MCP Integration

Ze includes an MCP (Model Context Protocol) server that makes the BGP daemon AI-ready. Any AI assistant (Claude, GPT, or custom agents) can connect via MCP and fully control Ze -- the same operations available through the CLI are accessible programmatically through typed tools.

AI-Ready BGP Operations

The MCP server exposes typed tools with structured parameters, so AI assistants can manage BGP without parsing CLI output:

Tool	Description
`ze_announce`	Announce routes with typed parameters (origin, next-hop, communities, prefixes)
`ze_withdraw`	Withdraw routes
`ze_peers`	Monitor peer state, ASN, uptime
`ze_peer_control`	Teardown, pause, resume, flush peers
`ze_execute`	Run any CLI command -- full daemon control
`ze_commands`	List all available daemon commands
`ze_reference`	Full machine-readable reference for this daemon (commands, RPC endpoints, dispatch keys, plugins, families, services); same JSON as `ze help ai --json`. Call first to discover capabilities.

Additional tools are auto-generated from the command registry at runtime. Every registered YANG command and plugin command appears as a typed MCP tool with an action enum and optional arguments and peer parameters. New commands are exposed automatically without code changes.

The ze_execute tool is the key to full control: anything you can do in ze cli (interactive or ze cli -c for one-shot commands), an AI can do via MCP. This includes:

Route management: bgp peer * update text origin set igp nhop set 1.1.1.1 nlri ipv4/unicast add 10.0.0.0/24
RIB queries: show bgp rib received, show bgp rib sent, clear bgp rib in
Peer lifecycle: bgp peer * show, bgp peer 10.0.0.1 teardown 6, delete bgp peer-peer with ...
Configuration: request commit start window1, route changes, request commit end window1
Cache operations: cache list, cache forward
Event subscription: request subscribe bgp/update
Schema discovery: command-list, command-help <name>

Starting the MCP Server

ze start --mcp 8080 config.conf

Or via config:

environment {
    mcp {
        enabled true
        server main {
            ip 127.0.0.1
            port 8080
        }
    }
}

Environment variable overrides: ze.mcp.listen=ip:port, ze.mcp.enabled=true, ze.mcp.token=<secret>. Defaults to 127.0.0.1:8080 (security: local-only unless explicitly overridden via ze.mcp.listen). Bearer token auth available via --mcp-token flag, ze.mcp.token env var, or the config token leaf.

AI Command Reference

ze help ai
ze help ai api          # daemon API endpoints (ze-show:*, ze-set:*, ...)

Generates a machine-readable command reference from code, suitable for feeding to an AI as context. Lists all available commands with their parameters, descriptions, and examples. The legacy ze help --ai flag form is still accepted.

Example: AI-Driven Route Announcement

An AI assistant connected via MCP can:

Check peer state: ze_peers returns structured JSON with all peer status
Announce a route: ze_announce with origin=igp, next-hop=10.0.0.1, prefixes=[10.0.0.0/24]
Verify propagation: ze_execute with command show bgp rib sent peer peer1 family ipv4/unicast
Withdraw if needed: ze_withdraw with the same prefixes

All without parsing text output -- each tool returns structured data.

Elicitation (2025-06-18)

Tool handlers may ask the client for missing input mid-dispatch via session.Elicit. The server upgrades the POST reply to text/event-stream, sends an elicitation/create request over the same HTTP body, and resumes the handler when the client POSTs a response correlated by id. The client must advertise capabilities.elicitation at initialize for the server to prompt; otherwise handlers fail fast with a deterministic error. ze_execute illustrates the pattern: calling it with command="" on an elicit-capable client produces an elicitation/create asking which ze command to run.

See MCP Elicitation for the full flow.

Testing

ze-test mcp provides a functional test client with wait-established synchronization for CI pipelines and -- with --elicit plus elicit-accept/elicit-decline/elicit-cancel stdin directives -- covers the server-initiated elicitation flow.

See MCP Guide for details and MCP Remote Access for tunneling.