Quick start#
Launch PyFluent-MCP#
The simplest way to start the MCP server is the ansys-fluent-mcp console script or
the ansys.fluent.mcp module:
ansys-fluent-mcp
# equivalent
python -m ansys.fluent.mcp
It launches the server over STDIO (the default MCP transport) and waits for connections from MCP clients.
To run over streamable HTTP instead:
ansys-fluent-mcp --transport http --host 127.0.0.1 --port 8000
Connect to your IDE or client#
PyFluent-MCP works with multiple MCP-compatible clients. For setup information, see IDE and client configuration.
Claude Code (recommended for AI-assisted development)
Visual Studio Code with Copilot (for Visual Studio Code users)
Claude Desktop (macOS app)
Cursor and other MCP-compatible clients
Follow the basic workflow#
Connect to Fluent#
There are two ways to connect to Fluent once the MCP server is running.
Option 1: Launch a new Fluent session (recommended).
Ask your AI assistant to use the connect tool:
“Connect to Fluent and launch a new solver session.”
This starts a new Fluent process through PyFluent and connects to it automatically.
Option 2: Attach to an existing instance.
Ask your AI assistant to use the connect tool with IP and port:
“Connect to Fluent on localhost port 18500.”
This option is useful when Fluent is already running on a remote machine.
Inspect, generate, and execute#
Once Fluent is connected, the recommended loop for most setup tasks is:
Discover: Use
find_api,get_state, orlist_named_objectsto inspect the live settings tree.Validate: Use
validate_codeto pre-check Python snippets against the AST sandbox.Execute: Use
run_codeto apply the change to the live solver.Verify: Use
summarize_setuporsimulation_reportto confirm the result.
Use offline-only tools#
You can use several tools without a live Fluent session:
find_apisearches the bundled settings schema.get_helpreturns per-path help text.validate_codeperforms an AST pre-check only.
Consider example use cases#
Set up boundary conditions and solver settings with AI guidance.
Inspect mesh quality and cell counts on a loaded case.
Compare two case files to see what changed between versions.
Generate and debug PyFluent settings-API scripts interactively.
Next steps#
For an overview of available tools, see Overview.
For additional API reference, see Tools and capabilities.
For practical examples, browse Examples.
For configuration options, see Configuration.