summarize-mcp-worker

# summarize-mcp-worker A Cloudflare Worker exposing an MCP server for web page extraction and lightweight extractive summarization — no external model dependencies. ## MCP Tools | Tool | Description | |------|-------------| | `fetch_url` | Fetch a URL and return raw response metadata plus a cleaned text preview. | | `extract_clean` | Fetch a web page and extract the main text, title, headings, and readable paragraphs. Strips ads, nav, footers, and boilerplate. | | `summarize` | Generate an extractive summary from a URL or raw text. Picks the most representative sentences based on keyword frequency — no LLM needed. | | `extract_metadata` | Extract page metadata: title, description, canonical URL, byline, site name, language, and published time. | | `health` | Worker health check. | ## How It Works Content extraction uses a custom readability-style parser that runs entirely in the Worker. Summaries are extractive (sentence selection by keyword scoring), not abstractive — fast, deterministic, and no API calls. The MCP endpoint follows standard JSON-RPC (`/mcp`). ## Local Development ```bash npm install npx wrangler dev --local --port 8792 ``` ## Deploy ```bash npx wrangler deploy ``` ## Project Structure ``` summarize-mcp-worker/ ├── src/index.js # Worker entry + extraction + summarization + MCP handlers ├── wrangler.toml ├── package.json └── README.md ``` ## License This project is licensed under the GNU General Public License v3.0 — see the [LICENSE](LICENSE) file for details.