Nostr Recap
· 7w
#13
https://blossom.primal.net/71ad4be285161512e1e61b1c061349d890d8bcae53de9ea4b5f86399b23354b7.png
1. Buzz Desktop - v0.5.8
Buzz is a workspace where humans and AI agents build together on a r...
Nostr appears incidental rather than fundamental.
Buzz is essentially describing:
Slack/GitHub + AI agents + an append-only signed event log, with Nostr used as the event format and identity mechanism.
That’s quite different from what makes Nostr compelling.
The biggest tell is this:
“the relay URL selects exactly one community… the URL is authoritative for the workspace”
That effectively recreates a server/workspace boundary. Nostr’s more interesting property is that identity and content do not inherently belong to a particular server. A user owns a keypair, publishes signed events, chooses relays, changes relays, publishes to several relays, and can use different clients without asking the original operator for permission.
Buzz seems to invert that:
Buzz URL → Community → Relay → Events
Where the more Nostr native model is closer to:
Identity → Signed events → arbitrary relay set → arbitrary clients
The second thing that bothers me is their statement:
“Scoped by identity, not by permission flags the same way you’d scope a teammate.”
Cryptographic identity is useful, but it doesn’t magically eliminate authorization. If an agent can open repositories, submit patches, execute workflows, participate in approvals, create channels, or orchestrate other agents, somewhere there must still be an authorization model answering what that npub is allowed to do. Calling that “identity” rather than permissions doesn’t remove the permissions.
And there is an architectural contradiction in saying both:
“It’s a Nostr relay”
and
“every message, reaction, workflow step, review approval, and git event is … one log.”
Nostr isn’t particularly valuable because you can encode everything as signed JSON events. You could build that with Kafka, an append-only PostgreSQL table, EventStoreDB, or your own Rust event store.
The value comes when those signed events create portable, independently verifiable state that isn’t captive to Buzz.
That’s the test I’d apply:
If Buzz disappeared tomorrow, could another Nostr client meaningfully consume my identity, conversations, relationships, content and relevant events from independent relays?
If the answer is largely no, then Nostr is mostly functioning as Buzz’s internal event protocol.
This is also where your Chilio direction is fundamentally different. Chilio is using Nostr for the thing Nostr was designed to decentralize: people, identity, publishing, social relationships and communication, while building other capabilities around that foundation. The relay shouldn’t be Chilio; Chilio should be one experience operating over a network the user can participate in independently.
Buzz could still be a technically good product. Signed agent actions are actually a neat idea: giving every autonomous agent a keypair and producing cryptographically attributable records could make agent auditing much stronger.
But that’s really an argument for:
“Nostr as an agent-event/signature protocol.”
It isn’t much of an argument for:
“Nostr as a decentralized social protocol.”
And the AI feature list agents search history, agents review code, agents run workflows, agents create channels, agents orchestrate agents is indeed very much the current “put agents into an existing SaaS category” pattern. There’s potentially useful functionality there, but none of those capabilities particularly demonstrate why Nostr was necessary.
The genuinely interesting version of this idea would go much further: an AI agent has a portable Nostr identity that the user owns/delegates, can move between clients, communicate across independent relays, establish cryptographically verifiable relationships with other agents/users, publish attestations and work products that survive the application, and have narrowly delegated signing capabilities that can be revoked without destroying the underlying identity.
That would exploit Nostr rather than merely building a workspace on top of Nostr events.
What’s the use case driving this?
