About @anycast/mcp
@anycast/mcp is an MCP server for managing Anycast Platform agents, connectivity, groups, tokens, and webhooks. It gives MCP-compatible clients tools and resources for operating the network.
Key features
- 01Manage agents and registration tokens
- 02Organize agents into groups
- 03Inspect P2P connections and statistics
- 04Manage webhook destinations and events
- 05Read agent, connection, and stats resources
Use cases
- 01Check online agents and their connection history
- 02Create or revoke agent registration tokens
- 03Review P2P connection counts and bandwidth
- 04Set up webhook destinations for event delivery
MCP support
Exposes 19 tools and resources to MCP clients for managing Anycast Platform agents, connections, groups, tokens, and webhooks.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
18Saved tool list · updated 3w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| status | enum | No | Filter by agent statusONLINE | OFFLINE | CONNECTING |
| search | string | No | Search by agent name |
| limit | number | No | Max results (default 20) |
Details
From the project README@anycast/mcp
Connect AI assistants to the Anycast network. Manage agents, P2P connections, and webhook delivery directly from Claude, Cursor, or any MCP-compatible client.
Sign up free · Documentation · npm
MCP (Model Context Protocol) server for the Anycast Platform. Enables AI assistants like Claude, Cursor, and Copilot to manage agents, webhooks, and P2P connectivity.
Installation
npm install -g @anycast/mcp
Configuration
Set your Anycast API key:
export ANYCAST_API_KEY=your-api-key-here
export ANYCAST_API_URL=https://agents.anycast.com # optional
Usage with Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"anycast": {
"command": "anycast-mcp",
"env": {
"ANYCAST_API_KEY": "your-api-key-here",
"ANYCAST_API_URL": "https://agents.anycast.com"
}
}
}
}
Usage with Claude Code
Add to your project's .claude/settings.json:
{
"mcpServers": {
"anycast": {
"command": "npx",
"args": ["@anycast/mcp"],
"env": {
"ANYCAST_API_KEY": "your-api-key-here"
}
}
}
}
Usage with Cursor
Add to Cursor settings > MCP Servers:
{
"anycast": {
"command": "anycast-mcp",
"env": {
"ANYCAST_API_KEY": "your-api-key-here"
}
}
}
Available Tools
Agents
| Tool | Description |
|---|---|
list_agents | List all agents with status, IP, version |
get_agent | Get agent details by ID |
update_agent | Update agent name or labels |
delete_agent | Delete agent (disconnects connections) |
get_agent_connections | Get connection history for an agent |
Tokens
| Tool | Description |
|---|---|
list_tokens | List registration tokens (prefixes only) |
create_token | Create new registration token |
revoke_token | Revoke a token |
Groups
| Tool | Description |
|---|---|
list_groups | List agent groups |
create_group | Create a new group |
add_agent_to_group | Add agent to group |
remove_agent_from_group | Remove agent from group |
Connectivity
| Tool | Description |
|---|---|
get_discovery | Get rendezvous server URL |
list_connections | List P2P connections |
get_connection_stats | Bandwidth and connection counts |
Webhooks (Ramps)
| Tool | Description |
|---|---|
list_destinations | List webhook destinations |
create_destination | Create webhook destination |
list_events | List webhook events |
Resources
| URI | Description |
|---|---|
anycast://agents | All registered agents |
anycast://connections | Recent connections (24h) |
anycast://stats | Platform statistics |
Examples
Ask Claude:
"List all my online agents"
"Create a registration token called 'production' with a limit of 50 uses"
"Show me the connection stats for the last week"
"Create a webhook destination at https://api.example.com/webhooks"
"Which agents are in the 'production' group?"
License
MIT
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.