slack-mcp-server

Cloudflare Browser Toolkit

Use Cloudflare Browser Rendering from this repo to run resilient page checks and captures when local browser automation is blocked.

Prerequisites

• CLOUDFLARE_ACCOUNT_ID set
• CLOUDFLARE_API_TOKEN or CF_TERRAFORM_TOKEN set
• Token needs Browser Rendering access
• Account API tokens are supported (recommended for this workflow)

Verify Access

npm run cf:browser -- verify

The verifier checks account-token auth (/accounts/{account_id}/tokens/verify) first and falls back to user-token verification automatically.

Quick Commands

Rendered HTML:

npm run cf:browser -- content "https://example.com"

Markdown extract:

npm run cf:browser -- markdown "https://example.com"

Link extraction:

npm run cf:browser -- links "https://example.com"

Selector scrape:

npm run cf:browser -- scrape "https://example.com" --selectors "h1,.card a"

Screenshot:

npm run cf:browser -- screenshot "https://example.com" --out ./tmp/example.png --type png

PDF:

npm run cf:browser -- pdf "https://example.com" --out ./tmp/example.pdf

Structured JSON:

npm run cf:browser -- json "https://example.com" --schema '{"title":"string","links":["string"]}'

Notes

• This toolkit is for page rendering, extraction, and evidence capture.
• It does not manage third-party account logins for posting workflows.

This site is open source. Improve this page.