project-advisors-mcp
# Project Advisors MCP Server
An MCP (Model Context Protocol) server providing diagnostic tools to analyze software architecture, security, REST API compliance, and dependencies.
Supports: Python, Rust, Go, Node.js, Java, Swift (iOS), and Kotlin (Android).
> **Inspiration:** The design and concept of these diagnostic advisors were heavily inspired by the **BootUI** platform.
## Setup
```bash
npm install
npm run build
```
## Running the Server
The server runs via stdio:
```bash
npm start
```
## Available Tools
* `run_architecture_scan`
* `run_rest_api_scan`
* `run_security_scan`
* `run_vulnerability_scan`
* `run_pentest_scan`
* `get_telemetry_stream`
* `run_mcp_design_scan`
## Using the Agent Skill
This project includes a custom AI Agent Skill located at `.agents/skills/project-advisor/SKILL.md`. Agents like Antigravity will automatically use it when you ask them to "audit codebase" or "give project health score".
## How to use this MCP Server
To use this library with your AI Coding Assistant, you need to configure the assistant to spawn this server via its `stdio` transport mechanism.
### 1. Claude Desktop Configuration
Add the following to your Claude Desktop config file (e.g. `~/Library/Application Support/Claude/claude_desktop_config.json` on Mac):
```json
{
"mcpServers": {
"project-advisors": {
"command": "node",
"args": ["/Users/ragu/Code/project-advisors/dist/index.js"]
}
}
}
```
*Note: Make sure to run `npm run build` first so the `dist/index.js` file exists, and update the absolute path to point to your repository.*
### 2. Cursor IDE Configuration
1. Open Cursor Settings -> **Features** -> **MCP**.
2. Click **+ Add new MCP server**.
3. Set the Name to `project-advisors`.
4. Set the Type to `command`.
5. Set the Command to `node /absolute/path/to/project-advisors/dist/index.js`.
6. Click **Save**.
### 3. Example AI Prompts
Once connected, you can ask your AI Agent to run audits on your current workspace:
* *"Run an architecture scan on this project and tell me if there are any boundary violations."*
* *"Check my REST APIs for compliance and security issues."*
* *"Give me an overall health score for this project using the project advisor tools."*
TDQS
Scored across 7 tools
Each tool targets a distinct concern (dependencies, endpoints, runtime telemetry, MCP design, architecture, REST APIs, security config), but some overlap exists between security scan, vulnerability scan, and pentest scan. Descriptions clarify boundaries, so misselection is unlikely but possible.
Six of seven tools follow the 'run_*_scan' pattern, which is predictable and consistent. The outlier 'get_telemetry_stream' uses a different verb and noun style, creating a minor deviation but not causing confusion.
Seven tools is well within the ideal range for a specialized scanning server. Each tool covers a distinct aspect of project advisory, and the count feels neither sparse nor bloated.
The tool surface covers the main advisory domains: dependencies, security config, architecture, REST APIs, and runtime telemetry. Minor gaps like container scanning or infrastructure checks exist, but the core workflows are well covered.