Skip to main content

Simple MCP client that connects to your AgenticWerx MCP server to retrieve rules

Project description

AgenticWerx MCP Client

PyPI version Python Support License: MIT

A Model Context Protocol (MCP) client that connects to the AgenticWerx Lambda MCP server using JSON-RPC 2.0 protocol to provide code analysis and rule management capabilities.

🚀 Quick Start

The AgenticWerx MCP Client supports two modes:

CLI Mode - Direct Command Line Usage

# Get marketplace rules
uvx agenticwerx-mcp-client@latest --api-key YOUR_KEY get-rules

# Get custom rules (Pro/Pro Plus)
uvx agenticwerx-mcp-client@latest --api-key YOUR_KEY get-custom-rules --language python

# Get team rules (Pro Plus teams)
uvx agenticwerx-mcp-client@latest --api-key YOUR_KEY get-team-rules --category security

# Analyze code
uvx agenticwerx-mcp-client@latest --api-key YOUR_KEY analyze-code --file script.py

# Analyze code snippet
uvx agenticwerx-mcp-client@latest --api-key YOUR_KEY analyze-code \
  --code "print('hello')" --language python

Features:

  • 🔍 Auto-detects programming language from file extensions
  • 📦 Automatically chunks large files (>8KB) for analysis
  • 📊 Aggregates results from multiple chunks
  • 🎯 Supports 20+ programming languages

See CLI_USAGE.md for detailed CLI documentation.

MCP Server Mode - For IDE Integration

Add this configuration to your MCP-compatible IDE (Kiro, Amazon Q Developer, etc.):

{
  "mcpServers": {
    "agenticwerx": {
      "command": "uvx",
      "args": ["agenticwerx-mcp-client@latest"],
      "env": {
        "AGENTICWERX_API_KEY": "your-api-key-here"
      }
    }
  }
}

Get Your API Key

  1. Visit AgenticWerx Dashboard
  2. Navigate to API Keys section
  3. Create a new API key
  4. Use it in CLI or MCP configuration

🛠️ Available Tools

MCP Server Mode Tools

get_rules

Get AgenticWerx marketplace rules from subscribed packages.

Parameters:

  • packageId (optional): Specific package ID to filter rules
  • language (optional): Filter by programming language (e.g., typescript, python, go)
  • framework (optional): Filter by framework (e.g., react, nextjs, express)
  • category (optional): Filter by category (e.g., security, api-design, testing)
  • severity (optional): Filter by severity (critical, high, medium, low)
  • appliesTo (optional): Filter by applies to (e.g., frontend, backend, api, all)
  • search (optional): Search in rule titles and descriptions
  • ruleIds (optional): Get specific rules by ID
  • detailed (optional): Get full rule details (instructions, rationale, tags). Default: false
  • includeContent (optional): Get complete rule content (markdown/JSON). Default: false
  • includePatterns (optional): Include regex patterns (requires detailed=true). Default: false
  • limit (optional): Maximum rules to return (1-200). Default: 50
  • offset (optional): Skip N rules for pagination. Default: 0

get_custom_rules

Get user-uploaded custom rules (Pro/Pro Plus only).

Custom rules are rules you've uploaded to your AgenticWerx account. These are private to your account and can be used alongside marketplace rules.

Parameters: Same filtering options as get_rules (except packageId):

  • language, framework, category, severity, appliesTo, search, ruleIds
  • detailed, includeContent, includePatterns
  • limit, offset

get_team_rules

Get team-shared custom rules (Pro Plus teams only).

Team rules are custom rules shared across your Pro Plus team. These are visible to all team members and can be used alongside marketplace and custom rules.

Parameters: Same filtering options as get_rules (except packageId):

  • language, framework, category, severity, appliesTo, search, ruleIds
  • detailed, includeContent, includePatterns
  • limit, offset

analyze_code

Analyze code using AgenticWerx rules from the server.

Parameters:

  • code (required): Code snippet to analyze
  • language (optional): Programming language
  • packageIds (optional): Array of package IDs to use for analysis

Example:

{
  "tool": "analyze_code",
  "code": "print('hello world')",
  "language": "python",
  "packageIds": ["stripe-integration-excellence-pack"]
}

The tool connects to AgenticWerx services and retrieves the rules, which are then processed and returned to your IDE.

External MCP Server Tools 🆕

Connect to external MCP servers (GitHub, AWS Docs, PostgreSQL, etc.) alongside AgenticWerx rules.

list_external_mcp_servers

Discover available external MCP servers from the AgenticWerx registry.

Parameters: None

Returns: List of pre-configured servers (GitHub, AWS Docs, PostgreSQL, etc.)

get_external_mcp_server

Get connection details for a specific external MCP server.

Parameters:

  • serverId (required): Server ID from registry (e.g., "github", "aws-documentation")

Returns: Command, arguments, and required environment variables

connect_external_server

Connect to an external MCP server. Supports two modes:

Mode A - Using serverId (Recommended):

{
  "serverId": "github",
  "env": {
    "GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_xxx"
  }
}

Mode B - Manual Connection:

{
  "name": "custom-server",
  "command": "python",
  "args": ["my_server.py"],
  "env": {
    "CUSTOM_VAR": "value"
  }
}

Features:

  • ✅ Automatic persistence - connections restored on restart
  • ✅ Multiple simultaneous connections
  • ✅ Tool discovery and execution
  • ✅ Support for custom servers

list_external_connections

List all currently connected external MCP servers and their available tools.

disconnect_external_server

Disconnect from an external MCP server.

Parameters:

  • serverName (required): Name of the server to disconnect

call_external_tool

Call a tool on a connected external MCP server.

Parameters:

  • serverName (required): Name of the connected server
  • toolName (required): Name of the tool to call
  • arguments (optional): Arguments to pass to the tool

See EXTERNAL_MCP_SERVERS.md for detailed documentation.

🔗 Simple Connection

This client acts as a simple bridge between your IDE and AgenticWerx services. It retrieves rules and passes them back to your IDE for code analysis.

🔧 Installation Methods

Method 1: UVX (Recommended)

No installation needed! Your IDE will automatically download and run the client:

uvx agenticwerx-mcp-client@latest --api-key your_key_here

Method 2: pip install

pip install agenticwerx-mcp-client
agenticwerx-mcp-client --api-key your_key_here

Method 3: From Source

git clone https://github.com/agenticwerx/mcp-client.git
cd mcp-client
pip install -e .
agenticwerx-mcp-client --api-key your_key_here

📋 IDE Configuration Examples

Kiro IDE

{
  "mcpServers": {
    "agenticwerx": {
      "command": "uvx",
      "args": ["agenticwerx-mcp-client@latest", "--api-key", "${AGENTICWERX_API_KEY}"],
      "env": {
        "AGENTICWERX_API_KEY": "your-api-key-here"
      }
    }
  }
}

Amazon Q Developer

{
  "mcpServers": {
    "agenticwerx": {
      "command": "uvx",
      "args": ["agenticwerx-mcp-client@latest", "--api-key", "${AGENTICWERX_API_KEY}"],
      "env": {
        "AGENTICWERX_API_KEY": "your-api-key-here"
      }
    }
  }
}

VS Code (with MCP extension)

{
  "mcp.servers": {
    "agenticwerx": {
      "command": "uvx",
      "args": ["agenticwerx-mcp-client@latest", "--api-key", "${AGENTICWERX_API_KEY}"],
      "env": {
        "AGENTICWERX_API_KEY": "your-api-key-here"
      }
    }
  }
}

🔒 Security & Privacy

  • API Key Security: Your API key is only used to authenticate with AgenticWerx services
  • Code Privacy: Code analysis happens securely through encrypted connections
  • No Data Storage: Your code is analyzed in real-time and not stored on our servers
  • Local Processing: The MCP client runs locally on your machine

🚀 Features

  • Simple Connection: Connects your IDE to AgenticWerx services
  • Rule Retrieval: Fetches rules from the server
  • MCP Compatible: Works with any MCP-compatible IDE
  • Zero Configuration: Just add your API key
  • Lightweight: Minimal overhead, just passes data through

📊 Example Output

{
  "tool": "analyze",
  "packageId": "stripe-integration-excellence-pack",
  "rules": {
    "rules": [
      {
        "id": "rule-1",
        "name": "Security Rule",
        "description": "Prevents security vulnerabilities",
        "pattern": "eval\\(",
        "message": "Avoid using eval() as it can lead to code injection"
      }
    ],
    "metadata": {
      "package_name": "Security Rules",
      "version": "1.0.0",
      "total_rules": 1
    }
  }
}

🛠️ Technical Requirements

Runtime Requirements

  • Python 3.8+
  • httpx >= 0.25.0
  • mcp >= 1.0.0
  • pydantic >= 2.0.0

Installation

# Via uvx (recommended)
uvx agenticwerx-mcp-client@latest --api-key your_key_here

# Via pip
pip install agenticwerx-mcp-client

📚 Documentation

🆘 Support

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

Note: This is a proprietary package developed and maintained exclusively by AgenticWerx. We do not accept external contributions at this time.

🔄 Changelog

v1.0.0 (2025-01-XX)

  • Initial release
  • Full MCP protocol support
  • Rule retrieval tools
  • Multi-language support
  • Real-time code feedback

Built with ❤️ by the AgenticWerx Team

Making code quality accessible to every developer, in every IDE, for every language.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

agenticwerx_mcp_client-1.5.3.tar.gz (38.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agenticwerx_mcp_client-1.5.3-py3-none-any.whl (44.3 kB view details)

Uploaded Python 3

File details

Details for the file agenticwerx_mcp_client-1.5.3.tar.gz.

File metadata

  • Download URL: agenticwerx_mcp_client-1.5.3.tar.gz
  • Upload date:
  • Size: 38.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.5

File hashes

Hashes for agenticwerx_mcp_client-1.5.3.tar.gz
Algorithm Hash digest
SHA256 ec2171a6027f5152d2630a56572842a18cdb0cede4d633400f7ef1b29af55a50
MD5 2ebdff87b54531dbe14c3aacff426b84
BLAKE2b-256 2d277005e90bec15bd03f41a0b0933a6bc018890363ec96cb13379ebb58277c6

See more details on using hashes here.

File details

Details for the file agenticwerx_mcp_client-1.5.3-py3-none-any.whl.

File metadata

File hashes

Hashes for agenticwerx_mcp_client-1.5.3-py3-none-any.whl
Algorithm Hash digest
SHA256 dcbb548c5e92a1d4bf303cf01eaccbe36c96eab2cda170abae18b5bbc58112bb
MD5 00f93c5458f1413f15e88d9639c34fed
BLAKE2b-256 4cc4e08d6c23823788f3ec1d1aaa80c51ae08db044944e322dccb98c0427ac89

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page