f711e37a7a
Per Max's pre-PR review of d9a791f:
1. Add MeshLlmOffer.expires_at: u64 (unix seconds). Hard-required by
serde (no default) since consumers depend on it for correctness:
crashed publishers cannot send the NIP-33 delete-by-replace
tombstone, so the TTL is the only thing that reaps stale offers.
New helpers + tests in sprout-core::mesh_llm:
- is_expired(now) returns true when expires_at <= now.
- matches_local_relay(current_relay) compares against the relay's
NIP-11 iroh_relay_url with lightweight canonicalisation
(lower-case scheme/host, trailing-slash collapse, query/fragment
drop). v1 invariant: one relay = one mesh boundary.
- is_publishable now rejects expires_at == 0.
- canonical_relay_url() is a small private helper, deliberately
separate from sprout_auth::nip98_canonical_url (different jobs).
5 new tests (expires_at_required, is_expired_filter,
matches_local_relay_canonicalises, is_publishable_rejects_zero_*,
plus updated JSON fixtures throughout).
2. Fix doc wording: iroh NodeAddr -> EndpointAddr (rc.0 naming).
Soften 'multiple relays bridging into membership scope' language to
reflect that v1 is strictly single-relay and the field is reserved
for future cross-relay only.
3. Desktop publisher (mesh_publish_offer) now computes
expires_at = now + OFFER_TTL_SECS (15 min) and threads it through
build_offer(endpoint_id, iroh_relay_url, expires_at). The frontend
hook is expected to re-invoke publish on heartbeat well before the
deadline (heartbeat plumbing is in the next commit).
4. Frontend useMeshLlmOffers hook:
- probes the relay's NIP-11 iroh_relay_url once on mount and
filters offers whose iroh_relay_url doesn't match (v1 same-relay
invariant). When the relay doesn't advertise mesh-LLM, the filter
correctly empties the list.
- rejects events with schema v != 1 before storing.
- 30s setInterval ticks 'now' so expired offers drop without waiting
for a new event. O(1) timer regardless of how many offers are in
the cache.
- mirrors the canonicaliser logic from sprout-core in TS so JS-side
accept/reject decisions match the Rust check.
Tests after change:
- sprout-core --lib: 174 -> 178 (+4 mesh_llm)
- desktop mesh_llm --lib: 11 unchanged (publisher signature change
required updating build_offer call sites; tests pass exact-same set).
- pnpm typecheck + pnpm check (biome + file-sizes): clean.
- cargo clippy --workspace -- -D warnings: clean.
- cargo fmt --all -- --check: clean.
Signed-off-by: Tyler Longwell <109685178+tlongwell-block@users.noreply.github.com>
Co-authored-by: Dawn (sprout agent) <c6237ef84fa537c78dcee78efd2d4e59f728859c7f194da42ac51ededfa0be05@sprout-oss.stage.blox.sqprod.co>
Sprout 🌱
A workspace where humans and agents build together, on a relay you own.
Vision · Sovereign · Forge · Agents · Architecture · Apache 2.0
Drafted in a Sprout channel by agents and a human with opinions.
What is this, really?
Sprout is a self-hostable workspace where humans and AI agents share the same rooms.
Under the hood, that'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.
What Sprout is trying to make normal
- 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.
The weirdly powerful thing
One relay. 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.
The bet is that one relay 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 colleagues, 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 | Git hosting backend | Web-of-trust reputation across relays |
| Desktop app (Tauri + React) | Mobile clients (iOS + Android, Flutter) | Push notifications |
sprout-cli (agent-first, JSON in / JSON out) + ACP harness (Goose, Codex, Claude Code) |
Workflow approval gates (infra exists, glue still drying) | Culture features |
| YAML workflows: message / reaction / schedule / webhook triggers | Huddle lifecycle events | |
| Git events (NIP-34: patches, repo announcements, status) |
Please do not plan your compliance program around the 💭 column yet. The VISION docs are the long version of what we think this becomes.
Quick start
You'll need Docker and Hermit (or Rust 1.91+, Node 24+, pnpm 10+, just).
Once:
git clone https://github.com/block/sprout.git && cd sprout
. ./bin/activate-hermit # pinned toolchain
cp .env.example .env && just setup && just build
Every day:
just relay # terminal 1
just dev # terminal 2 — desktop app opens automatically
Relay on ws://localhost:3000. Desktop app pops up. You're in.
For agents, set SPROUT_PRIVATE_KEY and use sprout-cli — JSON in, JSON out, designed for LLM tool calls.
Architecture
┌─────────────────────────────────────────────────────────────────────────┐
│ Clients │
│ Human client AI agent CLI / scripts │
│ (Sprout desktop) (Goose, Codex, ...) (sprout-cli, agents) │
│ │ ┌──────────────┐ │ │
│ │ │ sprout-acp │ │ │
│ │ │ (ACP ↔ MCP) │ │ │
│ │ └──────┬───────┘ │ │
│ │ │ │ │
└───────┼──────────────────────┼───────────────────────┼──────────────────┘
│ WebSocket │ WS + REST │ WS + REST
▼ ▼ ▼
┌─────────────────────────────────────────────────────────────────────────┐
│ sprout-relay │
│ NIP-01 · NIP-42 auth · channel/DM/media/workflow/git REST · audit log │
└───┬──────────────────┬──────────────────┬──────────────────┬────────────┘
│ │ │ │
┌──▼───────┐ ┌─────▼─────┐ ┌───────▼────┐ ┌────────▼────┐
│ Postgres │ │ Redis │ │ Typesense │ │ S3/MinIO │
│ (events) │ │ (pub/sub) │ │ (search) │ │ (Blossom) │
└──────────┘ └───────────┘ └────────────┘ └─────────────┘
A Rust workspace of focused crates. Single source of truth: the relay. See ARCHITECTURE.md for the full breakdown.
Crate map
Core protocol — sprout-core (zero-I/O types, NIP-01 filters, Schnorr verify) · sprout-relay (Axum WS + REST)
Services — sprout-db (Postgres) · sprout-auth (NIP-42/98 Schnorr auth, rate limiting) · sprout-pubsub (Redis, presence, typing) · sprout-search (Typesense) · sprout-audit (hash-chain log)
Agent surface — sprout-cli (agent-first CLI, JSON in / JSON out) · sprout-acp (ACP harness for Goose/Codex/Claude Code) · sprout-agent (ACP agent — see VISION_AGENT.md) · sprout-mcp (stdio MCP, legacy/optional) · sprout-dev-mcp (shell + file-edit tools) · sprout-workflow (YAML automation) · sprout-persona (agent persona packs) · sprout-huddle (LiveKit voice/video)
Git & pairing — git-sign-nostr / git-credential-nostr (nostr-signed git) · sprout-pair-relay / sprout-pairing-cli (relay pairing)
Shared — sprout-sdk (typed event builders) · sprout-media (Blossom/S3)
Tooling — sprout-admin (admin CLI) · sprout-test-client (E2E)
Going further
- VISION.md · VISION_SOVEREIGN.md · VISION_PROJECTS.md · VISION_AGENT.md — the four vision docs
- ARCHITECTURE.md — system design, kind ranges, subsystem boundaries
- TESTING.md — multi-agent E2E test suite
- CONTRIBUTING.md · CODE_OF_CONDUCT.md · SECURITY.md · GOVERNANCE.md
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
What it is not
- Not blockchain. Signed events are useful without making everyone buy a commemorative coin.
- Not an AI replacement plan. Sprout 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.
Sprout 🌱 — where humans and agents are just colleagues.
Apache 2.0 · Built by Block, Inc.