Connect OpenGraph.io to Cursor so your AI assistant can preview URLs, extract clean content, capture webpage context, and inspect metadata directly inside your coding workflow. One click to install.
Cursor's AI agent becomes dramatically more useful when it can work with URLs. These tools are discovered automatically — just ask in natural language.
See a page's title, description, image, favicon, and Open Graph metadata — debug your own site's link previews or check a competitor's without leaving your editor.
Convert any URL to clean, readable Markdown with boilerplate stripped. Pull API docs or articles into your project as context — no HTML parsing required.
Take full-page or viewport-sized screenshots of any webpage. Compare competitor UIs, verify deployments, or generate visual documentation inline.
Ask the agent to check whether a page has complete Open Graph and Twitter Card tags before you ship it or share it.
Run an instant quality check on a single URL, or kick off a full site audit, and get back a prioritized list of SEO and social-sharing fixes.
Ask the agent to create an architecture diagram, icon, or social card from a prompt, then export it straight into your project.
Click Add to Cursor above for the fastest path, or paste the config below yourself — hosted is selected by default.
~/.cursor/mcp.json for global access or .cursor/mcp.json in your project root.{ "mcpServers": { "opengraph": { "url": "https://mcp.opengraph.io/mcp" } } }
Works in either global or project mcp.json. Sign in when Cursor opens the browser tab.
Once the tools appear, paste this into Cursor's chat to confirm everything is wired up correctly.
Use OpenGraph.io to preview this URL and summarize the title, description, image, and metadata.Cursor supports two config locations. Choose based on whether you need tools everywhere or just for a specific project.
The AI agent automatically picks the right MCP tool based on your request. These prompts work in Cursor's chat, Composer, or inline editing.
Scrape or convert documentation pages to Markdown and have the agent summarize endpoints, parameters, and code examples inline.
Screenshot competitor pages, analyze their design patterns, and generate implementation code — all in one flow.
Run a pre-launch audit and validate OG tags and social sharing previews without switching to a browser.
The MCP server runs locally as a subprocess of Cursor. Your App ID stays in your config file and is never sent to the AI model.
Your App ID is stored in mcp.json on your machine — never exposed to the AI model or transmitted externally.
The MCP server runs as a child process of Cursor on your machine. No cloud hosting, no shared infrastructure.
Communication uses standard input/output between Cursor and the MCP server. No open ports, no network exposure.
The server doesn't cache responses locally or maintain session state. Every request is stateless and ephemeral.
Most issues are solved by one of these steps. If you're still stuck, check the full troubleshooting guide.
node --version in your terminal. Restart Cursor after updating your PATH.Cmd+Shift+U) and select “MCP Logs” from the dropdown for detailed error messages.The same opengraph-io-mcp package powers every client below. Only the config file format differs.
Browse every capability across Web Data, Site Audit, and Image Generation.
Copy-paste prompts for previewing, extracting, capturing, and auditing URLs.
The full overview of OpenGraph.io's MCP server, security model, and client support.
Full setup documentation and configuration reference for every client.
Over 1B URLs processed
Powering previews, metadata, screenshots, extraction, embeds, and audits.
