Comie.dev

Product Introduction

1. Definition: Comie.dev is a production context layer and MCP (Model Context Protocol) server for AI-powered coding agents. It is a developer tool that bridges the gap between AI coding assistants and live production environments.
2. Core Value Proposition: It exists to give AI coding agents like Cursor, Claude Code, and Codex real-time, read-only access to production data, enabling them to debug, analyze, and reason about code with actual system context, thereby solving the "AI blindness to production" problem.

Main Features

1. Universal MCP Integration: Comie.dev functions as a globally installed MCP server. It works seamlessly with any MCP-compatible AI coding environment without requiring specific IDE plugins. You install it once via a terminal command, and it becomes available across your entire AI coding workflow (Claude Code, Cursor, etc.).
2. Multi-Stack Production Connectors: The platform offers secure, read-only integrations with a wide array of production tools across categories. This includes databases (PostgreSQL, MongoDB, Redis, Supabase, Convex), observability platforms (Sentry, Datadog), analytics tools (PostHog), and infrastructure providers (Railway). It acts as a unified query layer for these disparate systems.
3. Scoped, Secure Access Setup: Comie.dev emphasizes security and simplicity in setup. Users generate scoped, read-only API tokens or credentials for their production tools. Comie uses these tokens to retrieve context on-demand for the AI agent, ensuring no permanent storage of raw production data and adhering to the principle of least privilege.

Problems Solved

1. Pain Point: AI coding agents operate in a vacuum, only aware of static codebases and lacking context from live systems. This leads to suggestions that may not account for real database schemas, live error rates, user behavior patterns, or infrastructure constraints.
2. Target Audience: The primary users are software engineers, DevOps professionals, and development teams who actively use AI-assisted coding tools (Cursor, Claude Code, etc.) and need to debug production issues, validate feature implementations against real data, or understand system behavior during development.
3. Use Cases: Essential scenarios include: debugging a production incident by letting the AI inspect related logs, traces, and database state; validating a database query change by having the AI analyze the current schema and sample data; understanding user impact by correlating code changes with analytics events from PostHog or Sentry.

Unique Advantages

1. Differentiation: Unlike traditional observability or database GUI tools built for humans, Comie.dev is built specifically as a context provider for AI agents. It does not replace tools like Sentry or Datadog but sits atop them, translating their data into context an AI can use. It is also not a standalone AI tool but an enhancer for existing agentic coding environments.
2. Key Innovation: Its core innovation is the application of the Model Context Protocol (MCP) to production system awareness. By packaging secure, read-only connections to multiple production systems as a standardized MCP server, it creates a pluggable "production context layer" that any compatible AI agent can immediately leverage, enabling a new category of production-aware coding assistance.

Frequently Asked Questions (FAQ)

1. Is Comie.dev secure for production data? Yes, Comie.dev is designed with security first: it operates in read-only mode by default, uses scoped access tokens you control, and does not permanently store your raw production data, acting only as a secure conduit for contextual information.
2. How does Comie.dev work with Cursor or Claude Code? Comie.dev installs as a local MCP server on your machine. Once configured with your production keys, AI agents in Cursor, Claude Code, and other MCP-compatible tools can request context through it, allowing them to pull in real-time logs, database schemas, and analytics data directly into your coding session.
3. What is the difference between Comie and an observability tool? Observability tools (Sentry, Datadog) are for human monitoring and alerting. Comie.dev is an AI context layer that allows coding agents to query and reason across data from your existing observability stack, databases, and analytics tools to assist with debugging and development tasks directly in your IDE.
4. Can Comie.dev write to my database or change production? No, Comie.dev is strictly read-only. Its integrations are configured to only permit data retrieval (e.g., SELECT queries, log fetching). It cannot execute INSERT, UPDATE, DELETE, or any administrative commands on your connected production systems.