the
transition's end. Theme-independent; no per-theme CSS touched; the
huddle drawer's `overflow-visible` is preserved.

## Before / after

Left 420px of the app with the sidebar collapsed. Before = unpatched
`origin/main` @ 69107dc; after = this branch. Same seeded state, same
build pipeline (`build:e2e` between checkouts).

| theme | before (ghost sidebar over the rail) | after (rail clean: A /
B / + visible) |
|---|---|---|
| vesper |
![before-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-vesper.png)
|
![after-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-vesper.png)
|
| buzz |
![before-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz.png)
|
![after-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz.png)
|
| buzz-dark |
![before-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz-dark.png)
|
![after-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz-dark.png)
|

Before shots: ghost `⌘K` search chip + blue active-item pill painted
over the rail column; on vesper the opaque panel hides the rail buttons
entirely. After: the rail's community buttons (A, B) and `+` are visible
and clickable in all three themes.

Reported by Thomas P in #buzz-bugs:
buzz://message?channel=e62570dd-33ad-42c5-b92b-75f2689f9694&id=9ea401ca1d009f555ca4324e136f8d8d8156db2f8afa3ff89fd038d2c16260f7

cc @klopez4212 — this touches the layout your block#4281/block#5478 work shaped;
please confirm it doesn't defeat the huddle drawer or glass intentions.
The change deliberately hides only the *offcanvas-collapsed* container,
nothing in the expanded path.

## Test plan

- [x] New Playwright regression spec `sidebar-offcanvas-rail.spec.ts`
(buzz / buzz-dark / vesper): collapsed sidebar must be `visibility:
hidden` + `pointer-events: none`, community rail stays visible and
interactive. **Fails on unpatched build** (verified), passes with the
fix.
- [x] Full desktop unit suite: 4,954 pass / 0 fail
- [x] `pnpm typecheck`, `pnpm check` (biome + file-size ratchet +
px-text + pubkey-truncation) green
- [x] Before/after screenshots above captured via the e2e harness on
both builds

Signed-off-by: Thomas Petersen <[email protected]>
Co-authored-by: Wintermute <165f0c871dd2586bb18b6aa109eeaf57bb2132ff4d27b10120f4368a0f627022@buzz.block.builderlab.xyz>

* ci: refresh Buzz 0.5.14 fork baseline

Signed-off-by: Ross Hartmann <[email protected]>

* fix(acp): preserve same-turn action ordering

Signed-off-by: Ross Hartmann <[email protected]>

---------

Signed-off-by: Thomas Petersen <[email protected]>
Signed-off-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Signed-off-by: Wes <[email protected]>
Signed-off-by: kenny lopez <[email protected]>
Signed-off-by: Kenny Lopez <[email protected]>
Signed-off-by: Princess Donut <b238ea756dee4d98afa5883fc7f1de61eeabe65bf700e3a5a5a80db5e42e2c2b@buzz.block.builderlab.xyz>
Signed-off-by: Ross Hartmann <[email protected]>
Co-authored-by: thomaspblock <[email protected]>
Co-authored-by: Wintermute <3f1797424fd9ad6653a83665c660517777cd7f8c228c0d5907f49e01537f3ca5@buzz.block.builderlab.xyz>
Co-authored-by: Tyler <[email protected]>
Co-authored-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Co-authored-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>
Co-authored-by: Release Automation <[email protected]>
Co-authored-by: klopez4212 <[email protected]>
Co-authored-by: Fast Fizz <2df81cb51f05a9d5387ef24d7b9ecb8fcdfcd1c70ffabc67061c9596e1b5b1c4@buzz.block.builderlab.xyz>
Co-authored-by: Princess Donut <b238ea756dee4d98afa5883fc7f1de61eeabe65bf700e3a5a5a80db5e42e2c2b@buzz.block.builderlab.xyz>
Co-authored-by: Wintermute <165f0c871dd2586bb18b6aa109eeaf57bb2132ff4d27b10120f4368a0f627022@buzz.block.builderlab.xyz>

…ng over the community rail (block#5947)

## Summary

Collapsing the sidebar left a phantom copy of it painted over the
community/relay rail — opaquely on flat themes (vesper et al., which
made the rail look *removed*), and as ghost fragments (muted search-box
fill, truncated channel-name tails) on the Buzz themes whose chrome is
intentionally transparent for the gradient.

**Cause:** block#4281 made the app-sidebar layer `overflow-visible` (the
huddle drawer needs to escape it). That removed the ancestor clipping
the offcanvas collapse relied on: the sidebar slides to `left:
-sidebar-width` but kept painting, exactly over the `z-0` rail (`z-10`
sidebar layer).

**Fix:** the offcanvas-collapsed sidebar container is now `invisible` +
`pointer-events-none`, with `visibility` added to the transition list so
the 200 ms slide-out still animates and the flip happens only at the
transition's end. Theme-independent; no per-theme CSS touched; the
huddle drawer's `overflow-visible` is preserved.

## Before / after

Left 420px of the app with the sidebar collapsed. Before = unpatched
`origin/main` @ 69107dc; after = this branch. Same seeded state, same
build pipeline (`build:e2e` between checkouts).

| theme | before (ghost sidebar over the rail) | after (rail clean: A /
B / + visible) |
|---|---|---|
| vesper |
![before-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-vesper.png)
|
![after-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-vesper.png)
|
| buzz |
![before-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz.png)
|
![after-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz.png)
|
| buzz-dark |
![before-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz-dark.png)
|
![after-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz-dark.png)
|

Before shots: ghost `⌘K` search chip + blue active-item pill painted
over the rail column; on vesper the opaque panel hides the rail buttons
entirely. After: the rail's community buttons (A, B) and `+` are visible
and clickable in all three themes.

Reported by Thomas P in #buzz-bugs:
buzz://message?channel=e62570dd-33ad-42c5-b92b-75f2689f9694&id=9ea401ca1d009f555ca4324e136f8d8d8156db2f8afa3ff89fd038d2c16260f7

cc @klopez4212 — this touches the layout your block#4281/block#5478 work shaped;
please confirm it doesn't defeat the huddle drawer or glass intentions.
The change deliberately hides only the *offcanvas-collapsed* container,
nothing in the expanded path.

## Test plan

- [x] New Playwright regression spec `sidebar-offcanvas-rail.spec.ts`
(buzz / buzz-dark / vesper): collapsed sidebar must be `visibility:
hidden` + `pointer-events: none`, community rail stays visible and
interactive. **Fails on unpatched build** (verified), passes with the
fix.
- [x] Full desktop unit suite: 4,954 pass / 0 fail
- [x] `pnpm typecheck`, `pnpm check` (biome + file-size ratchet +
px-text + pubkey-truncation) green
- [x] Before/after screenshots above captured via the e2e harness on
both builds

Signed-off-by: Thomas Petersen <[email protected]>
Co-authored-by: Wintermute <165f0c871dd2586bb18b6aa109eeaf57bb2132ff4d27b10120f4368a0f627022@buzz.block.builderlab.xyz>
Signed-off-by: bhargavms <[email protected]>

## Problem

Humans cannot interrupt agents in huddles: agent TTS keeps playing over
a talking human (reported by @tlongwell-block, 2026-08-20). The full
requirement: **any human talking — local or remote, any input mode, any
audio rig — must interrupt any agent on the huddle.**

## This is a restoration, not a new feature

- **`b29c8cdaa` (#4281, 2026-08-04) deleted working local VAD
barge-in.** The pre-image shipped `BARGE_IN_DEBOUNCE_FRAMES = 20` (320
ms sustained speech cancels TTS), live at the production call site
(`pipeline.rs` passed `Some(tts_cancel)` unconditionally). The same day,
`ce3cf3cd2` (#4694) flipped the default input mode from VAD to
push-to-talk, which masked the loss.
- **`068a83b09` (#5671, 2026-08-13) removed the remaining TTS-awareness
plumbing from STT** (deliberately, to keep transcribing over agent audio
— a good change). Consequence: this PR is a re-plumb through the
`PlaybackCoordinator` from #6341, not a revert.
- A gap that **never** worked is also closed: PTT-mode users who open
the mic via the mute button (`manual_mic_unmuted` postdates the
deletion) transcribed fine but could not barge in.

## Design

All floor state lives under the single `PlaybackCoordinator` lock —
onset acceptance, epoch bump, synthesis invalidation, player
replacement, and output lease are one committed transition. No cancel
flag can be observed out of order with the playback state it describes.

- **Human floor**: local + per-peer remote ownership with epoch
invalidation. Onset cancels playback by queue replacement; late
synthesis for a stale epoch cannot append or restart.
- **Local onset (VAD)**: on an **isolated output route** (all CoreAudio
output-stream terminals report headphones), a confirmed short onset
interrupts immediately — no echo path exists. On a **coupled route**
(speakers, unknown, virtual, mixed), the restored **20-frame / 320 ms
sustained-speech debounce** discriminates a real human from speaker
bleed; the deleted code's comment records that 80 ms was tried and
false-triggered on laptop speakers. Route classification is queried
fresh at each onset (never cached — default-device re-routing mid-huddle
would strand a stale verdict).
- **Mic-open gate is per-frame**: barge-in observes on any frame where
the mic is actually open — pure VAD mode, or PTT with the mic manually
unmuted (key-held frames defer to the shortcut's own cancel).
- **Remote onset**: sustained non-DTX frames from a peer enter the same
persistent floor (independent of whether playback is live — a human
speaking while TTS is idle blocks late-arriving synthesis from starting
over them). Release on sustained DTX/absence, peer departure, and
recv-loop exit, with guards so a vanished peer cannot wedge the floor.
- **Output lease**: accepted appends renew an `Active` lease;
drain/cancel/onset start a 100 ms tail hangover (conservative against
measured ~12 ms/~1 ms CoreAudio tails), so speaker-tail bleed in the
just-drained window cannot self-trigger the coupled path.

## Known limitations (phase 2 pointers)

- Coupled-route mid-output barge-in pays the 320 ms debounce; a
playback-reference echo discriminator would shorten it.
- A speakers-rig participant's bleed can enter their mic and hold the
floor for other machines (bounded by release debounce).
- Non-macOS routes classify as coupled (fail-safe).

## Verification

- Full `buzz-desktop --lib` suite at head `b0459ae4a`: **2707 passed, 0
failed, 18 ignored** (pinned cargo 1.95.0).
- Exact CI recipe `just desktop-tauri-clippy`: PASS at head; base arm at
merge-base `b728a2af3` confirms the two
`#[allow(clippy::too_many_arguments)]`s cover branch-caused threshold
crossings (human_floor threading), not inherited noise.
- Regression tests pin: 20-frame threshold + reset-on-gap, short-coupled
rejection, sustained-coupled acceptance, coupled-idle acceptance,
remote-idle delayed-TTS rejection, output-tail hangover boundary (during
= rejected, after = accepted), isolated onset, per-frame mic-open gate
truth table, PTT+manual-unmute sustained coupled acquisition.
- `LocalBargeIn::observe` is covered as two joined halves (gate truth
table in `local_barge_in.rs`, floor transition in `tts_playback.rs`);
its body is a straight-line wrapper around a live CoreAudio query, left
uninjected deliberately.
- Coverage precision (mutation-verified):
`manual_open_ptt_sustained_speech_acquires_coupled_floor` pins the gate
→ 20-frame debounce → acquire → floor-blocked chain on the
coupled-**idle** cell. The live-output override leg is carried by
`sustained_coupled_speech_overrides_live_output_suppression`
(tts_playback.rs); the joiner's `sustained_coupled` argument is not
load-bearing there (flipping it to `false` leaves the test green, while
shortening the debounce by one frame turns it red).
- Live arms in progress: pre-regression build `b29c8cdaa^` staged to
confirm the deleted mechanism worked; two-endpoint remote-leg test
pending a second human.

## Commits

1. `fb681a5a9` — restore human barge-in (coordinator floor, lease, route
isolation, remote floor, 320 ms coupled debounce)
2. `b4265418e` — enable barge-in for manually opened mics (per-frame
gate; closes the PTT-open-mic gap)
3. `886489f2b` — extract local barge-in policy module (file-size
ratchet; also hoists the CoreAudio route query from per-frame to
per-onset, named in the commit message)
4. `b0459ae4a` — two targeted clippy allows for the widened worker
signatures

## Credits

Built by **Wren**. Regression archaeology and the PTT-open-mic gap by
**Dawn** (who also killed her own first fix as vacuous and caught a
clippy blocker before it hit CI). Review blockers (coordinator
serialization, idle-onset floors, output lease) by **Mari**. Live rig
verification by **Max**. Coordination and verification by **Eva**.
Opened by Eva with Tyler's explicit direction; commits carry agent
trailers.

---------

Signed-off-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Co-authored-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>

The 10 ms playout tick uses MissedTickBehavior::Delay, which never
shortens the ticks that follow a missed one. Windows timers default to
a 15.6 ms resolution and tokio intervals fire ~14.6 ms late there on
average (tokio-rs/tokio#5021), so the loop settles at ~62 of the 100
pulls/s the pipeline needs. The per-peer rodio queues run dry (audible
gaps, dropped words) while NetEq stays full and time-compresses
playback (metallic, sped-up voices) - on every peer, regardless of
network quality. Matches the choppy-audio reports in block#2652; block#4281
changed the drop threshold but not the tick rate.

Measured on Windows 11 with a standalone reproduction of this loop:
62.4 ticks/s with Delay at default resolution, 100.2 ticks/s with
timeBeginPeriod(1) raised for the loop lifetime and Burst making up
missed ticks. Catch-up bursts are absorbed by the existing queue
recovery (hysteresis 10-4, emergency trim at 30).

Signed-off-by: kaalph <[email protected]>

The 10 ms playout tick uses MissedTickBehavior::Delay, which never
shortens the ticks that follow a missed one. Windows timers default to
a 15.6 ms resolution and tokio intervals fire ~14.6 ms late there on
average (tokio-rs/tokio#5021), so the loop settles at ~62 of the 100
pulls/s the pipeline needs. The per-peer rodio queues run dry (audible
gaps, dropped words) while NetEq stays full and time-compresses
playback (metallic, sped-up voices) - on every peer, regardless of
network quality. Matches the choppy-audio reports in block#2652; block#4281
changed the drop threshold but not the tick rate.

Measured on Windows 11 with a standalone reproduction of this loop:
62.4 ticks/s with Delay at default resolution, 100.2 ticks/s with
timeBeginPeriod(1) raised for the loop lifetime and Burst making up
missed ticks. Catch-up bursts are absorbed by the existing queue
recovery (hysteresis 10-4, emergency trim at 30).

Signed-off-by: kaalph <[email protected]>

## Summary

- open Huddles in a focused companion window with a clean handoff back
to the in-app drawer and backing channel
- redesign the participant film strip, sidebar control, transcript
surface, and themed shell treatment
- preserve microphone and device control across windows, start agent
voice on the first reply, and show agent speaking activity in the film
strip
- give each agent a distinct session voice, beginning with the
configured default, plus compact per-agent text-to-speech and voice
controls
- enroll only agents explicitly mentioned or deliberately added through
an agent panel into the live Huddle roster
- keep temporary Huddle channels out of the sidebar unless the user
explicitly brings one into the main app
- remove Huddle-only avatar policy badges and filter short silence or
noise segments before speech-to-text posts

## Why

The previous flow exposed the temporary channel as product UI, obscured
who was present or speaking, and split transcript and audio state
between the main and companion windows. This keeps backing channels as
implementation details unless a user explicitly brings a Huddle into the
app, while sharing the live conversation and audio lifecycle across both
surfaces. Agent participants now join only after an explicit invitation,
distinct voices make multi-agent Huddles easier to follow, and short
microphone noise no longer becomes stray transcript messages.

## Validation

- `pnpm check`
- `pnpm build:e2e`
- `pnpm exec playwright test tests/e2e/huddle-transcription.spec.ts
--project=smoke` (13 passed)
- Huddle sidebar visibility unit coverage (4 passed)
- focused managed-agent and persona-mention E2E coverage (2 passed)
- `pnpm test` (3,910 passed)
- `cargo clippy --manifest-path desktop/src-tauri/Cargo.toml
--all-targets -- -D warnings`
- `cargo test --manifest-path desktop/src-tauri/Cargo.toml` (2,093
passed, 14 ignored; 3 diagnostics passed)

---------

Signed-off-by: kenny lopez <[email protected]>
Signed-off-by: Wes <[email protected]>
Co-authored-by: Wes <[email protected]>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>

…ng over the community rail (block#5947)

## Summary

Collapsing the sidebar left a phantom copy of it painted over the
community/relay rail — opaquely on flat themes (vesper et al., which
made the rail look *removed*), and as ghost fragments (muted search-box
fill, truncated channel-name tails) on the Buzz themes whose chrome is
intentionally transparent for the gradient.

**Cause:** block#4281 made the app-sidebar layer `overflow-visible` (the
huddle drawer needs to escape it). That removed the ancestor clipping
the offcanvas collapse relied on: the sidebar slides to `left:
-sidebar-width` but kept painting, exactly over the `z-0` rail (`z-10`
sidebar layer).

**Fix:** the offcanvas-collapsed sidebar container is now `invisible` +
`pointer-events-none`, with `visibility` added to the transition list so
the 200 ms slide-out still animates and the flip happens only at the
transition's end. Theme-independent; no per-theme CSS touched; the
huddle drawer's `overflow-visible` is preserved.

## Before / after

Left 420px of the app with the sidebar collapsed. Before = unpatched
`origin/main` @ 8eb5cb5; after = this branch. Same seeded state, same
build pipeline (`build:e2e` between checkouts).

| theme | before (ghost sidebar over the rail) | after (rail clean: A /
B / + visible) |
|---|---|---|
| vesper |
![before-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-vesper.png)
|
![after-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-vesper.png)
|
| buzz |
![before-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz.png)
|
![after-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz.png)
|
| buzz-dark |
![before-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz-dark.png)
|
![after-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz-dark.png)
|

Before shots: ghost `⌘K` search chip + blue active-item pill painted
over the rail column; on vesper the opaque panel hides the rail buttons
entirely. After: the rail's community buttons (A, B) and `+` are visible
and clickable in all three themes.

Reported by Thomas P in #buzz-bugs:
buzz://message?channel=e62570dd-33ad-42c5-b92b-75f2689f9694&id=9ea401ca1d009f555ca4324e136f8d8d8156db2f8afa3ff89fd038d2c16260f7

cc @klopez4212 — this touches the layout your block#4281/block#5478 work shaped;
please confirm it doesn't defeat the huddle drawer or glass intentions.
The change deliberately hides only the *offcanvas-collapsed* container,
nothing in the expanded path.

## Test plan

- [x] New Playwright regression spec `sidebar-offcanvas-rail.spec.ts`
(buzz / buzz-dark / vesper): collapsed sidebar must be `visibility:
hidden` + `pointer-events: none`, community rail stays visible and
interactive. **Fails on unpatched build** (verified), passes with the
fix.
- [x] Full desktop unit suite: 4,954 pass / 0 fail
- [x] `pnpm typecheck`, `pnpm check` (biome + file-size ratchet +
px-text + pubkey-truncation) green
- [x] Before/after screenshots above captured via the e2e harness on
both builds

Signed-off-by: Thomas Petersen <[email protected]>
Co-authored-by: Wintermute <165f0c871dd2586bb18b6aa109eeaf57bb2132ff4d27b10120f4368a0f627022@buzz.block.builderlab.xyz>

…#6431)

## Problem

Humans cannot interrupt agents in huddles: agent TTS keeps playing over
a talking human (reported by @tlongwell-block, 2026-08-20). The full
requirement: **any human talking — local or remote, any input mode, any
audio rig — must interrupt any agent on the huddle.**

## This is a restoration, not a new feature

- **`af68c34e7` (block#4281, 2026-08-04) deleted working local VAD
barge-in.** The pre-image shipped `BARGE_IN_DEBOUNCE_FRAMES = 20` (320
ms sustained speech cancels TTS), live at the production call site
(`pipeline.rs` passed `Some(tts_cancel)` unconditionally). The same day,
`5a361f99e` (block#4694) flipped the default input mode from VAD to
push-to-talk, which masked the loss.
- **`5214a0755` (block#5671, 2026-08-13) removed the remaining TTS-awareness
plumbing from STT** (deliberately, to keep transcribing over agent audio
— a good change). Consequence: this PR is a re-plumb through the
`PlaybackCoordinator` from block#6341, not a revert.
- A gap that **never** worked is also closed: PTT-mode users who open
the mic via the mute button (`manual_mic_unmuted` postdates the
deletion) transcribed fine but could not barge in.

## Design

All floor state lives under the single `PlaybackCoordinator` lock —
onset acceptance, epoch bump, synthesis invalidation, player
replacement, and output lease are one committed transition. No cancel
flag can be observed out of order with the playback state it describes.

- **Human floor**: local + per-peer remote ownership with epoch
invalidation. Onset cancels playback by queue replacement; late
synthesis for a stale epoch cannot append or restart.
- **Local onset (VAD)**: on an **isolated output route** (all CoreAudio
output-stream terminals report headphones), a confirmed short onset
interrupts immediately — no echo path exists. On a **coupled route**
(speakers, unknown, virtual, mixed), the restored **20-frame / 320 ms
sustained-speech debounce** discriminates a real human from speaker
bleed; the deleted code's comment records that 80 ms was tried and
false-triggered on laptop speakers. Route classification is queried
fresh at each onset (never cached — default-device re-routing mid-huddle
would strand a stale verdict).
- **Mic-open gate is per-frame**: barge-in observes on any frame where
the mic is actually open — pure VAD mode, or PTT with the mic manually
unmuted (key-held frames defer to the shortcut's own cancel).
- **Remote onset**: sustained non-DTX frames from a peer enter the same
persistent floor (independent of whether playback is live — a human
speaking while TTS is idle blocks late-arriving synthesis from starting
over them). Release on sustained DTX/absence, peer departure, and
recv-loop exit, with guards so a vanished peer cannot wedge the floor.
- **Output lease**: accepted appends renew an `Active` lease;
drain/cancel/onset start a 100 ms tail hangover (conservative against
measured ~12 ms/~1 ms CoreAudio tails), so speaker-tail bleed in the
just-drained window cannot self-trigger the coupled path.

## Known limitations (phase 2 pointers)

- Coupled-route mid-output barge-in pays the 320 ms debounce; a
playback-reference echo discriminator would shorten it.
- A speakers-rig participant's bleed can enter their mic and hold the
floor for other machines (bounded by release debounce).
- Non-macOS routes classify as coupled (fail-safe).

## Verification

- Full `buzz-desktop --lib` suite at head `b0459ae4a`: **2707 passed, 0
failed, 18 ignored** (pinned cargo 1.95.0).
- Exact CI recipe `just desktop-tauri-clippy`: PASS at head; base arm at
merge-base `8f2b659fe` confirms the two
`#[allow(clippy::too_many_arguments)]`s cover branch-caused threshold
crossings (human_floor threading), not inherited noise.
- Regression tests pin: 20-frame threshold + reset-on-gap, short-coupled
rejection, sustained-coupled acceptance, coupled-idle acceptance,
remote-idle delayed-TTS rejection, output-tail hangover boundary (during
= rejected, after = accepted), isolated onset, per-frame mic-open gate
truth table, PTT+manual-unmute sustained coupled acquisition.
- `LocalBargeIn::observe` is covered as two joined halves (gate truth
table in `local_barge_in.rs`, floor transition in `tts_playback.rs`);
its body is a straight-line wrapper around a live CoreAudio query, left
uninjected deliberately.
- Coverage precision (mutation-verified):
`manual_open_ptt_sustained_speech_acquires_coupled_floor` pins the gate
→ 20-frame debounce → acquire → floor-blocked chain on the
coupled-**idle** cell. The live-output override leg is carried by
`sustained_coupled_speech_overrides_live_output_suppression`
(tts_playback.rs); the joiner's `sustained_coupled` argument is not
load-bearing there (flipping it to `false` leaves the test green, while
shortening the debounce by one frame turns it red).
- Live arms in progress: pre-regression build `af68c34e7^` staged to
confirm the deleted mechanism worked; two-endpoint remote-leg test
pending a second human.

## Commits

1. `fb681a5a9` — restore human barge-in (coordinator floor, lease, route
isolation, remote floor, 320 ms coupled debounce)
2. `b4265418e` — enable barge-in for manually opened mics (per-frame
gate; closes the PTT-open-mic gap)
3. `886489f2b` — extract local barge-in policy module (file-size
ratchet; also hoists the CoreAudio route query from per-frame to
per-onset, named in the commit message)
4. `b0459ae4a` — two targeted clippy allows for the widened worker
signatures

## Credits

Built by **Wren**. Regression archaeology and the PTT-open-mic gap by
**Dawn** (who also killed her own first fix as vacuous and caught a
clippy blocker before it hit CI). Review blockers (coordinator
serialization, idle-onset floors, output lease) by **Mari**. Live rig
verification by **Max**. Coordination and verification by **Eva**.
Opened by Eva with Tyler's explicit direction; commits carry agent
trailers.

---------

Signed-off-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Co-authored-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>

…ng over the community rail (block#5947)

## Summary

Collapsing the sidebar left a phantom copy of it painted over the
community/relay rail — opaquely on flat themes (vesper et al., which
made the rail look *removed*), and as ghost fragments (muted search-box
fill, truncated channel-name tails) on the Buzz themes whose chrome is
intentionally transparent for the gradient.

**Cause:** block#4281 made the app-sidebar layer `overflow-visible` (the
huddle drawer needs to escape it). That removed the ancestor clipping
the offcanvas collapse relied on: the sidebar slides to `left:
-sidebar-width` but kept painting, exactly over the `z-0` rail (`z-10`
sidebar layer).

**Fix:** the offcanvas-collapsed sidebar container is now `invisible` +
`pointer-events-none`, with `visibility` added to the transition list so
the 200 ms slide-out still animates and the flip happens only at the
transition's end. Theme-independent; no per-theme CSS touched; the
huddle drawer's `overflow-visible` is preserved.

## Before / after

Left 420px of the app with the sidebar collapsed. Before = unpatched
`origin/main` @ 69107dc; after = this branch. Same seeded state, same
build pipeline (`build:e2e` between checkouts).

| theme | before (ghost sidebar over the rail) | after (rail clean: A /
B / + visible) |
|---|---|---|
| vesper |
![before-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-vesper.png)
|
![after-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-vesper.png)
|
| buzz |
![before-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz.png)
|
![after-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz.png)
|
| buzz-dark |
![before-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz-dark.png)
|
![after-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz-dark.png)
|

Before shots: ghost `⌘K` search chip + blue active-item pill painted
over the rail column; on vesper the opaque panel hides the rail buttons
entirely. After: the rail's community buttons (A, B) and `+` are visible
and clickable in all three themes.

Reported by Thomas P in #buzz-bugs:
buzz://message?channel=e62570dd-33ad-42c5-b92b-75f2689f9694&id=9ea401ca1d009f555ca4324e136f8d8d8156db2f8afa3ff89fd038d2c16260f7

cc @klopez4212 — this touches the layout your block#4281/block#5478 work shaped;
please confirm it doesn't defeat the huddle drawer or glass intentions.
The change deliberately hides only the *offcanvas-collapsed* container,
nothing in the expanded path.

## Test plan

- [x] New Playwright regression spec `sidebar-offcanvas-rail.spec.ts`
(buzz / buzz-dark / vesper): collapsed sidebar must be `visibility:
hidden` + `pointer-events: none`, community rail stays visible and
interactive. **Fails on unpatched build** (verified), passes with the
fix.
- [x] Full desktop unit suite: 4,954 pass / 0 fail
- [x] `pnpm typecheck`, `pnpm check` (biome + file-size ratchet +
px-text + pubkey-truncation) green
- [x] Before/after screenshots above captured via the e2e harness on
both builds

Signed-off-by: Thomas Petersen <[email protected]>
Co-authored-by: Wintermute <165f0c871dd2586bb18b6aa109eeaf57bb2132ff4d27b10120f4368a0f627022@buzz.block.builderlab.xyz>

…ng over the community rail (block#5947)

## Summary

Collapsing the sidebar left a phantom copy of it painted over the
community/relay rail — opaquely on flat themes (vesper et al., which
made the rail look *removed*), and as ghost fragments (muted search-box
fill, truncated channel-name tails) on the Buzz themes whose chrome is
intentionally transparent for the gradient.

**Cause:** block#4281 made the app-sidebar layer `overflow-visible` (the
huddle drawer needs to escape it). That removed the ancestor clipping
the offcanvas collapse relied on: the sidebar slides to `left:
-sidebar-width` but kept painting, exactly over the `z-0` rail (`z-10`
sidebar layer).

**Fix:** the offcanvas-collapsed sidebar container is now `invisible` +
`pointer-events-none`, with `visibility` added to the transition list so
the 200 ms slide-out still animates and the flip happens only at the
transition's end. Theme-independent; no per-theme CSS touched; the
huddle drawer's `overflow-visible` is preserved.

## Before / after

Left 420px of the app with the sidebar collapsed. Before = unpatched
`origin/main` @ 69107dc; after = this branch. Same seeded state, same
build pipeline (`build:e2e` between checkouts).

| theme | before (ghost sidebar over the rail) | after (rail clean: A /
B / + visible) |
|---|---|---|
| vesper |
![before-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-vesper.png)
|
![after-vesper](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-vesper.png)
|
| buzz |
![before-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz.png)
|
![after-buzz](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz.png)
|
| buzz-dark |
![before-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--before-buzz-dark.png)
|
![after-buzz-dark](https://raw.githubusercontent.com/block/buzz/3f98c576e062e51d976940725d84b4e0be7fd53c/pr-5947--after-buzz-dark.png)
|

Before shots: ghost `⌘K` search chip + blue active-item pill painted
over the rail column; on vesper the opaque panel hides the rail buttons
entirely. After: the rail's community buttons (A, B) and `+` are visible
and clickable in all three themes.

Reported by Thomas P in #buzz-bugs:
buzz://message?channel=e62570dd-33ad-42c5-b92b-75f2689f9694&id=9ea401ca1d009f555ca4324e136f8d8d8156db2f8afa3ff89fd038d2c16260f7

cc @klopez4212 — this touches the layout your block#4281/block#5478 work shaped;
please confirm it doesn't defeat the huddle drawer or glass intentions.
The change deliberately hides only the *offcanvas-collapsed* container,
nothing in the expanded path.

## Test plan

- [x] New Playwright regression spec `sidebar-offcanvas-rail.spec.ts`
(buzz / buzz-dark / vesper): collapsed sidebar must be `visibility:
hidden` + `pointer-events: none`, community rail stays visible and
interactive. **Fails on unpatched build** (verified), passes with the
fix.
- [x] Full desktop unit suite: 4,954 pass / 0 fail
- [x] `pnpm typecheck`, `pnpm check` (biome + file-size ratchet +
px-text + pubkey-truncation) green
- [x] Before/after screenshots above captured via the e2e harness on
both builds

Signed-off-by: Thomas Petersen <[email protected]>
Co-authored-by: Wintermute <165f0c871dd2586bb18b6aa109eeaf57bb2132ff4d27b10120f4368a0f627022@buzz.block.builderlab.xyz>

…#6431)

## Problem

Humans cannot interrupt agents in huddles: agent TTS keeps playing over
a talking human (reported by @tlongwell-block, 2026-08-20). The full
requirement: **any human talking — local or remote, any input mode, any
audio rig — must interrupt any agent on the huddle.**

## This is a restoration, not a new feature

- **`b29c8cdaa` (block#4281, 2026-08-04) deleted working local VAD
barge-in.** The pre-image shipped `BARGE_IN_DEBOUNCE_FRAMES = 20` (320
ms sustained speech cancels TTS), live at the production call site
(`pipeline.rs` passed `Some(tts_cancel)` unconditionally). The same day,
`ce3cf3cd2` (block#4694) flipped the default input mode from VAD to
push-to-talk, which masked the loss.
- **`068a83b09` (block#5671, 2026-08-13) removed the remaining TTS-awareness
plumbing from STT** (deliberately, to keep transcribing over agent audio
— a good change). Consequence: this PR is a re-plumb through the
`PlaybackCoordinator` from block#6341, not a revert.
- A gap that **never** worked is also closed: PTT-mode users who open
the mic via the mute button (`manual_mic_unmuted` postdates the
deletion) transcribed fine but could not barge in.

## Design

All floor state lives under the single `PlaybackCoordinator` lock —
onset acceptance, epoch bump, synthesis invalidation, player
replacement, and output lease are one committed transition. No cancel
flag can be observed out of order with the playback state it describes.

- **Human floor**: local + per-peer remote ownership with epoch
invalidation. Onset cancels playback by queue replacement; late
synthesis for a stale epoch cannot append or restart.
- **Local onset (VAD)**: on an **isolated output route** (all CoreAudio
output-stream terminals report headphones), a confirmed short onset
interrupts immediately — no echo path exists. On a **coupled route**
(speakers, unknown, virtual, mixed), the restored **20-frame / 320 ms
sustained-speech debounce** discriminates a real human from speaker
bleed; the deleted code's comment records that 80 ms was tried and
false-triggered on laptop speakers. Route classification is queried
fresh at each onset (never cached — default-device re-routing mid-huddle
would strand a stale verdict).
- **Mic-open gate is per-frame**: barge-in observes on any frame where
the mic is actually open — pure VAD mode, or PTT with the mic manually
unmuted (key-held frames defer to the shortcut's own cancel).
- **Remote onset**: sustained non-DTX frames from a peer enter the same
persistent floor (independent of whether playback is live — a human
speaking while TTS is idle blocks late-arriving synthesis from starting
over them). Release on sustained DTX/absence, peer departure, and
recv-loop exit, with guards so a vanished peer cannot wedge the floor.
- **Output lease**: accepted appends renew an `Active` lease;
drain/cancel/onset start a 100 ms tail hangover (conservative against
measured ~12 ms/~1 ms CoreAudio tails), so speaker-tail bleed in the
just-drained window cannot self-trigger the coupled path.

## Known limitations (phase 2 pointers)

- Coupled-route mid-output barge-in pays the 320 ms debounce; a
playback-reference echo discriminator would shorten it.
- A speakers-rig participant's bleed can enter their mic and hold the
floor for other machines (bounded by release debounce).
- Non-macOS routes classify as coupled (fail-safe).

## Verification

- Full `buzz-desktop --lib` suite at head `b0459ae4a`: **2707 passed, 0
failed, 18 ignored** (pinned cargo 1.95.0).
- Exact CI recipe `just desktop-tauri-clippy`: PASS at head; base arm at
merge-base `b728a2af3` confirms the two
`#[allow(clippy::too_many_arguments)]`s cover branch-caused threshold
crossings (human_floor threading), not inherited noise.
- Regression tests pin: 20-frame threshold + reset-on-gap, short-coupled
rejection, sustained-coupled acceptance, coupled-idle acceptance,
remote-idle delayed-TTS rejection, output-tail hangover boundary (during
= rejected, after = accepted), isolated onset, per-frame mic-open gate
truth table, PTT+manual-unmute sustained coupled acquisition.
- `LocalBargeIn::observe` is covered as two joined halves (gate truth
table in `local_barge_in.rs`, floor transition in `tts_playback.rs`);
its body is a straight-line wrapper around a live CoreAudio query, left
uninjected deliberately.
- Coverage precision (mutation-verified):
`manual_open_ptt_sustained_speech_acquires_coupled_floor` pins the gate
→ 20-frame debounce → acquire → floor-blocked chain on the
coupled-**idle** cell. The live-output override leg is carried by
`sustained_coupled_speech_overrides_live_output_suppression`
(tts_playback.rs); the joiner's `sustained_coupled` argument is not
load-bearing there (flipping it to `false` leaves the test green, while
shortening the debounce by one frame turns it red).
- Live arms in progress: pre-regression build `b29c8cdaa^` staged to
confirm the deleted mechanism worked; two-endpoint remote-leg test
pending a second human.

## Commits

1. `fb681a5a9` — restore human barge-in (coordinator floor, lease, route
isolation, remote floor, 320 ms coupled debounce)
2. `b4265418e` — enable barge-in for manually opened mics (per-frame
gate; closes the PTT-open-mic gap)
3. `886489f2b` — extract local barge-in policy module (file-size
ratchet; also hoists the CoreAudio route query from per-frame to
per-onset, named in the commit message)
4. `b0459ae4a` — two targeted clippy allows for the widened worker
signatures

## Credits

Built by **Wren**. Regression archaeology and the PTT-open-mic gap by
**Dawn** (who also killed her own first fix as vacuous and caught a
clippy blocker before it hit CI). Review blockers (coordinator
serialization, idle-onset floors, output lease) by **Mari**. Live rig
verification by **Max**. Coordination and verification by **Eva**.
Opened by Eva with Tyler's explicit direction; commits carry agent
trailers.

---------

Signed-off-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Co-authored-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>