Release notes#
This section contains the release notes for PyFluent-MCP.
0.5.0 - September 25, 2026#
Update flit-core requirement from <4,>=3.2 to >=3.2,<5 in the pip-deps group |
|
Bump the actions group with 15 updates |
|
Bump the pre-commit group across 1 directory with 2 updates |
|
Bump the pip-deps group with 2 updates |
|
Bump astral-sh/ruff-pre-commit from v0.16.3 to 0.16.4 in the pre-commit group |
|
Bump pydantic from 2.13.4 to 2.13.5 in the pip-deps group |
|
Bump the pre-commit group with 2 updates |
|
Bump the pip-deps group with 4 updates |
|
Bump sphinx-autodoc-typehints from 3.13.5 to 3.13.6 in the pip-deps group |
|
Bump codecov/codecov-action from 7.0.0 to 7.1.0 in the actions group |
0.4.0 - August 14, 2026#
Integrate meshing workflow |
0.3.0 - August 04, 2026#
0.2.1 - July 31, 2026#
0.2.0 - July 20, 2026#
Guard _import_litellm against the tokenizers |
|
Unified LLM provider configuration, transport diagnostics, and validator/active-path hardening |
|
Python version support according to PyAnsys guidelines |
|
Skip allowed_values() on inactive command arguments to avoid Scheme "object is not active" transcript spam |
|
Slim the Fluent MCP leaf - remove in-server LLM codegen, normalize the tool surface |
Update CHANGELOG for v0.1.0 |
0.1.0 - July 10, 2026#
Guard _import_litellm against the tokenizers |
|
Raise unit-test coverage above 80%, expand the live-Fluent integration tests, and remove bandit skips as part of the technical review. |
|
Add numpy-style docstrings across the public API so the autoapi reference documentation renders complete signatures and parameter descriptions. |
|
Add Anthropic message normalization for LiteLLM transport |
|
Probe APIs for Settings APIs |
|
Deterministic path-discovery + write-target resolver for the Solve leaf |
Update and align Ansys action version to v10.3.3 |
0.1.0 - Initial release#
Initial release of the standalone Fluent Solve MCP server plus the shared
|
|
22-tool MCP surface: connection, schema discovery, named objects, codegen, execution, reporting, and domain tools. |
|
PyFluent backend with pluggable backend entry-point seam. |
|
Offline-first settings schema catalog and LLM codegen pipeline. |
|
Model- and provider-agnostic LLM transport via |
TLS certificate verification enabled by default for outbound LLM and retrieval calls. |
|
Network retrievers honor |