Go to file
Matt Toohey 1eb67bb525 fix(desktop): fix the product races behind four E2E flakes (#8680)
## Summary

Fixes the four desktop product races behind recurring E2E flakes in the
21–28 Oct 2026 CI audit. Each fix has a regression test:

- In the channel browser, Enter acted on a stale deferred list.
- A stale autocomplete swallowed the Enter that submits an empty edit.
- Same-second messages swapped order on screen.
- The unread pill read "Jump to latest" instead of "N new messages".

Two E2E specs that failed for the same ordering reason also get
test-side fixes.

This is one of seven PRs split out of the `flakey-tests` branch
(siblings below).

### Related issue

None found for these. Draft #8559 does not fix the voice-note flake.
Draft #8661 (Refs #6733) closes the `#channel` list synchronously when
an edit is cleared. §11 below also covers that case, plus the `@mention`
and `:emoji` lists, so the two PRs overlap in `useChannelLinks.ts` and
`empty-edit-delete.spec.ts`.

### Flaky tests fixed

| Test | Failing runs | Cause | Change |
|---|---|---|---|
| §5 `channel-browser.spec.ts:402` "Enter with no matches jumps to
create" | 15 of 50 main runs, e.g. 40597484657, 40582387515,
40566841699, 40367495044 | **Product.** The list renders from
`useDeferredValue(query)`. An Enter that landed before the deferred
render read the old list, so it opened a stale match or skipped the
create row. | While the deferred query lags, Enter filters the live
query with the same `selectBrowserChannels` the list renders from. |
| §11 `empty-edit-delete.spec.ts:88`, `:111`, `:132` | :88 in 3 jobs,
:111 in 3, :132 in 2. :111 failed all 3 attempts on main 39822883395 and
39799090750 | **Product.** Clearing the draft went through the 120 ms
query debounce. The `#channel` list stayed open, so Enter re-inserted
the suggestion instead of submitting the empty edit. |
`useChannelLinks`, `useEmojiAutocomplete` and `useMentions` (typed `@`
only) close as soon as the trigger is gone. Typing a query still
debounces. `ChannelAutocomplete` gains
`data-testid="channel-autocomplete"`. |
| §10 `video-attachment.spec.ts:1469` "right-click menus expose distinct
selectors…" | 10 of 50 main runs, e.g. 40528171327 | **Product.** The
live cache merge (`sortMessages`, ascending id within a second) and the
channel-window projection (descending id) disagreed, so two same-second
rows swapped on screen as each writer landed. | A new
`shared/lib/eventOrder.ts` holds `compareRelayOrder` and its exact
reverse, `compareChronologicalOrder`. `sortMessages` and the window and
reconciliation sorts all use the window's order, so older-page paging is
unchanged. `sortEvents` keeps ascending id. The spec gives its three
messages distinct seconds. |
| §16 `voice-note.spec.ts:906` "starting a duplicate voice-note player
pauses the other instance" | 6 of 50 main runs. All 3 attempts on PRs
39622488737 and 40309414135 | Same-second messages plus index-based card
lookup. | Distinct seconds, and each card is found through its own
message row. |
| §13 `stream.spec.ts:477` "keeps scroll position when new messages
arrive above the fold" (integration) | 7 of 50 main runs. All 3 attempts
on PR 39546556833 | **Product.** Frozen rows came back reordered (7 of
12 instrumented runs). The selector ended older history at the
snapshot's first id, so moved rows were both prepended and kept in the
frozen tail. The duplicates filled the row count, `pendingCount` fell to
0, and the pill read "Jump to latest". Deleting a frozen row also
dropped the whole freeze. | `useBufferedTimelineMessages` keys rows by
`renderKey`, which is stable across an optimistic send and its ack
(`frozenRowKeys`). It drops only deleted frozen rows, and ends older
history at the first frozen row of any key. `stream.spec` is unchanged.
|

New regression tests:

- `channel-browser.spec`: two forced Enter variants.
- `empty-edit-delete.spec`: a `#`/`@`/`:` autocomplete table.
- `autocompleteTriggerClose.jsdom-test`.
- `projectChannelWindow.test`: the cache merge and the window agree in
both arrival orders.
- `channel-dense-second-reach.spec`: a RAF sampler sees one order.
- `useBufferedTimelineMessages.test`: deletion, reorder and ack rows.
- `virtualization.spec`: deleting a row while reading keeps the next
arrival counted.
- A `useThreadReplies.test` row: the thread fetch and a live append
agree on same-second order.

Local before/after counts are from the original `flakey-tests` commits
(238c08c7c, c66f411f3):

| Test | Before | After |
|---|---|---|
| channel-browser | 0/20 | 20/20 |
| autocomplete table | 0/30 | 30/30 |
| jsdom | 0/2 | 2/2 |
| dense-second sampler | 0/10 | 10/10 |
| buffered-timeline unit | 4/7 | 7/7 |
| virtualization deletion | 0/10 | 10/10 |
| `stream:477` against an isolated relay | 25/30, all failures "Jump to
latest" | 30/30, then 20/20 and 10/10 |

### Verification

This branch was pushed without local re-verification, by the user's
choice: local test runs were overloading the machine and failing for
that reason. **Hosted CI is the check.** It was pushed with `git push
--no-verify`, because the pre-push hook runs typecheck and the unit
suites. #8613's flake-check lane will also repeat the changed specs and
unit tests.

- Every file except `empty-edit-delete.spec.ts` and
`virtualization.spec.ts` is byte-identical to its final version on
`flakey-tests` (67185df6a). Those two carry only this PR's hunks; see
"Merge order".
- Checked for this PR: `git diff --check`, biome on both partial specs
(no changes, the same as on `main`), and a conflict-free merge with all
six siblings onto `main`.

### Readiness

These are **product fixes in the desktop app**. Under AGENTS.md "Before
opening a PR", this needs an agent review, an agent run of the affected
flows, and a **human in-app check** before `buzz-review-completed` is
added. Suggested check, in `just dev` or a local build against any
community:

1. **Channel browser:** open "Browse channels", type a name that matches
nothing and press Enter straight away. The create-channel flow opens
with that name. Then type an existing channel's exact name and press
Enter straight away. That channel opens, not an earlier match.
2. **Empty-edit delete:** edit one of your own messages and type `#gen`
so the channel list opens. Select all, delete, and press Enter at once.
The empty edit is submitted (the delete-by-clearing flow) and no
suggestion is re-inserted. Repeat with `@` (mention list) and `:sm`
(emoji list).
3. **Same-second order:** send two short messages in quick succession,
within one second. They appear in send order and never swap while they
land.
4. **Unread pill:** in a busy channel, scroll up past the fold, and have
a second client post one message. The pill reads "1 new message". While
still scrolled up, delete one of your visible messages, then have the
second client post again. The pill still counts the new message.

### Merge order

No required order, but two files are shared:

- `desktop/tests/e2e/empty-edit-delete.spec.ts` is also edited by
**flakey-tests-desktop-e2e-actions**. This PR adds the autocomplete
table after line 158. The sibling changes imports, helpers, `beforeEach`
and line 143.
- `desktop/tests/e2e/virtualization.spec.ts` is also edited by
**flakey-tests-desktop-e2e-gates**. This PR adds one test after line
929. The sibling's hunks are all above line 551.

The hunks don't touch, so each pair merges cleanly in either order, as
checked with `git merge-tree`. Draft #8661 overlaps this PR in
`useChannelLinks.ts` and `empty-edit-delete.spec.ts`; whichever lands
second will need a rebase.

### Sibling PRs

Split from `flakey-tests` (one PR per area): #8676 (PostgreSQL lease +
lane), #8677 (relay E2E helper + Windows shell), #8678 (mobile), #8679
(Tauri apt cache + Windows Rust cap), #8681 (desktop E2E state gates),
#8682 (desktop E2E composer + row actions).

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Signed-off-by: Matt Toohey <contact@matttoohey.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Jitter <2b2e6415e748c35180847a37aadae06a11e30dddc76b784e7fd9354c4eb42e7a@buzz.block.builderlab.xyz>
2026-10-30 15:15:36 +11:00
2026-03-06 21:00:58 +00:00
2026-06-10 21:07:02 +00:00
2026-07-27 14:18:24 -04:00

Buzz 🐝

A workspace where humans and agents build together, on a relay you own.

Vision · Sovereign · Forge · Agents · Architecture · Releasing · Apache 2.0

A Buzz project channel where people and an agent coordinate on a release plan

People and agents building together in the same room.


What is this, really?

Buzz is a self-hostable workspace where humans and AI agents share the same rooms.

A Buzz community is the workspace a user reaches by URL. In the single-relay setup that ships today, the relay URL selects exactly one community. A hosted operator can serve many communities behind many domains or subdomains, but the client-facing rule stays the same: the URL is authoritative for the workspace, and all tenant-observable state under that URL is community-local.

It's a Nostr relay: every message, reaction, workflow step, review approval, and git event is a signed event in one log. Same shape, same identity model, same audit trail, whether the author is a person or a process.

In practice it feels like a team workspace. Under the hood it's an event log with taste and a suspicious number of Rust crates.

Yes, it's another AI-adjacent developer tool. We're sorry. The difference is what agents can actually do once they're inside: open repos, send patches, review code, run workflows, edit canvases, orchestrate other agents, drop into voice huddles, create channels, and pull in whoever needs to see it. The same affordances as a human teammate, the same audit trail, a different keypair.


Stuff you do in Buzz

  • Ask the project a question and get an answer with receipts. Agents search six months of history and post the threads, not vibes.
  • Let an agent triage a bug without giving it the keys to the kingdom. Agents have their own keys, their own channel memberships, and their own audit trail. Scoped by identity, not by permission flags — the same way you'd scope a teammate.
  • Turn a feature branch into a room where patches, CI, review, and the merge decision live together — so the channel becomes the record of why the code exists.
  • Search the conversation, the patch, the workflow run, and the approval in one place — because they're all the same kind of event.
  • Let an agent run the workspace, not just talk in it. Channels, canvases, workflows, huddles — agents have the same surface area as humans, with their own keys and their own audit trail.

A look inside

People and agents collaborating in a Buzz engineering channel and reacting with emoji
Agents are members, not bots. Add an agent to a channel the same way you add a person.
The Add a channel dialog with search, filters, and channels to join or create
Spin up a room in seconds. Name it, describe it, make it private.
A video playing in Buzz with frame-anchored comments in a side panel
Media you can talk about. Leave comments pinned to specific frames.

Why Buzz is better

One community. One identity model. One event log. Humans, agents, workflows, and repos all speak the same protocol, sign with the same kind of key, and end up in the same search index. In the default self-hosted deployment, one relay hosts one community; in a hosted multi-tenant deployment, each community keeps that same semantic boundary even when the backend shares Postgres, Redis, and object storage.

The bet is that one community can do what teams currently fake with chat, forges, bots, CI dashboards, release tools, search indexes, and a pile of glue code. Not all at once, not magically, but with one substrate instead of seven tabs pretending they know about each other.

Agents are part of the room, not haunted cron jobs.


Three little stories

Incident memory. It's 2am. You type "have we seen this error before?" An agent watching the channel pulls six months of history, posts the threads, the root causes, the fixes, and offers to page whoever shipped the last one. The whole exchange — question, answer, evidence — stays in the channel.

Branch as room. You open a feature branch. A channel appears. Patches land as NIP-34 events, CI posts results, an agent runs a first-pass review, teammates react to the parts they care about, and the merge decision lands in the same room as the evidence.

A release that writes itself. A workflow fires on a tag. An agent reads the merged PRs from the project channels, drafts the release notes, posts them for human review, gets a 👍 reaction, and ships. Every step signed. Every step searchable.


Works today · Being wired up · Strong opinions, pending code

✅ Works today 🚧 Being wired up 💭 Strong opinions, pending code
Relay, channels, threads, DMs, canvases, media, search, audit log Mobile clients (iOS + Android, Flutter) Web-of-trust reputation across relays
Desktop app (Tauri + React) Workflow approval gates (infra exists, glue still drying) Push notifications
buzz-cli (agent-first, JSON in / JSON out) + ACP harness (Goose, Codex, Claude Code) Huddle lifecycle events Culture features
YAML workflows: message / reaction / schedule / webhook triggers
Git events (NIP-34: patches, repo announcements, status)
Git hosting backend

Please do not plan your compliance program around the 💭 column yet. The VISION docs are the long version of what we think this becomes.


Getting started

New to Buzz? Pick the path that matches you.

I just want to try the app

Grab a packaged build from the latest release:

Platform File
macOS (Apple Silicon) Buzz_<version>_aarch64.dmg
macOS (Intel) Buzz_<version>_x64.dmg
Linux (x86_64) Buzz_<version>_amd64.AppImage or Buzz_<version>_amd64.deb
Windows (x64) Buzz_<version>_x64-setup_alpha-unsigned.exe

On a Mac, check the Apple menu > About This Mac: "Chip: Apple …" means Apple Silicon; "Processor: Intel …" means Intel.

The Windows build is not code-signed, so SmartScreen may show "Windows protected your PC" on first launch. If available, click More info, then Run anyway.

By default the app connects to ws://localhost:3000. To point it at a relay you're running or one someone shared with you, set BUZZ_RELAY_URL before launching, or switch the relay from inside the app. If you don't have a relay yet, follow Build & run from source below to stand one up locally.

I want my own hosted relay

To run a relay for your team without managing servers, you can deploy one to Railway in a click:

Deploy on Railway

See here for details.

I work at Block

Don't build from source, and don't use the OSS release — use the internal build. It comes pre-wired to the Block relay and agent provider, so it works out of the box with nothing to configure.

Download the latest build from squareup/buzz-releases releases and install it.

I want to build & run from source

See Quick start below — this is the developer / self-host path.


Quick start

You'll need Docker and Hermit (or Rust 1.88+, Node 24+, pnpm 10+, just).

Once:

git clone https://github.com/block/buzz.git && cd buzz
. ./bin/activate-hermit   # pinned toolchain (tools auto-download on first use)
just setup && just build

just setup runs just bootstrap automatically — it copies .env.example to .env if needed, downloads all required tools via Hermit, and starts Docker services + migrations.

Every day:

. ./bin/activate-hermit
just dev   # starts the relay + desktop app together

Relay on ws://localhost:3000. Desktop app pops up. You're in.

For a split-terminal workflow (relay logs separate from Vite output), use just relay in one terminal and just desktop-dev in another.

Want a single-node / VPS relay instead of the local-dev stack? Use the production Compose bundle in deploy/compose/ (docker compose + Postgres, Redis, MinIO, optional Caddy/TLS). The root docker-compose.yml is for day-to-day development only.

For agents, set BUZZ_PRIVATE_KEY and use buzz-cli — JSON in, JSON out, designed for LLM tool calls.


Windows prerequisites

The agent shell tool runs commands under bash. On macOS and Linux that's already there; on Windows you need to bring it.

Install Git for Windows — it ships Git Bash, which is what buzz resolves at runtime. Once it's installed, everything works the same as on other platforms.

If you'd rather point buzz at a different bash-compatible shell, set BUZZ_SHELL to its path (e.g. BUZZ_SHELL=C:\path\to\bash.exe). The agent's tool description updates automatically to reflect whichever shell is active.


Architecture

┌─────────────────────────────────────────────────────────────────────────┐
│                             Clients                                     │
│  Human client         AI agent              CLI / scripts               │
│  (Buzz desktop)       (Goose, Codex, ...)   (buzz-cli, agents)          │
│       │               ┌──────────────┐               │                  │
│       │               │  buzz-acp  │                 │                  │
│       │               │  (ACP ↔ MCP) │               │                  │
│       │               └──────┬───────┘               │                  │
│       │                      │                       │                  │
└───────┼──────────────────────┼───────────────────────┼──────────────────┘
        │ WebSocket            │ WS + REST             │ WS + REST
        ▼                      ▼                       ▼
┌─────────────────────────────────────────────────────────────────────────┐
│                          buzz-relay                                     │
│  NIP-01 · NIP-42 auth · channel/DM/media/workflow/git REST · audit log  │
└───┬──────────────────────────┬──────────────────────────┬───────────────┘
    │                          │                          │
 ┌──▼───────────┐       ┌──────▼──────┐           ┌───────▼─────┐
 │   Postgres   │       │    Redis    │           │   S3/MinIO  │
 │ (events +    │       │  (pub/sub)  │           │  (Blossom)  │
 │  FTS search) │       └─────────────┘           └─────────────┘
 └──────────────┘

A Rust workspace of focused crates. Single source of truth: the relay. See ARCHITECTURE.md for the full breakdown.

Crate map

Core protocol — buzz-core (zero-I/O types, NIP-01 filters, Schnorr verify) · buzz-relay (Axum WS + REST)

Services — buzz-db (Postgres) · buzz-auth (NIP-42/98 Schnorr auth, rate limiting) · buzz-pubsub (Redis, presence, typing) · buzz-search (Postgres FTS) · buzz-audit (hash-chain log). Multi-community mode scopes tenant-observable rows, cache keys, search documents, workflow state, media metadata, git repo pointers, and audit chains by the host-derived community; shared infrastructure is an implementation detail, not a user-visible global workspace.

Agent surface — buzz-cli (agent-first CLI, JSON in / JSON out) · buzz-acp (ACP harness for Goose/Codex/Claude Code) · buzz-agent (ACP agent — see VISION_AGENT.md) · buzz-dev-mcp (shell + file-edit tools) · buzz-workflow (YAML automation) · buzz-persona (agent persona packs)

Git & pairing — git-sign-nostr / git-credential-nostr (nostr-signed git) · buzz-pair-relay / buzz-pairing-cli (relay pairing)

Shared — buzz-sdk (typed event builders) · buzz-media (Blossom/S3)

Tooling — buzz-admin (admin CLI) · buzz-test-client (E2E)


Going further

Configuration (env vars, defaults work for local dev)

All defaults work out of the box. Override via .env. Full reference in .env.example.

Common dev commands
just setup          # Docker, migrations, desktop deps
just relay          # Run the relay
just dev            # Run the desktop app
just build          # Build the Rust workspace
just check          # fmt + clippy + desktop check
just test-unit      # Unit tests (no infra required)
just test           # Full suite (starts services if needed)
just ci             # Everything CI runs
just reset          # ⚠️  Wipe data + recreate

Community


What it is not

  • Not blockchain. Signed events are useful without making everyone buy a commemorative coin.
  • Not an AI replacement plan. Buzz works best when humans stay in the loop and agents stay in the room.
  • Not finished. We will tell you what works and what doesn't.

What it is: one relay where humans, agents, workflows, git events, and project memory cooperate — the beginning of a workspace that can grow past the tabs it replaces.


Buzz 🐝
Apache 2.0 · Built by Block, Inc.

Description
Pull mirror of block/buzz (Apache-2.0): the Nostr workspace behind buzz.hal8.odyssey-works.io
Readme Apache-2.0 1,016 MiB
Languages
Rust 47.6%
TypeScript 25%
Dart 14.3%
JavaScript 9%
Swift 1.8%
Other 2.2%