Skip to main content
Glama
Sign In
Curizx MCP Server

Curizx MCP Server

by Rayhan55555
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Remote

Curizx MCP Server

Curizx AI Protocol: Model Context Protocol Python: 3.11+ License: MIT Tests: 17 Passed

"Intelligent Tools for AI-Powered Development"
A standalone Model Context Protocol (MCP) server engineered by Rayhan | Curizx AI Architecture.

💡 What it Does

Curizx MCP Server connects any AI assistant (Claude, Cursor, Windsurf) to enterprise-grade developer capabilities. It empowers coding agents with deep GitHub repository intelligence, static AST code analysis, real-time security vulnerability scanning, rate-limited web searches, official doc querying, and automated test generation.

Related MCP server: architectos

🎬 Demo

┌───────────────────────────────────────────────────────────────┐
│                      CURIZX MCP SERVER                        │
│                                                               │
│   Claude / Cursor ──[ STDIO JSON-RPC ]──> FastMCP Server      │
│                                                │              │
│       ┌─────────────────┬──────────────────────┴───────┐      │
│       ▼                 ▼                             ▼      │
│   GitHub Tools     Code Intelligence             Search & Docs│
│   - Analyze Repo   - AST Complexity Analysis     - Web Search │
│   - Review PR      - Security Vulnerability Scan - Docs Search│
│   - Find Issues    - Automated Pytest Generator               │
│   - Gen README     - Code Explainer (3 levels)                │
└───────────────────────────────────────────────────────────────┘

(Demo GIF: assets/curizx-demo.gif placeholder)

⚡ Quick Install

Get up and running in 3 commands:

git clone https://github.com/rayhan/curizx-mcp-server.git
cd curizx-mcp-server && pip install -r requirements.txt
python server.py

🤖 Client Configurations

1. Claude Desktop Config

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "curizx-intelligence": {
      "command": "python",
      "args": [
        "/ABSOLUTE/PATH/TO/curizx-mcp-server/server.py"
      ],
      "env": {
        "GITHUB_TOKEN": "ghp_your_token_here",
        "LOG_LEVEL": "INFO"
      }
    }
  }
}

2. Cursor IDE Config

Add to .cursor/mcp.json or through Cursor Settings > MCP Servers:

{
  "mcpServers": {
    "curizx": {
      "command": "python",
      "args": ["server.py"]
    }
  }
}

🛠️ All 12 MCP Tools

Every tool strictly returns execution time in milliseconds and never leaks secret credentials.

Category

Tool

Input Parameters

What it Does

GitHub

github_analyze_repo

repo (str)

Fetches repo info, detects tech stack, counts files, reads README summary

GitHub

github_review_pr

repo (str), pr_number (int)

Gets PR diff, changed files, analyzes code quality issues and smells

GitHub

github_find_issues

repo (str), path (str, optional)

Scans for missing docstrings, complex functions, unused imports, TODOs

GitHub

github_generate_readme

repo (str)

Reads repo structure and produces professional standard README.md

Code

code_analyze

code (str), language (str)

Full AST cyclomatic complexity analysis, function metrics, code smells

Code

code_security_scan

code (str), language (str)

Scans for SQLi, hardcoded secrets, command injection (secrets masked)

Code

code_suggest_tests

code (str), language (str)

Generates comprehensive pytest templates marked # [REVIEW REQUIRED]

Search

web_search

query (str), max_results (int=5)

DuckDuckGo search with per-session rate limit (10 req/min)

Search

docs_search

query (str), technology (str)

Scoped search for Python, FastAPI, Next.js, React, Postgres, Docker, GitHub

Utility

summarize_text

text (str), max_length (int=200)

Extracts key points, strips conversational fluff, scores sentence importance

Utility

generate_docs

code (str), doc_type (str)

Generates Google-style docstring, REST api spec, or module readme

Utility

explain_code

code (str), level (str)

Explains code logic for beginner, intermediate, or expert developers

🔒 Security Standards

  1. Zero Credential Exposure: Secrets like OpenAI, Anthropic, or GitHub tokens are strictly masked (e.g. sk-ab****, ghp_****).

  2. Path Traversal Guards: All file paths are sanitized against ../ directory escapes.

  3. No Shell Invocations: Zero shell=True or arbitrary command executions.

  4. Rate Limiting: Rolling 60-second window protecting search endpoints.

⚙️ Environment Variables

Copy .env.example to .env:

# GitHub Personal Access Token (for private repos & 5000 req/hr rate limit)
GITHUB_TOKEN="ghp_your_token_here"

# Server Host and Port
MCP_HOST="0.0.0.0"
MCP_PORT=8000

# Logging Level
LOG_LEVEL="INFO"

# Environment
ENVIRONMENT="production"

🧪 Development & Testing

# Run standalone verification test suite
python run_tests.py

# Run pytest with async support
pytest -v --cov=tools

🐳 Docker Deployment

cd docker
docker-compose up -d --build

📜 License

Distributed under the MIT License. Created with excellence by Rayhan | Curizx AI Architecture.

Tool Schema Changelog

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

Loading tool schema changelog…

    This server cannot be deployed

    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Resources

    Related MCP Connectors

    Related MCP Servers