Bottom line: Routing24 exposes its state and actions via a JavaScript-based interface, allowing Claude Cowork and future WebMCP-compatible agents to use route optimization as an integrated tool.
Routing24, a free route optimization application, can now be controlled via Claude Cowork as a WebMCP agent. This enables AI assistants to perform complex tasks such as data import, geocoding, and optimization explanations directly within the agent interface.
Routing24.com provides companies with free route planning and optimization features. Previously, browser agents could not interact directly with the application, which proved problematic for several typical AI tasks: data preparation from CSV and Excel, validation of geocoding quality with feedback to users, support for the steep learning curve for business users, and explainability in route optimization decisions.
The solution was to make Routing24’s state model and actions available via a `window` object for JavaScript-based agents. This was initially developed for integration testing with Chrome DevTools Protocol, but then proved perfectly suited for Claude Cowork with JavaScript tool and browser extension: data import, geocoding, optimization, and explanation can now be controlled directly via agent prompts.
This approach has been standardized as W3C WebMCP (Web Model Context Protocol, currently in Chrome Preview). This makes Routing24 compatible with Claude Cowork now and will allow it to work with any WebMCP-compatible agent in the future.
Source: github.com · Published July 4, 2026
Lumi AI News — AI-assisted curation pursuant to Art. 50 EU AI Act. Paraphrase and classification by Lumi News Pipeline v1.7.3.