mcp-lindoai
Official
# @lindoai/mcp-server
MCP (Model Context Protocol) server for [Lindo AI](https://lindo.ai) — create websites, pages, and blog posts with AI directly from Claude, Cursor, Kiro, Windsurf, and other MCP-compatible tools.
Listed in the official [MCP Registry](https://registry.modelcontextprotocol.io/?q=io.github.lindoai%2Fmcp-server) as **`io.github.lindoai/mcp-server`**.
[](https://glama.ai/mcp/servers/lindoai/mcp-server) [](https://smithery.ai/servers/lindoai/mcp-server)
## Setup
Add to your MCP client configuration:
```json
{
"mcpServers": {
"lindo": {
"command": "npx",
"args": ["-y", "@lindoai/mcp-server"],
"env": {
"LINDO_API_KEY": "your_api_key_here"
}
}
}
}
```
### Where to add this config
| Client | Config file |
|--------|------------|
| Claude Desktop | `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS) |
| Claude Code | `.claude/settings.json` |
| Cursor | `.cursor/mcp.json` |
| Kiro | `.kiro/settings/mcp.json` |
| Windsurf | MCP settings in IDE |
### Getting your API key
1. Go to [app.lindo.ai](https://app.lindo.ai)
2. Navigate to Workspace Settings → API Keys
3. Create a new API key
4. Copy the key and paste it as `LINDO_API_KEY`
Requires a **Business** or **Whitelabel** plan.
## Available Tools
### Websites
- **create_website** — Create a new website using AI from a text prompt
- **list_websites** — List all websites in your workspace
### Pages
- **edit_website** — Edit an existing website (across its pages) using AI
- **publish_blog** — Publish a static blog post with markdown content
### Blog
- **create_blog** — Generate a blog post on a website using AI
### Clients
- **create_client** — Create a new client
- **list_clients** — List all clients
- **assign_website** — Assign a website to a client
- **generate_magic_link** — Generate a client login link
### Credits
- **allocate_credits** — Allocate credits to a client
## Scheduling
The `create_website`, `edit_website`, and `create_blog` tools support an optional `schedule_at` parameter (ISO 8601 datetime) to schedule the workflow for a future time.
## Remote MCP Server
You can also connect via the remote MCP endpoint without installing anything:
```
https://mcp.lindo.ai/mcp
```
## Links
- [Lindo AI](https://lindo.ai)
- [API Documentation](https://lindo.ai/docs)
- [Dashboard](https://app.lindo.ai)
TDQS
Scored across 50 tools
Tools are broadly separated by resource (workspace, client, website, page, blog) and action, and the async workflow/batch variants are clearly scoped. Minor ambiguity exists between create_blog (AI-generated) and publish_blog (static markdown), and get_credits vs get_client_credits requires reading descriptions to distinguish.
Most tools follow a consistent verb_noun pattern (list_websites, create_client, delete_page, unpublish_blog), with batch/check variants following the same convention. Small inconsistencies like get_workspace_team vs add_workspace_team_member and unprefixed get_credits vs get_client_credits keep it from a 5.
At 50 tools, this is an extreme count for a single MCP server; even with a broad CMS scope, the agent surface is very large. Many workflow starters and their polling/batch counterparts inflate the surface and would be better grouped behind fewer tools or explicit sub-resources.
The surface covers nearly full lifecycles for workspaces, clients, websites, pages, and blogs, including CRUD, publishing, analytics, team management, credits, custom domains, and async workflows with batch status polling. No obvious dead ends or critical missing operations exist for the stated platform domain.