Tools

The MCP Server provides a comprehensive set of tools that support every stage of the modeling workflow, from discovery and creation to solving and model management. These tools are designed to work together, allowing users to move seamlessly from one step to the next without leaving their development environment.

At a high level, the tools are organized into five categories:

  • Examples and Templates tools
  • Analysis tools
  • Solving tools
  • Diagnostics tools
  • Model Management tools

Examples and Templates Tools

Examples and Templates tools help users get started quickly. You can search a library of nearly 200 pre-built models, retrieve complete examples, or generate structured templates for new models. This makes it easy to either adapt an existing solution or begin building from a guided starting point.

  • search_examples — Search approximately 200 example models by keyword or type, including optimization, simulation, datamining/data science, calculation/decision tables, and flow/multi-stage pipelines.
  • get_example — Retrieve a specific example model with full definition and metadata.
  • get_model_template — Get a bare JSON scaffold for a model type, including blank, optimization, simulation, sim-optimization, data-science, workflow, decision-table, and power-bi.

Analysis Tools

Analysis tools provide insight into how models are constructed. They can identify the type of model, list variables and constraints, and highlight potential issues. This is especially useful when working with unfamiliar models or troubleshooting errors.

  • describe_model — Analyze model structure, including type and subtype detection, component inventory, and issue identification.
  • about_rason — Get information about RASON capabilities, model types, the Frontline Systems ecosystem, and server information.

Solving Tools

Solving tools allow you to execute models directly through the MCP Server. You can run models synchronously for quick results or submit longer-running jobs for asynchronous processing. Once complete, results can be retrieved and analyzed within your workflow.

  • solve_model — Solve a model synchronously using inline RASON, RASON files, or Excel workbooks.
  • submit_solve — Submit a model for asynchronous solving for long-running models.
  • get_solve_status — Check solving status and progress.
  • get_solve_result — Retrieve the result once solving is complete.
  • stop_solve — Stop a running solve. Intermediate results may be available.

Diagnostics Tools

Diagnostics tools help identify and resolve issues in models, particularly when optimization problems are infeasible or unbounded. These tools can suggest which constraints may need to be adjusted, making it easier to refine and improve model performance.

  • diagnose_model — Run diagnostics synchronously. This is useful for infeasible or unbounded optimization models.
  • submit_diagnose — Submit a model for asynchronous diagnostics.

Model Management Tools

Model Management tools support working with models in your RASON account. You can list, retrieve, create, update, and delete models, as well as manage versions and designate a champion model for use in production scenarios.

  • list_models — List models in your account with optional filters by type and kind.
  • get_model — Retrieve a model definition as RASON JSON.
  • post_model — Create a new model from a RASON JSON definition.
  • post_model_with_files — Upload a model with file attachments, such as RASON and data files, or Excel files.
  • put_model — Update an existing model or create a new version.
  • put_model_with_files — Update a model with file attachments.
  • set_champion — Set or unset the champion, or default, version of a model.
  • delete_model — Delete models with flexible filtering by kind, type, and force option.

Together, these tools transform the MCP Server from a simple execution engine into a full-featured modeling platform, enabling users to build, understand, solve, and manage models in a single, connected workflow.