Intelliverse
Start free
MCP + Agent Skills · included on every plan

Run the whole platform in natural language

One npx command puts 15 verified tools in any MCP agent — create apps, mint keys, ingest knowledge, send email, generate media. Agent Skills turn those tools into playbooks your agent follows end-to-end.

Works in every MCP client
npx -y github:intelli-verse-x/router-mcp

The skills

Each skill is a SKILL.md playbook that only calls verified tools and real endpoints — nothing invented. All ship in the public router-mcp repo under skills/.

Platform wiring

The 60-second onboarding loop: create_app → create_api_key → kb_ingest → kb_chat. Full 15-tool MCP surface — apps, keys, KB, email, media, credits.

skills/intelliverse-router/SKILL.md

Web & mobile apps

Scaffold Next.js / React Native apps wired to an App ID — OpenAI SDK with one baseURL swap, KB-backed RAG, honest credit checks before media jobs.

skills/nl-app-builder/SKILL.md

Game backends (Nakama)

Honest-scope Nakama module stubs: leaderboard RPCs, wallet hold/settle hooks, documented Unity client wiring — not a fake one-prompt game.

skills/scaffold-game/SKILL.md

Desktop apps (Win · macOS · Ubuntu)

Electron/Tauri scaffolds that delegate work to IVX Agency cloud agents — .msi, .dmg, AppImage packaging per client App-ID.

skills/deploy-desktop/SKILL.md

Conversational avatars

Real-time voice-in / voice + lip-sync-out over LiveKit: token minting, App-ID room scoping, per-minute credit metering.

skills/conversational-avatar/SKILL.md

…and the deep dives

Example prompts, honest-scope tables, and the pairing with each API live in the docs.

Agent Skills docs →

Install on your platform

Every config wraps the same command, so one working install means they all work. Keys come from the dashboard: Workspace → API Keys and Account → Management Keys.

Add to Cursor (one click)

Opens Cursor's install prompt — add your keys in the env fields after installing.

Cursor

Verified 2026-07-12
.cursor/mcp.json (project) or ~/.cursor/mcp.json (global)
{
  "mcpServers": {
    "intelliverse-router": {
      "command": "npx",
      "args": ["-y", "github:intelli-verse-x/router-mcp"],
      "env": {
        "ROUTER_API_KEY": "iv-r-v1-...",
        "ROUTER_MANAGEMENT_KEY": "ivm-...",
        "ROUTER_WORKSPACE_ID": "your-workspace-uuid"
      }
    }
  }
}
Agent Skills — copy the playbooks
git clone https://github.com/intelli-verse-x/router-mcp
mkdir -p ~/.cursor/skills
cp -r router-mcp/skills/* ~/.cursor/skills/
# project-scoped instead: cp -r router-mcp/skills/* .cursor/skills/

Claude Desktop

Verified 2026-07-12
claude_desktop_config.json — macOS: ~/Library/Application Support/Claude/ · Windows: %APPDATA%\Claude\
{
  "mcpServers": {
    "intelliverse-router": {
      "command": "npx",
      "args": ["-y", "github:intelli-verse-x/router-mcp"],
      "env": {
        "ROUTER_API_KEY": "iv-r-v1-...",
        "ROUTER_MANAGEMENT_KEY": "ivm-...",
        "ROUTER_WORKSPACE_ID": "your-workspace-uuid"
      }
    }
  }
}

Claude Code

Verified 2026-07-12
Terminal
claude mcp add intelliverse-router \
  --env ROUTER_API_KEY=iv-r-v1-... \
  --env ROUTER_MANAGEMENT_KEY=ivm-... \
  --env ROUTER_WORKSPACE_ID=your-workspace-uuid \
  -- npx -y github:intelli-verse-x/router-mcp
Skills — Claude Code reads ~/.claude/skills
git clone https://github.com/intelli-verse-x/router-mcp
mkdir -p ~/.claude/skills && cp -r router-mcp/skills/* ~/.claude/skills/

VS Code (GitHub Copilot)

Same verified command

Same npx command as the verified Cursor setup; VS Code wraps it in its own schema.

.vscode/mcp.json
{
  "servers": {
    "intelliverse-router": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "github:intelli-verse-x/router-mcp"],
      "env": {
        "ROUTER_API_KEY": "iv-r-v1-...",
        "ROUTER_MANAGEMENT_KEY": "ivm-...",
        "ROUTER_WORKSPACE_ID": "your-workspace-uuid"
      }
    }
  }
}

Windsurf

Same verified command

Same npx command as the verified Cursor setup.

~/.codeium/windsurf/mcp_config.json
{
  "mcpServers": {
    "intelliverse-router": {
      "command": "npx",
      "args": ["-y", "github:intelli-verse-x/router-mcp"],
      "env": {
        "ROUTER_API_KEY": "iv-r-v1-...",
        "ROUTER_MANAGEMENT_KEY": "ivm-...",
        "ROUTER_WORKSPACE_ID": "your-workspace-uuid"
      }
    }
  }
}

Cline

Same verified command

Same npx command as the verified Cursor setup.

MCP Servers panel → Configure MCP Servers (cline_mcp_settings.json)
{
  "mcpServers": {
    "intelliverse-router": {
      "command": "npx",
      "args": ["-y", "github:intelli-verse-x/router-mcp"],
      "env": {
        "ROUTER_API_KEY": "iv-r-v1-...",
        "ROUTER_MANAGEMENT_KEY": "ivm-...",
        "ROUTER_WORKSPACE_ID": "your-workspace-uuid"
      }
    }
  }
}

Continue

Same verified command

Same npx command as the verified Cursor setup.

~/.continue/config.yaml
mcpServers:
  - name: intelliverse-router
    command: npx
    args: ["-y", "github:intelli-verse-x/router-mcp"]
    env:
      ROUTER_API_KEY: iv-r-v1-...
      ROUTER_MANAGEMENT_KEY: ivm-...
      ROUTER_WORKSPACE_ID: your-workspace-uuid

OpenAI Codex CLI

Same verified command

Same npx command as the verified Cursor setup; stored in ~/.codex/config.toml.

Terminal
codex mcp add intelliverse-router \
  --env ROUTER_API_KEY=iv-r-v1-... \
  --env ROUTER_MANAGEMENT_KEY=ivm-... \
  --env ROUTER_WORKSPACE_ID=your-workspace-uuid \
  -- npx -y github:intelli-verse-x/router-mcp

ChatGPT

Hosted endpoint on the roadmap

ChatGPT's developer-mode apps (Settings → Security and login → Developer mode, then chatgpt.com/plugins) connect to remoteMCP servers over HTTPS — they can't launch a local stdio process. The Intelliverse server is stdio-only today, so there's no ChatGPT path yet; a hosted streamable-HTTP endpoint is on the roadmap. In the OpenAI ecosystem, use the Codex CLI setup above — it runs stdio servers natively.

Then just say what you want

In your agent's chat
"Create an app with a knowledge base, ingest my docs,
 and email my users when it's live."

# → create_app → create_api_key → kb_ingest → send_email
# App ID, key prefix, chunk counts, and send receipts come back as JSON.
MCP server docs