Skip to content

webx docs

The web toolkit for AI coding agents. Scrape, search, crawl, index, extract, and serve — one static binary, zero required services, MIT licensed.

One binary

go install github.com/kasyap1234/webx/cmd/webx@latest — no Node, no Python env, no required services. Runs on a laptop, a VPS, or air-gapped.

The full pipeline

Search → scrape → index → query → watch → verify → archive. Every stage writes to the same SQLite file it reads from.

Agent-native

17-tool MCP server, generated agent skills, --fit token-saver extraction, and chunk metadata with est_tokens.

Firecrawl-compatible

/v1 and /v2 routes accept Firecrawl request shapes — point their SDKs at your webx server and they keep working.