mitra-mcp
A modular MCP server integrating Clockify, WakaTime, and Azure DevOps for time tracking, work item management, and time logging via a unified API.
README
Mitra MCP Server
Mitra is a modular, stateless Model Context Protocol (MCP) server that integrates Clockify, WakaTime, and Azure DevOps. It enables developers and local AI assistants to fetch active projects, manage Azure DevOps work items (cards), and log time entries directly to Clockify using a unified workflow.
The server is fully stateless: individual team members supply their credentials (API keys, workspace IDs, Personal Access Tokens) via request headers (in remote/SSE mode) or local environment variables (in stdio mode).
Code Architecture
Mitra uses an integration-per-folder architecture with auto-discovery. Each integration is self-contained — its client, tools, prompts, and context live together in one folder:
src/mitra/
├── __init__.py
├── server.py # FastMCP instance + auto-discovery (never needs editing)
├── cli.py # CLI entrypoint (never needs editing)
│
├── core/ # Shared infrastructure
│ ├── __init__.py
│ ├── context.py # Base credential resolution helpers
│ └── registry.py # Auto-discovery engine
│
└── integrations/ # ← Each integration is one self-contained folder
├── __init__.py
│
├── clockify/ # Clockify time tracking
│ ├── __init__.py # register(mcp) entry point
│ ├── client.py # ClockifyClient (API wrapper)
│ ├── tools.py # @mcp.tool() definitions
│ ├── prompts.py # @mcp.prompt() + @mcp.resource()
│ └── context.py # Context vars, HTTP headers, resolvers
│
├── wakatime/ # WakaTime coding activity
│ ├── __init__.py
│ ├── client.py
│ ├── tools.py
│ └── context.py
│
├── azure_devops/ # Azure DevOps work items
│ ├── __init__.py
│ ├── client.py
│ ├── tools.py
│ ├── prompts.py
│ └── context.py
│
└── workflows/ # Cross-integration composite tools
├── __init__.py
├── linkage.py # Clockify ↔ Azure DevOps linkage
├── fill_clockify.py # Composite fill-timesheet tools
└── prompts.py # Unified Mitra agent guide
Adding a New Integration (Developer Guide)
Adding a new integration (for example, Jira or Github) requires creating one folder — no other files need to be modified:
mkdir -p src/mitra/integrations/jira
1. Create the entry point (__init__.py):
# src/mitra/integrations/jira/__init__.py
from mitra.integrations.jira.tools import register_tools
def register(mcp):
register_tools(mcp)
2. Create the API client (client.py):
# src/mitra/integrations/jira/client.py
class JiraClient:
def __init__(self, api_key: str): ...
async def list_issues(self, project: str): ...
3. Create the tools (tools.py):
# src/mitra/integrations/jira/tools.py
from mitra.integrations.jira.client import JiraClient
def register_tools(mcp):
@mcp.tool()
async def jira_list_issues(project: str, api_key: str) -> list:
"""Lists Jira issues for a project."""
client = JiraClient(api_key)
return await client.list_issues(project)
4. (Optional) Add credential headers (context.py):
# src/mitra/integrations/jira/context.py
import contextvars
from mitra.core.context import resolve_credential
request_jira_api_key = contextvars.ContextVar("jira_api_key", default=None)
HEADERS = {"x-jira-api-key": request_jira_api_key}
def get_jira_api_key():
return resolve_credential(request_jira_api_key, "JIRA_API_KEY")
That's it. The auto-discovery engine picks up your new folder automatically. The SSE middleware auto-collects your headers. No other files to touch.
Installation
Clone the repository and install in editable mode:
pip install -e .
Usage
1. Local stdio Mode (For Local IDEs / Claude Desktop)
In this mode, the server reads credentials directly from the shell environment.
Set the required environment variables:
export CLOCKIFY_API_KEY="your-clockify-api-key"
export CLOCKIFY_WORKSPACE_ID="your-clockify-workspace-id"
export WAKATIME_API_KEY="your-wakatime-api-key"
export AZURE_DEVOPS_PAT="your-azure-devops-pat"
export AZURE_DEVOPS_ORG="https://dev.azure.com/your-org"
Start the server:
mitra start --transport stdio
Claude Desktop Integration
To use Mitra locally with the Claude Desktop app, configure the server in your Claude Desktop configuration file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Add the following JSON snippet under the mcpServers key:
{
"mcpServers": {
"mitra": {
"command": "/absolute/path/to/your/venv/bin/mitra",
"args": [
"start",
"--transport",
"stdio"
],
"env": {
"CLOCKIFY_API_KEY": "your-clockify-api-key",
"CLOCKIFY_WORKSPACE_ID": "your-clockify-workspace-id",
"WAKATIME_API_KEY": "your-wakatime-api-key",
"AZURE_DEVOPS_PAT": "your-azure-devops-pat",
"AZURE_DEVOPS_ORG": "https://dev.azure.com/your-org"
}
}
}
}
[!NOTE] Make sure to replace
/absolute/path/to/your/venv/bin/mitrawith the actual path to themitraexecutable inside your Python virtual environment (e.g.,which mitra).
Claude Code (CLI) Integration
To use Mitra with the Claude Code CLI, register it using the claude mcp add command.
Run the following command in your terminal to configure the server (add the --scope user flag if you want it to be globally available across all projects):
claude mcp add mitra --scope user \
-e CLOCKIFY_API_KEY="your-clockify-api-key" \
-e CLOCKIFY_WORKSPACE_ID="your-clockify-workspace-id" \
-e WAKATIME_API_KEY="your-wakatime-api-key" \
-e AZURE_DEVOPS_PAT="your-azure-devops-pat" \
-e AZURE_DEVOPS_ORG="https://dev.azure.com/your-org" \
-- /absolute/path/to/your/venv/bin/mitra start --transport stdio
You can view active servers by typing /mcp inside your Claude Code session, or check the list using claude mcp list.
2. Remote SSE Mode (For Web Services)
In remote mode, the server is hosted as an HTTP app. Clients supply credentials on a per-request basis using HTTP headers:
X-Clockify-Api-KeyX-Clockify-Workspace-IdX-Wakatime-Api-KeyX-Azure-Devops-PatX-Azure-Devops-Org
Start the server:
mitra start --transport sse --host 127.0.0.1 --port 8000
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.