This isn’t added yet but would be pretty easy to do! Currently the tools in this MCP only read and query from relays but you could add some simple tools to create identity, create notes, and broadcast notes. Each chat could create its own ephemeral identity (or you could pass nsec to the LLM in the prompt to have a consistent identity?) View quoted note β†’
Of course we do now I can more effectively spy on you and everyone else through nostr relays using natural language prompts View quoted note β†’
YES View quoted note β†’
You can pass in relay param... Also give me a break I literally built it this morning lol image View quoted note β†’
test
Cool! View quoted note β†’
Wow! View quoted note β†’
Claude can talk to nostr relays
GitHub
GitHub - AustinKelsay/nostr-mcp-server: A Model Context Protocol (MCP) server that provides Nostr capabilities to AI agents
A Model Context Protocol (MCP) server that provides Nostr capabilities to AI agents - AustinKelsay/nostr-mcp-server
No Claude (or another model) can query nostr relays itself and understand what it’s getting back (certain kinds of notes at least) View quoted note β†’
Just one shotted https://hasmartyzapped.today/ Improved zap parsing and analytics! View quoted note β†’