Go to file
Tom Brow 934a16f2ff feat(mobile): message the Bestie from Siri and Shortcuts (#9517)
Adds a **Message my Bestie** action for Siri and Shortcuts. Siri asks
what to say, confirms with the Bestie's name ("Send “…” to Astra?"), and
posts the message in your DM with the Bestie, without opening Buzz. The
Bestie's reply arrives as a normal notification, in the same DM the Home
star opens.

This is the iOS 16+ App Shortcut from [ADR
0005](https://github.com/block/buzz/blob/mobile-siri-bestie-dm/docs/adr/0005-quick-send-uses-the-system-send-message-action.md).
Follow-ups, not in this PR: the iOS 27 `messages.sendMessage` schema
intent, naming other recipients, a separate "Start a chat with my
Bestie" intent for a new Solo chat, and opening Buzz with a draft when a
send fails.

### How it works

- The App Intent runs without opening the app and hands the text to Dart
over a method channel. It waits up to 20 s for Home to say it is ready,
and runs one send at a time.
- A background launch already starts the Flutter engine and builds Home
([ADR
0004](https://github.com/block/buzz/blob/mobile-siri-bestie-dm/docs/adr/0004-background-work-runs-in-the-app-engine.md)
records the measurement). So a small `QuickSendHost`, which the app
hands Home as an invisible background widget, takes the send. Nothing is
reimplemented natively.
- Before confirming, Swift asks Dart for the saved Bestie's pubkey and
name; the send then goes to exactly that agent, so a choice this phone
hasn't synced yet is caught at the confirmation.
- The send reuses the Home star's lookup for the Bestie DM (opening it
with the relay's find-or-create DM command when it isn't loaded) and
posts with the same `SendMessage` the composer uses.
- If Buzz has been in the background past its 5 s grace period, the
relay session has closed its socket. The send reopens it, then lets the
normal background pause start over.
- A message the relay rate-limits is retried (up to 3 tries); the relay
turned it away, so a retry can't duplicate it. Any other failure isn't
retried, since the message might have landed.
- The confirmation is pinned to the community and account it was shown
in: if Buzz switches either before the message goes out, the send is
refused.
- If the send fails, or Buzz can't tell whether it landed (no answer
from the relay), the message is saved as the Bestie DM's composer draft,
after any draft already there, and Siri says so ("Buzz isn't sure that
sent. Check your Bestie DM before trying again." when the outcome is
unknown). A successful send leaves the draft alone. Limits: if the DM
can't be found or opened, there is nowhere to save it and Siri says the
message wasn't saved, and the app being killed mid-send isn't covered.
- Refusals (no Bestie, offline, empty or too long, switched community)
come back as a short message Siri reads.

### Regressions

- Nothing existing changes behavior: the host only answers this method
channel, and the action does nothing until someone runs it. Its bugs
stay inside the quick send path.
- The Home star's DM lookup moved into `cachedBestieConversation`
unchanged, so the star behaves as before.
- A composer that is open on screen now shows a draft another writer
saved for it (here, a failed quick send) instead of overwriting it with
the next keystroke. It listens without rebuilding, and its own saves are
ignored.
- A background wake pays the cost of starting the app's Dart code, which
iOS already does for this launch.

### Before / after

| Before: no Buzz actions in Shortcuts | After |
| --- | --- |
| <img width="300" alt="Shortcuts Library before"
src="https://github.com/user-attachments/assets/b0870e3d-2823-4c7d-be0e-a242a3fdafd7">
| <img width="300" alt="Shortcuts Library after, listing Buzz Dev under
Apps"
src="https://github.com/user-attachments/assets/8c092006-76b9-4313-a401-68189f5674cf">
|

| Asks what to say | Confirms (names the Bestie) | Sent |
| --- | --- | --- |
| <img width="250" alt="Prompt"
src="https://github.com/user-attachments/assets/a2eaf44a-5287-48b6-9c5a-2aaefc90c0a8">
| Shows "Send “…” to *Bestie's name*?" (capture omitted: it shows an
internal agent name) | <img width="250" alt="Sent"
src="https://github.com/user-attachments/assets/07fe66e3-4480-4d06-be52-623416e567b8">
|

The DM result and a demo video show internal names, so they're shared
internally rather than here.

### Testing

- Added test, `quick_send_test.dart` journey: Home reports ready; Siri's
Bestie lookup is refused with no Bestie, then returns the chosen
Bestie's pubkey and name; a send confirmed for another community is
refused; an empty send is refused without publishing; a send while Buzz
is backgrounded and its socket closed reconnects, opens the DM with the
Bestie, survives a rate-limited first try, posts the message in that DM
tagged to the Bestie, and hands the session back to its background
pause; a second send goes to the same DM and leaves the person's draft
there as it was; a send whose answer is lost is reported as unsure and
kept in that draft after the person's own text, with its mentions; a
socket that drops just before the message goes out keeps it too; leaving
Home reports unready. Removing the rate-limit retry, the community check
or the draft save fails it.
- Extended the composer's share-delivery journey: a draft saved by
another writer while the composer is open shows in the composer and
survives the next keystroke, and survives an ordinary send that was
pending when it arrived and then fails. Removing the composer's listener
or its revision bump fails it.
- iPhone 17 Pro simulator (iOS 27), Builderlab community with a Bestie,
automated with AXe:
- Cold: Buzz terminated, ran the action from Shortcuts, the confirmation
named the Bestie, "Sent.", and the message was in the Bestie DM with its
reply. No Solo chat was created.
- Warm: Buzz in the background for 10 s, ran it again, "Sent.", same DM.
- At-risk flows: sent a typed message from the DM composer, went back to
Home, and the Home star reopened the same DM.
- `just mobile-check mobile-test` passes.






buzz-review-completed

---------

Signed-off-by: Tom Brow <tomb@squareup.com>
Co-authored-by: Tom Brow <tomb@squareup.com>
2026-11-15 09:06:25 -08:00
…
…
…
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.

Where to find things

Buzz's open-source code lives in two repositories. The desktop app has its own; the relay and the mobile client are here.

Component Home
Desktop app block/buzz-app
Mobile client (iOS + Android) This repo, in mobile/
Relay This repo, in crates/buzz-relay/

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 (in block/buzz-app) 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

The desktop app lives in block/buzz-app. Head there to install, build, or run it.

Need a relay for it to connect to? Deploy a hosted one, or follow Quick start 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 want to run the relay from source

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

I want to work on the mobile app

The Flutter client is in mobile/. See mobile/README.md for setup and run instructions.


Quick start

This stands up the relay locally. For the desktop app, see block/buzz-app; for the mobile app, see mobile/README.md.

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 relay   # starts the relay (and its Docker services if needed)

Relay on ws://localhost:3000. You're in.

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               │
│  (desktop, mobile)    (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. The human clients are the desktop app in block/buzz-app and the mobile app in mobile/. 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, JS deps
just relay          # Run the relay
just mobile-dev     # Run the mobile app in an iOS Simulator
just build          # Build the Rust workspace
just check          # Lint, formatting, and repo policy checks
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 1.1 GiB
Languages
Rust 47.9%
TypeScript 23%
Dart 16.8%
JavaScript 8.3%
Swift 1.9%
Other 2%