# Idea Reality MCP

Idea Reality MCP is an open-source MCP server that provides pre-build reality checks for AI coding agents. It automatically scans GitHub, Hacker News, npm, PyPI, Product Hunt, and Stack Overflow for existing projects, returning a 0-100 reality signal, duplicate likelihood, top competitors, evidence, and pivot suggestions before any code is written.

Canonical URL: https://a2a-mcp.org/entry/idea-reality-mcp

Language: en

Category: MCP Servers

Updated: 2026-03-23

## Overview
Idea Reality MCP (also known as idea-reality-mcp) is a lightweight, open-source **MCP-compatible server** developed by mnemox-ai. It solves a common pain point for AI coding agents and developers: spending hours or days building something only to discover it already exists.

The server acts as a protocol-layer tool that AI agents (like Claude or Cursor) can call automatically. It performs real-time searches across multiple public databases and returns objective, verifiable data so the agent can decide whether to proceed, pivot, or abandon the idea — *before writing a single line of code*.

It gained rapid popularity after its Show HN launch in February 2026, reaching hundreds of stars on GitHub.

## Key Features
- **Multi-Source Reality Scanning**: Queries GitHub (repos & stars), Hacker News (discussions), npm & PyPI (packages), Product Hunt (launches), and Stack Overflow.
- **Quantified Reality Signal**: Returns a 0-100 score indicating how "real" / saturated the idea is, along with duplicate likelihood.
- **Actionable Output**: Top similar projects, competitor analysis, evidence links, trend detection, and intelligent pivot suggestions.
- **Automatic Integration**: Designed for seamless use inside agent workflows — no manual searches required.
- **Lightweight & Zero-Config**: Runs via `uvx`, works with Claude Desktop, Cursor, Windsurf, and any MCP client.
- **Privacy-First**: Searches are performed on-demand with real API calls; no data stored.
- **MIT Licensed**: Fully open source and extensible.

## How It Works
When an AI agent receives a project idea or spec, it can call the `idea_check` tool provided by the MCP server. The server aggregates fresh data from public sources, analyzes overlap, and responds with structured JSON including:

- `reality_signal` (0-100)
- `duplicate_likelihood`
- List of top similar repos/projects with links and metrics
- Evidence summaries
- Pivot or differentiation recommendations

This keeps the agent grounded in reality and prevents wasted effort on saturated ideas.

## Quick Start
```bash
# One-command install and run
uvx idea-reality-mcp
```

Add to Claude / Cursor (example config):
```json
{
  "mcpServers": {
    "idea-reality": {
      "command": "uvx",
      "args": ["idea-reality-mcp"]
    }
  }
}
```

The agent will then automatically invoke the check during planning.

## Use Cases
- Preventing AI agents from duplicating popular open-source tools or startups
- Early-stage idea validation for solo developers and indie hackers
- Enterprise prompt engineering workflows that require grounded research
- Integration into autonomous coding agents, research agents, or product-building loops
- Complementing tools like Cursor, Claude Code, or other MCP-powered IDE experiences

## Background & Status
Created after the maintainer wasted 6 hours building a tool that already had 847 similar GitHub repos. Released as a practical MCP server rather than a SaaS dashboard. Actively maintained, with ongoing improvements to search coverage and scoring logic.

**GitHub Repository**: https://github.com/mnemox-ai/idea-reality-mcp (MIT License)

Also listed on MCP registries including smithery.ai, mcpservers.org, and others.

For the latest version, documentation, and web demo, visit the official GitHub repository.

## Tags

- mcp-server
- ai-agent
- reality-check
- idea-validation
- github-search
- hn-search
- npm
- pypi
- product-hunt
- mit
- python
- claude
- cursor

## Sources

- [Project website](https://github.com/mnemox-ai/idea-reality-mcp)