As a company, Foretellix is particularly bullish about multi-modal, foundation-model-based V&V assistants to automate processes, doing a growing part of what a V&V engineer does. To help engineers move faster without disrupting their flow, Foretellix is now offering customers a new way to integrate their testing capabilities directly into their favorite AI developer assistants, such as Claude Desktop, Cursor and Forellix’s own Foretify AI-Assistant. This approach is made possible by the Model Context Protocol, or MCP, a new standard that connects AI clients and AI Agents to external tools in a secure and flexible way.
Instead of switching between environments, engineers can interact with Foretellix platforms such as Foretify and V-Suites using simple prompts inside the AI assistants they already trust. They can ask to search for relevant scenarios, launch specific tests, or analyze results using natural language.
The assistant handles the request by talking to the MCP server, which connects locally to Foretellix tools running on the customer’s infrastructure. Rather than replacing existing tools, this is a new way to access and extend them, built to fit naturally into how engineers already work.
Part of a Broader Shift Toward AI-Native Engineering
The MCP server reflects a broader evolution in how engineers interact with complex systems. As AI assistants become a natural part of the developer toolkit, teams are looking for ways to connect those assistants to the tools they rely on every day.
Foretellix is at the forefront of this shift. Rather than requiring users to learn a new interface or adapt to a new workflow, we bring testing capabilities into the environments engineers already prefer. This approach makes it easier to adopt AI without introducing risk, overhead, or unnecessary complexity.
The Model Context Protocol is quickly gaining traction across the development community as a standard for integrating AI with software tools. As more AI clients and agent frameworks adopt MCP, Foretellix tools will remain compatible and easy to integrate into a growing ecosystem of developer-focused solutions.
This feature represents a step toward a future where testing, analysis, and even scenario creation can happen in conversation with an intelligent assistant that understands your toolchain and respects your constraints.
What You Can Do With the MCP Bridge
With the MCP server in place, engineers can already perform a wide range of tasks through their AI assistant without leaving their development environment.
In the example below, a developer using their AI assistant of choice can ask it to find a scenario involving a vehicle cut-in with a pedestrian crossing. The assistant queries the Foretify scenario library, surfaces available options, and allows the user to select and launch a test run.
Once the test is complete, the assistant can retrieve the results and assist with post-run analysis. Engineers can ask questions like “What affected the time to collision?” or “Were there any collisions in this run?” The assistant pulls metrics, performs calculations, and even generates visualizations based on the data with no scripts or manual coding required.
This approach is especially useful for ad hoc analysis and triage. Traditional dashboards are often too rigid or narrowly scoped to answer the specific questions engineers have in the moment. There is no one-size-fits-all solution for analyzing simulation runs, and teams frequently need custom views or insights depending on what they are investigating.
With the MCP bridge, the AI assistant becomes an analysis co-pilot, suggesting how to approach the results, what metrics to examine, and how to visualize key findings. Engineers can quickly investigate anomalies, explore edge case behavior, and generate tailored dashboards in minutes, all within the same AI interface they already use for documentation and coding support.
By reducing the steps involved in testing and analysis, the MCP integration frees engineers to focus more on high-value problem-solving and less on managing tools. It allows them to work at a higher level of abstraction, where they can explore complex questions, experiment more freely, and act on ideas without needing to pause for tool configuration or manual scripting.
This also opens up new possibilities across the development lifecycle, empowering engineers to run simulations, analyze results, and create visualizations on their own before involving additional teams. It enables earlier exploration, faster iteration, and more confident decision-making at every stage of development.
Built to Grow with Your Workflow
The MCP server already enables a range of useful tasks, but its design allows for much more. Foretellix is continuing to expand the capabilities exposed through this integration to support deeper analysis and more advanced testing workflows.
Engineers will be able to perform coverage gap detection, generate complete test plans, and receive AI-assisted guidance on improving scenario coverage. Support for OpenSCENARIO 2.0 DSL (OSC 2.0) including the ability to explain or generate OSC code using natural language will also be possible.
As the interface matures, Foretellix’s own Foretify AI Assistant will be able to do more tasks and the MCP connection will allow increased access to Foretify’s exposed APIs for external AI assistants. This means engineers will eventually be able to query and visualize trends across runs, compare results over time, and automate larger testing pipelines all through the same assistant interface.
This sets a long-term foundation for enabling AI-augmented development within the AV testing stack, with practical use cases available today and a clear path to more advanced functionality.
Try the MCP Integration and Streamline Your AV Testing Workflow
The MCP server can enhance existing development and validation processes. This integration allows engineers to work faster, ask better questions, and extract more value from the tools they already use, all without compromising control over sensitive data.
If your team is using Claude Desktop, Cursor, or another MCP-compatible AI assistant and you’re interested in learning more or sharing your thoughts, then please contact us to stay in the loop.
Foretellix’s Foretify AI Assistant is available for beta testing and we are actively partnering with teams who want to push the boundaries of what AI-assisted testing can look like, contact us to learn more.
