This community Workers MCP demo exposes image generation, page screenshots and image extraction through a local client adapter and Cloudflare backend.
This community Workers MCP demo exposes image generation, page screenshots and image extraction through a local client adapter and Cloudflare backend.
The Worker class implements image generation with FLUX.1 Schnell, page screenshots and extraction of image URLs from a page. Generated images and screenshots are stored in R2 and returned as URLs. The class defines these three methods; a separate README example about random numbers is not an implemented tool in this Worker.
The project requires Cloudflare Workers AI, R2 and Browser Rendering bindings for the respective tools. Configure the bucket's public delivery URL through BUCKET_CDN so returned image links can be opened. Use the project's dependencies and deployment instructions, then check the current service allowances for your account rather than assuming the demo runs at no cost.
The repository uses the workers-mcp adapter and provides a Claude installation command after deployment. The client connection launches a local bridge to the Worker backend; the source does not establish a public, universal Streamable HTTP MCP endpoint. Wrangler is the deployment tool, not the server package to install into an MCP client.
Try a screenshot of a test page you control and verify that the returned R2 URL is accessible. For generated images, compare the saved result with the requested prompt. Image extraction returns URLs found on the page, not copies of every image. Browser access and storage settings remain part of the deployment, and this community demo should be evaluated in its own environment before operational use.
Check the linked publisher documentation and your connected client's tool list. Names and availability can vary with the server version and account permissions.
| Tool | What it does |
|---|---|
| generateImage | Generate an image with Workers AI and store the result in R2. |
| screenshotURL | Capture a page screenshot and store it in R2. |
| extractImagesFromURL | Return image URLs extracted from a page. |
pnpx workers-mcp install:claudeFollow the repository setup and deploy your Worker first. Configure AI, R2_BUCKET, BROWSER and BUCKET_CDN on the backend; review the generated client configuration.
A Cloudflare account, deployed Worker, workers-mcp adapter, AI/R2/Browser bindings for the selected tools, and a configured BUCKET_CDN URL.
Enables AI assistants to control Spotify playback and manage playlists via the Model Context Protocol.
Allows AI assistants to query movie and TV show collections managed by Radarr and Sonarr.
Analyzes YouTube videos by extracting transcripts, summarizing content, and answering questions using AI.
Retrieves YouTube video URLs from a specified Watch Later playlist within a given timeframe.
Provides QQ Music search functionality via the Modular Control Protocol (MCP).
Extracts subtitles from YouTube videos using a provided link.