MCP Server

Control Rymi with natural language. The @rymi/mcp package connects your AI assistant directly to your Rymi account — create agents, inspect calls, validate configs, and go live without touching the dashboard or writing a single HTTP request.

Works with Claude Desktop, Claude Code, Cursor, Windsurf, or any MCP-compatible client.

How it works

You (natural language)
    ↓
AI client (Claude, Cursor, etc.)
    ↓  MCP tool calls
mcp.rymi.live  ←  your API key
    ↓
Rymi API

The MCP server translates your plain-English instructions into typed Rymi API calls. The AI figures out which tools to chain — you just describe what you want.

Quickstart

No install needed. Point your client at https://mcp.rymi.live and pass your API key as a Bearer token.

Claude Code (~/.claude.json)

{
  "mcpServers": {
    "rymi": {
      "type": "http",
      "url": "https://mcp.rymi.live/",
      "headers": {
        "Authorization": "Bearer rymi_your_secret_key"
      }
    }
  }
}

Claude Desktop

// ~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "rymi": {
      "type": "http",
      "url": "https://mcp.rymi.live/",
      "headers": {
        "Authorization": "Bearer rymi_your_secret_key"
      }
    }
  }
}

Get your API key from the Rymi dashboard → Settings → API Keys.

Available tools

These are the tools your AI can invoke once connected:

Tool	What it does
list_agents	List all your voice agents
get_agent	Fetch a single agent's full config
create_agent	Create a new voice agent from scratch
update_agent	Modify an existing agent's config
clone_agent	Duplicate an agent (great for A/B variants)
delete_agent	Permanently remove an agent
apply_agent_changes	Dry-run a change-set — validates without saving
validate_agent_publish	Check if an agent passes all publish requirements
publish_agent	Make an agent live (opt-in, see below)
list_calls_for_agent	Browse call history for an agent
list_llm_options	Discover available LLMs and voices
enrich_company	Auto-generate a company description from a URL
generate_agent_draft	Generate a full agent config from a plain-English prompt

Example prompts

Once connected, just talk to your AI client:

Build an agent from scratch

"Create a receptionist agent for BrightSmile Dental. Warm and professional tone. Grab the company info from brightsmile.example.com."

The AI will chain: list_llm_options → enrich_company → generate_agent_draft → create_agent

---

Tweak an existing agent

"Switch my Support agent to the Charon voice and update it to GPT-4o."

The AI will chain: list_agents → apply_agent_changes → update_agent

---

Validate before going live

"Is my agent abc-123 ready to publish?"

The AI calls: validate_agent_publish → surfaces any missing fields or config issues

---

Voice agents creating voice agents

A deployed Rymi agent with this MCP wired in can respond to a caller in real time: "Set up a new booking agent for my dental clinic" → the agent calls create_agent directly. No human in the loop required.

Enabling publish (opt-in)

publish_agent is disabled by default — an AI can't flip an agent to production without you explicitly opting in. To enable it, contact support@rymi.live or toggle it in your dashboard API key settings.

WARNING

Once enabled, publish_agent makes an agent immediately callable by end users. Only use it in supervised workflows.