The latest articles on DEV Community by Mathis Van Eetvelde (@mathisve). https://dev.to/mathisve https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F4231472%2F17428fbf-139e-45e8-a34b-a3c77d4c3116.jpg https://dev.to/mathisve en Mathis Van Eetvelde Sat, 07 Nov 2026 04:45:05 +0000 https://dev.to/mathisve/build-a-company-research-agent-using-a-wikipedia-and-sec-mcp-server-30af https://dev.to/mathisve/build-a-company-research-agent-using-a-wikipedia-and-sec-mcp-server-30af <p>I recently built a company research agent in Langflow using a Wikipedia and SEC EDGAR MCP server (called Dataville).</p> <p>It's a chatbot so you ask it to research a company (Walmart for example), and it will use the available tools in the MCP to fetch context about the company on Wikipedia, and query the SEC EDGAR database for the latest filings (like the annual and quarterly reports). Then it compiles a polished report with citations and all!</p> <p>We even wrote a blog post about it that goes in more detail about the tool call JSON responses which you can read <a href="https://www.dataville.com/blog/build-a-langflow-ai-agent-with-mcp-sec-edgar-and-wikipedia" rel="noopener noreferrer">here</a>! If you'd rather just take a look at the repository containing the Langflow flow, prompts and sample reports you can find that <a href="https://github.com/datavilleorg/langflow-company-research" rel="noopener noreferrer">here</a>.</p> <p>Full disclosure: I worked with Dataville on this project. Happy to answer any questions!</p> ai mcp langchain