MCP Tools DirectoryAbout this integration
Give it a store URL, get structured rows. A model writes the parser once; replays are free.
- Transport
- stdio
- Authentication
- api-key
- Initial setup
- credentials
- Runtime
- unattended
- Evidence
- documented
- Version
- 1.0.1
- Package
- scrapewright
- Last compatibility test
- Not independently tested
Connect your agent
scrapewrightPublisher README and executable manifest reviewed; package, target sites and model provider were not executed. Anti-bot walls are explicitly unsupported, and optional browser/model prerequisites remain caller responsibilities.
Capabilities: Storefront platform detection, Catalog and page extraction, Reusable selector synthesis, Optional browser-rendered crawling
Connected profiles
Additional details
io.github.Ozymandias-Owens-2/scrapewright
Source ↗ · Checked 2026-09-171.0.1
Source ↗ · Checked 2026-09-17CC0-1.0; package licenses are separate
Source ↗ · Checked 2026-09-17pypi
Source ↗ · Checked 2026-09-17scrapewright
Source ↗ · Checked 2026-09-191.0.1
Source ↗ · Checked 2026-09-19The documented MCP command exposes five tools to a client; known-platform and cached-recipe paths run deterministically, while custom sites may invoke the configured model provider.
Source ↗ · Checked 2026-09-19An Anthropic API key is needed only when an unknown site requires one-time recipe synthesis; supported deterministic storefront paths work without it.
Source ↗ · Checked 2026-09-19The client configuration launches the scrapewright CLI with the mcp subcommand through uvx.
Source ↗ · Checked 2026-09-19Publisher documentation reviewed; package and integration endpoint not executed or independently security-audited.
Source ↗ · Checked 2026-09-19