Skip to main content
Glama
YourAppSpace MCP Server

YourAppSpace MCP Server

Official
by yourappspace

YourAppSpace MCP Server

Give AI coding agents (Claude Code, Cursor, OpenAI Codex CLI, JetBrains AI Assistant) a structured backend with a built-in Visual Schema Designer, Data Grid UI, auto-generated REST APIs, and Row/Field-Level Security (RLS & FLS).

Live Demo Documentation Video Tutorial License: MIT


Why YourAppSpace MCP?

Most database MCP servers give an LLM raw SQL access to a database—risking destructive schema changes, leaking restricted columns, and leaving you with no UI to inspect or manage the data.

YourAppSpace Core solves this by pairing an AI-ready Entity MCP with a Visual Data Platform:

  • Visual Schema + Data Grid UI: Everything your AI agent creates via MCP (create_table, add_field, create_rows) immediately appears in a live visual schema designer and spreadsheet-like Data Grid UI with filtering and Excel/CSV export.

  • Safe by Design (RLS & FLS): Agents operate through the platform's Entity API governed by Row-Level Security and Field-Level Security instead of raw SQL—no accidental DROP TABLE or unauthorized field reads.

  • Instant REST APIs: Every table modeled by your agent automatically exposes documented REST endpoints.

  • Zero Vendor Lock-in: Run on YourAppSpace Cloud (https://core.yourappspace.com) or deploy on-premise via Docker.

Hosted Remote MCP (Zero-Install) — Coming Soon: In the next release of YourAppSpace Core, you will be able to connect directly to a Hosted Remote MCP endpoint (Streamable HTTP) straight from your workspace URL—requiring zero local installation, no Node.js, and no local processes. Until then (and for local Docker / air-gapped on-premise deployments), use this lightweight open-source STDIO connector.


Related MCP server: Apper MCP

Prerequisites

  • Node.js >= 20.0.0

  • A YourAppSpace Core workspace — click Try now on core.yourappspace.com for an instant demo workspace, or sign up for free.


Quick Start (Local STDIO Connector)

  1. Clone the repository and install dependencies:

    git clone https://github.com/yourappspace/yourappspace-mcp.git
    cd yourappspace-mcp
    npm install
  2. Connect your AI client using one of the configurations below.


Client Configuration

Replace <path_to_repo> with the absolute path to yourappspace-mcp, and set your workspace URL and credentials.

1. Claude Code (.mcp.json)

Create or update .mcp.json in your project root:

{
  "mcpServers": {
    "dev-entity": {
      "command": "node",
      "args": ["<path_to_repo>/server.js"],
      "env": {
        "MCP_ENTITY_BASE_URL": "https://core.yourappspace.com",
        "MCP_ENTITY_USER": "your_email@example.com",
        "MCP_ENTITY_PASSWORD": "your_password"
      }
    }
  }
}

Verify active tools inside Claude Code by running /mcp.

2. Cursor (.cursor/mcp.json)

Add to .cursor/mcp.json in your project or global ~/.cursor/mcp.json:

{
  "mcpServers": {
    "dev-entity": {
      "command": "node",
      "args": ["<path_to_repo>/server.js"],
      "env": {
        "MCP_ENTITY_BASE_URL": "https://core.yourappspace.com",
        "MCP_ENTITY_USER": "your_email@example.com",
        "MCP_ENTITY_PASSWORD": "your_password"
      }
    }
  }
}

3. OpenAI Codex CLI (~/.codex/config.toml)

[mcp_servers.dev-entity]
command = "node"
args = ["<path_to_repo>/server.js"]

[mcp_servers.dev-entity.env]
MCP_ENTITY_BASE_URL = "https://core.yourappspace.com"
MCP_ENTITY_USER = "your_email@example.com"
MCP_ENTITY_PASSWORD = "your_password"

4. JetBrains IntelliJ IDEA AI Assistant

  1. Open Settings → Tools → AI Assistant → Model Context Protocol (MCP).

  2. Click + Add and select STDIO.

  3. Paste the JSON configuration above (same format as Claude Code / Cursor) and click Apply.


Available MCP Tools (8 Tools)

Tool

Purpose

Modifies Data

list_tables

Lists accessible tables across user and system spaces

No

get_table

Retrieves table metadata, column types, and schema definition

No

create_table

Creates a new custom table in the active workspace

Yes

add_field

Adds a data column (INTEGER, STRING, DATETIME, etc.) or lookup relationship

Yes

read_rows

Queries records with WHERE / JSON filters, ordering, and pagination

No

create_rows

Inserts one or multiple records in batch mode

Yes

update_rows

Updates one or multiple records by primary key

Yes

delete_rows

Deletes one or multiple records by primary key

Yes


System Prompt for AI Agents (BASE_PROMPT.md)

This repository includes BASE_PROMPT.md, which gives AI agents full knowledge of YourAppSpace core data types, field widgets, and lookup relationships:

  • Claude Code: Copy BASE_PROMPT.md to CLAUDE.md in your project root.

  • Codex CLI: Copy BASE_PROMPT.md to AGENTS.md in your project root.

  • IntelliJ AI Assistant: Add it to Settings → Tools → AI Assistant → Prompt Library.


Environment Variables

Variable

Description

Example

MCP_ENTITY_BASE_URL

Base URL of your YourAppSpace instance (SaaS or Docker)

https://core.yourappspace.com or http://localhost:8080

MCP_ENTITY_USER

Email / username of the authorized workspace user

demo_manager@yourappspace.com

MCP_ENTITY_PASSWORD

Password for authentication

••••••••


Resources


License

MIT

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Turns your data model into a semantic layer for AI agents, automatically generating typed, discoverable tools with entity relationships and schema discovery.
      20 PyPI
      644
      Apache 2.0
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI assistants to build and update complete applications, including creating, updating, publishing, previewing, and configuring databases, security, permissions, and server-side integrations.
      Apache 2.0
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to read typed objects and propose, approve, and apply audited actions with permission-aware retrieval, enforcing visibility inside queries so each principal sees only authorized data and properties.
      MIT