Where developers grow together

doc2mcp

I built doc2mcp because AI coding agents are only as good as the documentation they can actually read.Today the workflow is broken 🤔 Your agent guesses an API, hallucinates a method that doesn't exist, or leans on stale training data. So you end up copy-pasting docs into context by hand, page by page, every single time.That's the gap doc2mcp closes ✅doc2mcp turns any documentation URL into a hosted, token-secured MCP server in under 60 seconds 🚀Paste a docs link → we crawl it (Mintlify, Docusaurus, Swagger, GitBook, plain HTML — all supported), preserve code blocks, run an understanding pass that detects auth patterns, endpoints, and workflows, and generate semantic MCP tools your agent can call directly.Then install it into Cursor, Claude, VS Code, or Windsurf with one command 🛠️ Your agent now answers from the real docs — with citations — instead of guessing.1. Crawls the docs and indexes every page2. Generates typed MCP tools (search, read, ask) automatically3. Hosts a token-secured MCP endpoint your agents can call4. One-click installs it into Cursor, Claude, VS Code & WindsurfIn the demo MCP, 40 pages became 3 tools at 97% confidence — no glue code.The core tool is ask_documentation: natural-language Q&A grounded in the crawled docs, with no API key needed from your users 🔐It's built for developers shipping with agent frameworks, teams whose users constantly ask "how do I use your API?", and anyone tired of their agent confidently inventing endpoints.It works two ways:• Web app → https://doc2mcp.site• CLI → npm i -g doc2mcp (convert, install, and chat with your docs from the shell): https://doc2mcp.site/cliIt's built for two people: developers who want to generate and self-host MCPs, and companies that want to publish official MCP infrastructure for their customers and AI agents (custom domains + company-attributed usage).Docs: https://doc2mcp.site/docsWould love your feedback — especially which docs site you'd want as an MCP first. I'll be here all day answering everything 🙏

AiraBreeze Mini Portable Cooler

AiraBreeze Mini Portable Cooler is a compact evaporative air cooler designed for personal use. Unlike traditional air conditioners that cool entire rooms using refrigerants and compressors, this portable device uses water-assisted evaporative cooling to deliver a refreshing airflow directly around the user.

Sanjeevani AI

Sanjeevani is an AI-driven healthcare triage platform designed to help patients quickly assess the urgency of their symptoms and connect with appropriate care. Built to bridge the gap between patients and doctors in real time, especially useful in resource-constrained or high-volume healthcare settings. Key Features: Bilingual voice input — supports Hindi and English, making it accessible to a wider range of patients including those less comfortable with typed English LLM-powered triage engine — uses a structured triage prompt to assess symptom severity and route patients accordingly Real-time doctor alerts — WebSocket architecture pushes urgent cases to doctors instantly, with queue replay-on-connect so no alerts are missed if a doctor reconnects Full-stack implementation — React frontend (with Web Speech API for voice) + FastAPI backend