The Dappier MCP Server is an implementation of the MCP (Model Context Protocol) server provided by Dappier, with the goal of enabling AI applications to quickly and freely access real-time web search and obtain high-quality data from trusted media brands, including news, financial markets, sports, entertainment, weather, and more, thereby building powerful AI agents.
Main capabilities
-
Real-time web search. Use Dappier's AI index to search the real-time web, with model ID am_01j06ytn18ejftedz6dyhz2b15. You can obtain real-time data such as global breaking news, weather forecasts and local updates, travel alerts and flight information, trending topics and viral content, online deals and shopping highlights, suitable for scenarios such as news agents, travel planning, and reminder bots.
-
Stock market insights. The model ID is am_01j749h8pbf7ns8r1bq9s2evrh, providing instant market data, financial headlines, and trading insights, suitable for portfolio dashboards, trading copilots, and investment tools. You can obtain real-time stock prices, financial news and company updates, trading signals and trends, market movement summaries, and AI analysis based on real-time data from Polygon.io.
-
AI content recommendations. Provides multiple AI models for specific domains, used for content discovery, summarization, and subscription feed generation. Including: Sports (real-time headlines, game recaps, expert analysis, model ID dm_01j0pb465keqmatq9k83dthx34), Lifestyle (health, entertainment, and daily inspiration, model ID dm_01j0q82s4bfjmsqkhs3ywm3x6y), Dog Care Assistant (training tips, health advice, behavior insights, model ID dm_01j1sz8t3qe6v9g8ad102kvmqn), Cat Expert (nutrition, play, grooming care, model ID dm_01j1sza0h7ekhaecys2p3y0vmj), Sustainable Lifestyle (environmental concepts, ethical choices, and green innovation, model ID dm_01j5xy9w5sf49bm6b1prm80m27), Hyperlocal News (local politics, culture, health, and multicultural updates, model ID dm_01jagy9nqaeer9hxx8z1sk1jx6). Each recommendation includes a title and summary, original publication date, trusted source and domain, image preview (if available), and a relevance score used for ranking. Advanced options can adjustthe entire search algorithm (semantic, most_recent, trending, etc.), focus results on a specific domain (ref), and adjust the number of results (similarity_top_k, num_articles_ref).
How to use
First, go to Dappier to register and generate an API Key. Then you can automatically install dappier-mcp to Claude Desktop via Smithery (uv must be installed first), or manually edit the configuration files of each client: Claude Desktop uses claude_desktop_config.json (the macOS path is ~/Library/Application Support/Claude/claude_desktop_config.json, and Windows is %APPDATA%\Claude\claude_desktop_config.json), and it can also be opened in the app via Claude > Settings > Developer > Edit Config; Cursor uses mcp.json (the global configuration on macOS is ~/.cursor/mcp.json, and on Windows is %USERPROFILE%.cursor\mcp.json; project-level configuration is placed in <project-root>/.cursor/mcp.json), and it can also be added via Settings > MCP > Add New Global MCP Server; Windsurf uses mcp_config.json (on macOS ~/.codeium/windsurf/mcp_config.json, and on Windows %USERPROFILE%.codeium\windsurf\mcp_config.json), edited via Settings > Cascade > Model Context Protocol (MCP) Servers > View raw config; after saving, you need to click Refresh in the MCP Servers section for the changes to take effect. In the configuration, you may need to enter the full path to the uvx executable (macOS/Lon Linux use which uvx, on Windows use where uvx to obtain it). In addition, the MCP inspector can be used to debug the server.
Use cases and limitations
This server works out of the box, requiring no training or fine-tuning, and can be used to build real-time, interactive tools such as news agents, travel planners, reminder bots, portfolio dashboards, trading copilots, and recommendation and subscription feed applications based on domain-specific content. When using it, you need to hold a Dappier API Key; the availability and specific capabilities of each data model are subject to the descriptions in marketplace.dappier.com. The project welcomes community contributions; you can refer to other MCP server examples and implementation patterns in modelcontextprotocol/servers to submit a Pull Request.
