uvx idea-reality-mcp

claude mcp add idea-reality -- uvx idea-reality-mcp

{
  "mcpServers": {
    "idea-reality": {
      "command": "uvx",
      "args": ["idea-reality-mcp"]
    }
  }
}

npx -y @smithery/cli install idea-reality-mcp --client claude

idea-reality setup

idea-reality config              # interactive menu
idea-reality config claude_code  # auto-installs
idea-reality config cursor       # prints Cursor config
idea-reality config raw_json     # generic MCP JSON

idea-reality doctor        # core checks (~2s)
idea-reality doctor --full # + API validation, all 6 sources

export GITHUB_TOKEN=ghp_...        # Higher GitHub API rate limits
export PRODUCTHUNT_TOKEN=...       # Enable Product Hunt (deep mode)

{
  "tool": "idea_check",
  "arguments": {
    "idea_text": "a CLI tool that converts Figma designs to React components",
    "depth": "deep"
  }
}

{
  "reality_signal": 72,
  "duplicate_likelihood": "high",
  "trend": "accelerating",
  "sub_scores": {
    "market_momentum": 73
  },
  "evidence": [
    {
      "source": "github",
      "type": "repo_count",
      "query": "figma react converter CLI",
      "count": 342
    },
    {
      "source": "github",
      "type": "max_stars",
      "query": "figma react converter",
      "count": 15000
    },
    {
      "source": "hackernews",
      "type": "mention_count",
      "query": "figma react",
      "count": 18
    },
    {
      "source": "npm",
      "type": "package_count",
      "query": "figma-react",
      "count": 56
    },
    {
      "source": "pypi",
      "type": "package_count",
      "query": "figma-react",
      "count": 23
    },
    {
      "source": "producthunt",
      "type": "product_count",
      "query": "figma react",
      "count": 8
    },
    {
      "source": "stackoverflow",
      "type": "question_count",
      "query": "figma react converter",
      "count": 120
    }
  ],
  "top_similars": [
    {
      "name": "figma-to-react/figma-to-react",
      "url": "https://github.com/figma-to-react/figma-to-react",
      "stars": 15000,
      "description": "Convert Figma designs to React components"
    }
  ],
  "pivot_hints": [
    "High competition. Consider a niche differentiator like supporting design tokens or specific component libraries.",
    "The leading project may have gaps in TypeScript support or accessibility features."
  ]
}

import httpx

response = httpx.post(
    "https://idea-reality-mcp.onrender.com/api/check",
    json={
        "idea_text": "AI code review tool",
        "depth": "deep"
    },
    timeout=30.0
)

result = response.json()
print(f"Reality Score: {result['reality_signal']}/100")
print(f"Trend: {result['trend']}")
print(f"Top Competitor: {result['top_similars'][0]['name']} ({result['top_similars'][0]['stars']} ⭐)")

curl -X POST https://idea-reality-mcp.onrender.com/api/check \
  -H "Content-Type: application/json" \
  -d '{
    "idea_text": "a markdown-based static site generator with live reload",
    "depth": "quick"
  }'

# User says: "I want to build a CLI tool for GitHub issue management"
# Agent calls:
{
  "tool": "idea_check",
  "arguments": {
    "idea_text": "CLI tool for GitHub issue management with labels and milestones",
    "depth": "deep"
  }
}

# If reality_signal > 80:
#   → Suggest niche differentiation or pivot
# If reality_signal 40-80:
#   → Validate unique features, check top_similars
# If reality_signal < 40:
#   → Green light, low competition

{
  "tool": "idea_check",
  "arguments": {
    "idea_text": "add real-time collaborative editing to my markdown editor",
    "depth": "quick"
  }
}

{
  "tool": "idea_check",
  "arguments": {
    "idea_text": "browser automation library using Chrome DevTools Protocol",
    "depth": "deep"
  }
}

When starting a new project, use the idea_check MCP tool to check if similar projects already exist.

name: Idea Reality Check
on:
  issues:
    types: [opened]

jobs:
  check:
    if: contains(github.event.issue.labels.*.name, 'proposal')
    runs-on: ubuntu-latest
    steps:
      - uses: mnemox-ai/idea-check-action@v1
        with:
          idea: ${{ github.event.issue.title }}
          github-token: ${{ secrets.GITHUB_TOKEN }}

claude mcp list  # verify installation
claude mcp remove idea-reality
claude mcp add idea-reality -- uvx idea-reality-mcp

export GITHUB_TOKEN=ghp_your_token_here
idea-reality doctor --full  # verify

export PRODUCTHUNT_TOKEN=your_token_here

# Python httpx client
httpx.post(..., timeout=30.0)

User: "I want to build a Rust-based SQL formatter with auto-fix"

Agent:
1. Call idea_check with depth="deep"
2. If reality_signal > 70:
   - Show top competitors (e.g. sqlformat, prettier-plugin-sql)
   - Suggest niches from pivot_hints (e.g. Rust performance angle)
3. If reality_signal < 50:
   - Proceed with project scaffolding

User: "Should I add Vim keybindings to my editor?"

Agent:
1. Call idea_check: "text editor with vim keybindings"
2. Check evidence[].count for npm/PyPI packages
3. High count → already solved, suggest integration
4. Low count → potential differentiator

User: "Is AI code review still worth building?"

Agent:
1. Call idea_check with depth="deep"
2. Check trend field
3. If "accelerating" → market growing, move fast
4. If "declining" → suggest pivot to specific niche

import httpx

ideas = [
    "AI-powered commit message generator",
    "Real-time Markdown collaboration",
    "GitHub issue templates manager"
]

async with httpx.AsyncClient() as client:
    tasks = [
        client.post(
            "https://idea-reality-mcp.onrender.com/api/check",
            json={"idea_text": idea, "depth": "quick"}
        )
        for idea in ideas
    ]
    results = await asyncio.gather(*tasks)

for idea, resp in zip(ideas, results):
    data = resp.json()
    print(f"{idea}: {data['reality_signal']}/100 ({data['trend']})")

def should_build(result: dict) -> str:
    score = result['reality_signal']
    trend = result['trend']
    
    if score < 40:
        return "BUILD: Low competition, novel idea"
    elif score < 70 and trend == "accelerating":
        return "BUILD WITH NICHE: Growing market, differentiate"
    elif score > 85:
        return "PIVOT: Saturated market"
    else:
        return "RESEARCH: Validate unique angle"

response = httpx.post(...)
verdict = should_build(response.json())