Model Context Protocol (MCP) Server Manual

A. Anthropic Claude Desktop

Add the following snippet to your claude_desktop_config.json file:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json
• Linux: ~/.config/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "tasleemat": {
      "command": "tasleemat-mcp",
      "args": []
    }
  }
}

B. Cursor AI IDE

Create or edit .cursor/mcp.json in your project root or user configuration:

{
  "mcpServers": {
    "tasleemat": {
      "command": "python3",
      "args": ["-m", "tasleemat.mcp"]
    }
  }
}

C. VS Code (Cline / Roo Code / Continue.dev)

In your extension settings under MCP Servers, declare:

{
  "mcpServers": {
    "tasleemat": {
      "command": "tasleemat-mcp",
      "args": [],
      "disabled": false,
      "autoApprove": [
        "tasleemat_list_forms",
        "tasleemat_get_form",
        "tasleemat_get_schema",
        "tasleemat_validate",
        "tasleemat_tailor_project",
        "tasleemat_calculate_eva"
      ]
    }
  }
}

1. tasleemat_list_forms

Lists available deliverables filtered by phase, sizing tier, or methodology pack.

// Client Request
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "tasleemat_list_forms",
    "arguments": {
      "phase": "03",
      "lang": "en"
    }
  }
}

// Server Response
{
  "jsonrpc": "2.0",
  "id": 1,
  "result": {
    "content": [
      {
        "type": "text",
        "text": "[{\"id\": \"PMO-03.01\", \"name\": \"Project Charter\", \"phase\": \"03\"}, ...]"
      }
    ]
  }
}

2. tasleemat_get_form

Fetches specific asset contents for a deliverable: template, guide, example, prompt, schema, or dictionary.

// Client Request
{
  "jsonrpc": "2.0",
  "id": 2,
  "method": "tools/call",
  "params": {
    "name": "tasleemat_get_form",
    "arguments": {
      "artifact_id": "PMO-03.01",
      "asset_type": "template",
      "lang": "en"
    }
  }
}

3. tasleemat_get_schema

Retrieves the complete JSON Schema Draft 2020-12 definition to enable client-side schema inspection or agent prompt injection.

// Client Request
{
  "jsonrpc": "2.0",
  "id": 3,
  "method": "tools/call",
  "params": {
    "name": "tasleemat_get_schema",
    "arguments": {
      "artifact_id": "PMO-04.08.02"
    }
  }
}

4. tasleemat_validate

Validates a structured JSON payload against the companion schema and returns exact JSON path errors.

// Client Request
{
  "jsonrpc": "2.0",
  "id": 4,
  "method": "tools/call",
  "params": {
    "name": "tasleemat_validate",
    "arguments": {
      "artifact_id": "PMO-03.01",
      "data": {
        "project_id": "PRJ-01",
        "project_name": "Cloud Platform",
        "sponsor": "Sarah Al-Ahmad"
      },
      "mode": "strict"
    }
  }
}

// Server Response
{
  "jsonrpc": "2.0",
  "id": 4,
  "result": {
    "content": [
      {
        "type": "text",
        "text": "{\"valid\": true, \"errors\": [], \"schema_id\": \"https://fakhr.me/schemas/pmo-03-01.json\"}"
      }
    ]
  }
}

5. tasleemat_tailor_project & 6. tasleemat_calculate_eva

tasleemat_tailor_project returns required deliverables for a tier/pack. tasleemat_calculate_eva computes attested EVM math metrics for PMO-06.05.

// Calculate EVA Metrics Request
{
  "jsonrpc": "2.0",
  "id": 6,
  "method": "tools/call",
  "params": {
    "name": "tasleemat_calculate_eva",
    "arguments": {
      "pv": 500000,
      "ev": 475000,
      "ac": 520000,
      "bac": 1200000
    }
  }
}

Tutorial 1: Autonomous PMO Deliverable Drafting in Cursor Agent Mode

Once .cursor/mcp.json is saved, switch Cursor Composer to Agent Mode and prompt:

Cursor will autonomously execute tasleemat_get_form, inspect constraints via tasleemat_get_schema, run tasleemat_validate to confirm schema conformance, and write the finalized markdown file.

Tutorial 2: Building a Standalone Python MCP Client

You can interact with tasleemat-mcp from any Python application via standard subprocessing:

import subprocess
import json

def call_mcp_tool(tool_name: str, arguments: dict) -> dict:
    # Invokes a tool on the Tasleemat MCP server via stdio JSON-RPC
    proc = subprocess.Popen(
        ["tasleemat-mcp"],
        stdin=subprocess.PIPE,
        stdout=subprocess.PIPE,
        stderr=subprocess.PIPE,
        text=True
    )
    
    # 1. Initialize Handshake
    init_req = {"jsonrpc": "2.0", "id": 1, "method": "initialize", "params": {"capabilities": {}}}
    proc.stdin.write(json.dumps(init_req) + "\n")
    proc.stdin.flush()
    proc.stdout.readline()  # consume init response
    
    # 2. Call Tool
    call_req = {
        "jsonrpc": "2.0",
        "id": 2,
        "method": "tools/call",
        "params": {"name": tool_name, "arguments": arguments}
    }
    proc.stdin.write(json.dumps(call_req) + "\n")
    proc.stdin.flush()
    
    response_line = proc.stdout.readline()
    proc.terminate()
    return json.loads(response_line)

# Example: tailor project for Tier 2 Agile
result = call_mcp_tool("tasleemat_tailor_project", {"tier": 2, "pack": "agile"})
print("Tailored Result:", result)