3 Servers · 107 Tools · Open Source

MCP Service

Open-source MCP servers by the Flaremity team, deployed on Cloudflare Workers. Connect your AI assistant to Atlassian, GitLab, and Datadog through the Model Context Protocol.

41 Atlassian · 44 GitLab · 22 Datadog

Explore Servers GitHub

MCP Servers

Three production-ready MCP servers, each deployed as a stateless Cloudflare Worker. Zero data stored. Instant global access.

41 tools

MCP Atlassian

27 Jira + 14 Confluence

Connect your AI to Jira and Confluence. Manage issues, search pages, transition workflows, and more.

IssuesSearchSprintsPagesSpacesComments

Dual (Token + OAuth)

Visit GitHub

44 tools

MCP GitLab

25 Read + 19 Write

Full GitLab integration. Projects, merge requests, pipelines, issues, and CI/CD job logs.

ProjectsMRsPipelinesIssuesLabelsGroups

PAT (Personal Access Token)

Visit GitHub

22 tools

MCP Datadog

Read-only · 11 domains

Monitor your infrastructure with AI. Dashboards, logs, metrics, incidents, APM traces, and more.

MonitorsDashboardsLogsMetricsIncidentsHosts

Dual-key (API + App Key)

Visit GitHub

How It Works

MCP servers on Cloudflare Workers — the simplest way to give AI access to your tools.

What is MCP?

The Model Context Protocol (MCP) is an open standard that lets AI assistants connect to external tools and data sources. Think of it as USB-C for AI — a universal interface between AI models and the services they need.

Instead of building custom integrations for each AI platform, MCP provides a single protocol that works with Claude, ChatGPT, Cursor, and any MCP-compatible client.

Why Cloudflare Workers?

Every MCP server runs as a stateless Cloudflare Worker.

🌍

Edge-First

Deployed to 300+ data centers. Your MCP server runs close to your users, worldwide.

📦

Stateless

No database, no sessions, no data retention. Pure proxy between AI and APIs.

⚡

Instant Deploy

Push to deploy in seconds. No containers, no VMs, no infrastructure to manage.

🚀

Zero Cold Start

Workers start in under 1ms. No boot time, no warming, always ready.

Architecture

Every request flows through the same simple pipeline.

Client

AI Assistant

Claude, ChatGPT, Cursor

→

MCP Server

Cloudflare Worker

Stateless proxy

→

Service

External API

Jira, GitLab, Datadog

Tech Stack

Every server shares the same battle-tested stack.

📝

TypeScript

Strict mode, zero runtime dependencies (except MCP SDK). Full type safety from schema to response.

☁️

Cloudflare Workers

V8 isolate runtime. 300+ edge locations, sub-millisecond cold starts, 100k+ requests/day free.

🔗

MCP SDK

Official @modelcontextprotocol/sdk with Streamable HTTP transport via Cloudflare Agents.

📦

Bun

Fast package manager and test runner. Replaces npm for installs and vitest for testing.

🛡️

Zod

Runtime schema validation for all tool inputs. Type-safe parameters with zero boilerplate.

🔄

Stateless Design

No database, no sessions, no KV storage. Every request is independent. Zero data retention.

300+ data centers. Zero cold starts. Every MCP server runs at the edge.

mcp.flaremity.com · GitHub · CF Workers Docs