EclipseWatcher

Overview

EclipseWatcher is a lightweight MCP (Model Context Protocol) server that turns an LLM agent into a real‑time monitor for the latest Berserk manga releases. By combining a Cloudflare‑aware scraper with an automatically described MCP API, it lets any Claude or Qwen‑powered assistant answer questions such as “What’s the newest chapter of Berserk?” without manual intervention. The server continually polls readberserk.com for new chapter listings and exposes two intuitive tools— and —so that agents can fetch the current chapter or check if a previously seen chapter is still the latest.

For developers, EclipseWatcher removes the need to write custom web‑scraping logic or manage authentication headers for Cloudflare. The built‑in wrapper handles the anti‑bot checks automatically, while BeautifulSoup parses the HTML into clean data structures. The FastMCP server then serialises these functions with clear descriptions, making them immediately usable by any function‑calling agent. This tight integration means a developer can spin up the MCP stack, launch an LLM server (e.g., vLLM), and start querying for new chapters with a single line of code.

Key capabilities include:

• Real‑time monitoring: The scraper runs on a configurable schedule, ensuring agents receive up‑to‑date information as soon as a new chapter appears.
• Agent readiness: The MCP API is auto‑described, allowing agents to discover and invoke the tools without manual prompt engineering.
• Cloudflare resilience: By leveraging , the server navigates Cloudflare protection seamlessly, avoiding downtime that would otherwise break monitoring.
• Extensibility: The design encourages adding new scraping targets or additional tools; developers can expose any web resource through the same MCP framework.

Typical use cases span fan communities, content aggregators, and news bots. A fan‑facing chatbot can instantly notify users when a new chapter is released, while an RSS feed generator could pull the latest titles into a website. In research or monitoring pipelines, EclipseWatcher can serve as a lightweight data source that feeds into larger analytics workflows.

In summary, EclipseWatcher provides a plug‑and‑play MCP solution that transforms static web pages into dynamic, agent‑driven data sources. Its combination of Cloudflare‑aware scraping, auto‑described tools, and seamless integration with LLM runtimes makes it a standout choice for developers looking to add real‑time manga updates—or any similar web content—to their AI assistants.