# ScrapeMole Pkgs: npm and PyPI package metadata for AI agents

> Pay-per-call package registry lookups: npm package metadata, version details, search, and download counts; PyPI project metadata, releases, and name lookup. Public registry data only (no auth, no private packages, no scraping of user profiles). Emails and tokens are stripped. USDC via x402, no accounts.

Pay with x402 v2 (exact, USDC on Base). Call without payment to get a 402 with the price; retry with an x402 client. Only 200 responses are charged.

## Endpoints

- GET https://pkgs.scrapemole.workers.dev/v1/npm/package ($0.002): Latest metadata for an npm package: description, homepage, repository, license, keywords, maintainers (names only), version count, latest version, publish times, and last-week downloads.
  - name (required): npm package name, optionally scoped (@scope/name).
- GET https://pkgs.scrapemole.workers.dev/v1/npm/version ($0.002): One published npm version: dependencies, peerDependencies, engines, dist tarball URL and integrity, and published time.
  - name (required): npm package name.
  - version (required): Exact version or "latest".
- GET https://pkgs.scrapemole.workers.dev/v1/npm/search ($0.003): Full-text search of the npm registry: matching package name, version, description, keywords and relevance score (publisher emails omitted).
  - q (required): Search text.
  - limit: Max results (1-20). Default 10.
- GET https://pkgs.scrapemole.workers.dev/v1/pypi/project ($0.002): PyPI project info: summary, home page, project URLs, license, requires_python, classifiers subset, current version and yank status (author emails redacted).
  - name (required): PyPI project name.
- GET https://pkgs.scrapemole.workers.dev/v1/pypi/release ($0.002): Files for one PyPI release: filename, packagetype, size, upload time, yank status and digests.
  - name (required): PyPI project name.
  - version (required): Exact version or "latest".
- GET https://pkgs.scrapemole.workers.dev/v1/pypi/search ($0.003): Resolve PyPI project names: exact and PEP 503–normalized variants (hyphen/underscore/dot), returning matching project summaries. Full-text web search is not used (PyPI XML-RPC search is retired; the HTML search page is challenge-gated).
  - q (required): Project name or name-like query.
  - limit: Max results (1-20). Default 10.

## MCP

- Remote MCP server (Streamable HTTP): POST https://pkgs.scrapemole.workers.dev/mcp. Same tools and prices; pay with the x402 MCP transport.

## Docs

- OpenAPI: https://pkgs.scrapemole.workers.dev/openapi.json
- Terms: https://pkgs.scrapemole.workers.dev/terms
- Website: https://scrapemole.pages.dev

## Also from ScrapeMole

- ScrapeMole (YouTube search/channel/transcripts, Google Trends): https://scrapemole-api-mainnet.scrapemole.workers.dev
- ScrapeMole Amazon products: https://scrapemole-api.vercel.app
- ScrapeMole Reader (page to Markdown, link previews, RSS feeds, links, sitemaps): https://reader.scrapemole.workers.dev
- ScrapeMole Domains (DNS, RDAP/WHOIS, availability, email security, SSL certificates, subdomains): https://domains.scrapemole.workers.dev
- ScrapeMole Onchain (Base/Ethereum prices, DEX prices, ENS/Basenames, gas, tokens, balances, transactions): https://onchain.scrapemole.workers.dev
- ScrapeMole Companies (SEC EDGAR: company profiles, financials, filings): https://companies.scrapemole.workers.dev
- ScrapeMole Wiki (Wikipedia summaries and search, Wikidata facts; CC BY-SA attributed): https://wiki.scrapemole.workers.dev
- ScrapeMole Pulse (ECB FX rates and conversion, time zones, weather forecasts, Hacker News): https://pulse.scrapemole.workers.dev
- ScrapeMole Arxiv (arXiv paper search, metadata, abstracts, category listings): https://arxiv.scrapemole.workers.dev
- ScrapeMole Social (Bluesky search/profiles/feeds, Hacker News search and threads, Lobsters): https://social.scrapemole.workers.dev