webindex

maxgfr/webindex

★ 0 stars TypeScript Browser/Web Updated today

Zero-dependency web-retrieval engine: turn a URL or a file into clean, citable text — HTML, PDFs through a six-rung ladder ending in OCR, office documents — and serve it over MCP. One vendorable engine.mjs, CLI + MCP server included. The web-side companion to codeindex.

View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "webindex": {
      "command": "npx",
      "args": [
        "-y",
        "maxgfr/webindex"
      ]
    }
  }
}

README Excerpt

Find pages with a local, keyless search stack, turn them into clean, citable text — HTML, PDFs through a six-rung ladder ending in OCR, and office documents — and serve that to an agent over MCP. Zero runtime dependencies. One ESM bundle plus one declaration file, plus a CLI. The web-side companion to [codeindex](https://github.com/maxgfr/codeindex): codeindex indexes

Tools (4)

readerwebindex_extractwebindex_fetchwebindex_search

Topics

ai-agentscontent-extractionesmfirecrawlocrpdf-extractiontext-extractiontypescriptvendoringweb-retrievalweb-scrapingzero-dependency

webindex

Quick Install

README Excerpt

Tools (4)

Topics

Related Browser/Web Servers