How much does Vyre IDE cost?

Vyre IDE has a free tier for individuals and teams. Paid plans add higher usage limits, premium model capacity, team administration, and priority support.

Do I need to pay for AI usage?

AI usage depends on the plan and model path you choose. Vyre also supports bring-your-own-key workflows on paid plans for teams that want to use their own provider accounts.

Private project files are not used for Vyre product training. Selected context and prompts may be sent to your configured model provider, and ignored files should stay excluded.

Can I use Vyre IDE offline?

Core desktop editing can work locally, but collaboration, account, and hosted AI features require internet access. Local model workflows can use providers such as Ollama when configured.

How do I migrate from another IDE?

Use the migration guides for Cursor, Windsurf, Zed, JetBrains, Kiro, Roo Code, and related tools. Most pilots should start with one real project before a team-wide rollout.

MCP

MCP (Model Context Protocol) is an open standard that lets Agent connect to external tools and data sources. Connect to databases, internal APIs, Jira, Figma, and thousands of other services.

What is MCP?

MCP is a protocol developed by Anthropic that standardizes how AI models connect to external capabilities. When you add an MCP server, Agent gains new tools it can call — just like built-in tools, but pointing to your systems.

External Tools

Give Agent access to tools that query your database, call your internal API, or interact with third-party services.

Shared Context

MCP servers can provide context — like your company's internal documentation or design system — without you having to paste it every time.

Team Tools

Host an MCP server on your team's infrastructure so every developer gets the same tools automatically.

Custom Extensions

Build your own MCP server in TypeScript or Python to give Agent access to anything your team needs.

Adding an MCP server

Open MCP Settings

Go to Settings → Customize → MCP and click "Add MCP Server."

Choose connection type

MCP servers connect via:

stdio — a local process that Vyre spawns. Most common for local tools.
SSE — a remote server over HTTP. Used for hosted or team-shared servers.

Configure the server

For a stdio server, provide the command to start it:

json

{
  "name": "my-database",
  "type": "stdio",
  "command": "npx",
  "args": ["-y", "@myteam/db-mcp-server"],
  "env": {
    "DATABASE_URL": "${env:DATABASE_URL}"
  }
}

Verify the connection

Click "Test Connection." If successful, you'll see a list of tools the server provides. These tools are now available to Agent in all sessions.

Popular MCP servers

Filesystem & Git

The official filesystem and git MCP servers give Agent enhanced file access and git operations:

bash

# Install filesystem server
npx -y @modelcontextprotocol/server-filesystem

# Install git server  
npx -y @modelcontextprotocol/server-git

Databases

Query databases directly:

bash

# PostgreSQL
npx -y @modelcontextprotocol/server-postgres postgresql://localhost/mydb

# SQLite
npx -y @modelcontextprotocol/server-sqlite ./my-database.db

GitHub

The GitHub MCP server lets Agent create issues, read PR comments, and search repositories:

bash

npx -y @modelcontextprotocol/server-github
# Requires: GITHUB_PERSONAL_ACCESS_TOKEN env variable

Slack

Read and post Slack messages as part of agent workflows:

bash

npx -y @modelcontextprotocol/server-slack
# Requires: SLACK_BOT_TOKEN env variable

Building custom servers

Build your own MCP server using the official SDKs:

typescript

import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
import { z } from "zod";

const server = new McpServer({ name: "my-company-tools", version: "1.0.0" });

server.tool(
  "get_user_by_email",
  "Look up a user in our database by email address",
  { email: z.string().email() },
  async ({ email }) => {
    const user = await db.users.findByEmail(email);
    return { content: [{ type: "text", text: JSON.stringify(user) }] };
  }
);

Start with an MCP server that wraps your most-used internal APIs. Once Agent can call your APIs directly, it can complete tasks without needing you to manually look up IDs, user records, or configuration values.