{"slug":"crawlbase","name":"Crawlbase MCP","provider_slug":"crawlbase","publisher":"official","publisher_name":"Crawlbase","kind":"provider-integration","description":"Retrieve HTML or Markdown, capture rendered pages, and revisit stored crawl results through Crawlbase tools.","tasks":["Read pages in a format suitable for the task.","Inspect JavaScript-rendered pages.","Retrieve previously stored HTML by request identifier."],"limitations":["The local profile needs separate Normal and JavaScript tokens for their respective operations.","Storage belongs to the token used for the crawl.","Documentation was reviewed; authenticated requests were not tested."],"capabilities":["scraping-extraction"],"connections":[{"id":"crawlbase-stdio","label":"Crawlbase local MCP","transport":"stdio","hosting":"local","endpoint":null,"package":"@crawlbase/mcp","auth":"Crawlbase normal token in CRAWLBASE_TOKEN and JavaScript token in CRAWLBASE_JS_TOKEN.","prerequisites":["Node.js 18 or later with npx.","Crawlbase Normal token; JavaScript token for rendering and screenshots."],"config_label":"Local stdio mcp.json — replace the credential placeholders","config":"{\n  \"mcpServers\": {\n    \"crawlbase-stdio\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"-y\",\n        \"@crawlbase/mcp\"\n      ],\n      \"env\": {\n        \"CRAWLBASE_TOKEN\": \"YOUR_CRAWLBASE_TOKEN\",\n        \"CRAWLBASE_JS_TOKEN\": \"YOUR_CRAWLBASE_JS_TOKEN\"\n      }\n    }\n  }\n}","source_ids":["crawlbase-docs","crawlbase-repo"]}],"tools":[{"name":"crawl","description":"Retrieve HTML for a URL.","task":"Inspect markup or save a page.","source_ids":["crawlbase-docs"]},{"name":"crawl_markdown","description":"Retrieve page content as Markdown.","task":"Read a page in the assistant.","source_ids":["crawlbase-docs"]},{"name":"crawl_screenshot","description":"Return a rendered PNG image.","task":"Check the visible page.","source_ids":["crawlbase-docs"]},{"name":"storage_get","description":"Retrieve a stored page by RID or URL.","task":"Reuse a recorded crawl.","source_ids":["crawlbase-docs"]},{"name":"storage_list","description":"List stored request identifiers with pagination.","task":"Locate saved results.","source_ids":["crawlbase-docs"]}],"commercial":{"account_required":true,"product":"Crawlbase Crawling API and Cloud Storage","billing":"Requests use the Crawlbase account credit balance. Charges depend on successful requests and selected features; storage and screenshots share the published rate card. MCP does not remove API usage or concurrency limits.","pricing_url":"https://crawlbase.com/pricing"},"sources":[{"id":"crawlbase-docs","label":"Crawlbase MCP tools and setup","url":"https://crawlbase.com/docs/ai-mcp","claims":["Package, tokens, tool names and screenshot storage behavior."],"checked_at":"2026-09-27"},{"id":"crawlbase-repo","label":"Official Crawlbase MCP repository","url":"https://github.com/crawlbase/crawlbase-mcp","claims":["Local stdio configuration and authentication."],"checked_at":"2026-09-27"},{"id":"crawlbase-storage","label":"Crawlbase Cloud Storage reference","url":"https://crawlbase.com/docs/cloud-storage","claims":["Persistence, token scope and retrieval formats."],"checked_at":"2026-09-27"},{"id":"crawlbase-pricing","label":"Crawlbase pricing","url":"https://crawlbase.com/pricing","claims":["Request credits and shared product billing."],"checked_at":"2026-09-27"},{"id":"crawlbase-package","label":"Crawlbase published MCP package metadata","url":"https://registry.npmjs.org/@crawlbase/mcp/latest","claims":["Package name, Node.js 18 runtime requirement and MIT license."],"checked_at":"2026-09-27"}],"checked_at":"2026-09-27","updated_at":"2026-09-27","review_due_at":"2026-10-27","verification":"docs-confirmed","test":null,"status":"published","withdrawal_reason":null,"repository_url":"https://github.com/crawlbase/crawlbase-mcp","license":"MIT","editorial":{"seo_title":"Crawlbase MCP Server: Crawl, Screenshot and Retrieve Pages","seo_description":"Set up Crawlbase MCP for Markdown, HTML and screenshots, then retrieve saved pages with the correct token. Includes storage and billing notes.","intro":"Crawlbase MCP fits investigations where fetching a page is only the first step. Its storage tools let an assistant return to saved crawl output. Keep the request identifier and token scope together so that later retrieval can find the same record.","selection":{"best_for":["Research that revisits saved page content.","Workflows comparing readable text with a rendered screenshot."],"not_for":["Expecting a screenshot file to be saved by the HTML storage option."],"source_ids":["crawlbase-docs","crawlbase-storage"]},"recommended_connection":{"connection_id":"crawlbase-stdio","reason":"The documented package exposes both token variables explicitly. This makes it easier to understand which credential supplies rendering and which storage collection a request uses.","source_ids":["crawlbase-repo"]},"workflow":{"title":"Fetch, inspect and reuse a page","steps":[{"title":"Select the output","body":"Use Markdown for reading or HTML for markup inspection. Enable storage when the page needs to be retrieved again.","tool_names":["crawl","crawl_markdown"],"source_ids":["crawlbase-docs"]},{"title":"Check rendering when needed","body":"Take a screenshot for a visual check. Preserve the returned request identifier and token_type with your notes.","tool_names":["crawl_screenshot"],"source_ids":["crawlbase-docs","crawlbase-storage"]},{"title":"Retrieve the saved version","body":"Locate the RID and fetch the stored record. Select the JavaScript token scope for a page originally captured with that token.","tool_names":["storage_list","storage_get"],"source_ids":["crawlbase-storage"]}]},"troubleshooting":[{"symptom":"A saved page cannot be found.","resolution":"Check the original token_type. Use use_js_token: true when retrieving from the JavaScript token storage silo.","source_ids":["crawlbase-docs","crawlbase-storage"]},{"symptom":"The local server does not start.","resolution":"Confirm Node.js meets the documented requirement and that the MCP client can launch npx with the configured environment.","source_ids":["crawlbase-repo"]}],"faqs":[{"question":"Does store: true save the screenshot image?","answer":"No. On screenshot calls it persists the rendered HTML. Save the returned PNG separately if the image is needed later.","source_ids":["crawlbase-docs"]},{"question":"Can I retrieve stored output without crawling again?","answer":"Yes. The storage retrieval tools address an existing record by its RID or URL, subject to the account storage settings.","source_ids":["crawlbase-storage"]},{"question":"Is the MCP package a separate unlimited scraping plan?","answer":"No. Calls consume the same Crawlbase products and balance. Check the current rate card and account limits before scaling a job.","source_ids":["crawlbase-pricing"]}]},"url":"https://workingproxysites.info/mcp/crawlbase","provider_url":"https://workingproxysites.info/t/80-crawlbase-proxies","review_due":false,"version":null}