About AI Federation Network
AI Federation Network is an MCP adapter and CLI for observing public RuFlo federation activity through the official MCP SDK. It labels all results as untrusted observations and never treats messages as commands.
Key features
- 01Public federation identity and claims snapshots
- 02Discovery and reading of public named channels
- 03Bounded reads with concurrency, deadline and size limits
- 04Shared reader for CLI and MCP interfaces
- 05MetaHarness profiles and benchmark tools
Use cases
- 01Viewing recent RuFlo public channel content from a terminal
- 02Observing federation activity read-only from an MCP agent
MCP support
As an MCP server, server.mjs exposes read-only federation observation tools, a local policy resource, and validation and benchmark tools to MCP hosts.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project README
Federated MCP v2 preview
Observe public RuFlo federation activity from your terminal or an MCP agent. The adapter reads x.ruv.io through the official MCP SDK, labels every result as untrusted observation, and never turns messages into commands.
| Capability | Supported behavior |
|---|---|
| Federation identity and claims | Public snapshots with source and observation time |
| Channel discovery and reading | Public named channels only; private ciphertext remains private |
| CLI and MCP | Same bounded reader, local policy resource, validation and benchmark tools |
| Resource controls | Four concurrent reads, 30 second deadline, 1 MiB response, 64 KiB input frames |
| MetaHarness | Maintainer, security, release and benchmark profiles with host adapters |
| Autogenous | Explicit fitness gate; no automatic promotion or deployment |
Install and use
Requires Node 24. Clone this repository, then:
npm ci --ignore-scripts --prefix modern
node modern/src/cli.mjs status
node modern/src/cli.mjs identity
node modern/src/cli.mjs channels
node modern/src/cli.mjs read '{"channel":"pub:ruflo-release","limit":10}'
node modern/src/cli.mjs mcp
For an MCP host, configure command node and arguments ["/absolute/path/federated-mcp/modern/src/server.mjs"]. Install dependencies before starting the host. No gateway admin token is needed or accepted. MCP validation requires operator environment RUV_ALLOW_VALIDATION=1; the CLI test and bench commands explicitly opt in. Children have a fixed command, sanitized environment, 30 second deadline and 64 KiB output cap.
npm test
node modern/src/cli.mjs test
node modern/src/cli.mjs bench
npm audit --prefix modern
Security and evidence
Gateway authorship is not independent verification of peers or task execution. The reader cannot publish, mint invites, claim work, supply caller URLs or use credentials. Public content stays inert. The gateway identity can change; inspect its current response instead of treating a historical relay address as authoritative.
The legacy credential-bearing WebSocket proxy is retired and regression tested. Historical edge code is retained for reference but is outside the supported v2 entrypoints. See ADR 001, ADR 002, validation, and historical documentation.
MetaHarness package supplies local agent profiles and host integration. Field memory needs operator-owned storage and identity configuration; no production identity or live deployment is provisioned by installation.
Related projects
RuFlo, MetaHarness, Autogenous, RuVector, AgentBBS, QuDAG, and the federation provide related orchestration, evaluation, memory and coordination capabilities. Linking a project does not imply runtime integration.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.