oas2mcp
=======
.. raw:: html
OpenAPI to agent-refined MCP
Convert OpenAPI and Swagger specifications into AI-refined FastMCP and LangGraph server surfaces.
.. grid:: 1 1 2 2
:gutter: 3
.. grid-item-card:: Quickstart
:link: quickstart
:link-type: doc
Install the package, point it at an OpenAPI spec, and generate output quickly.
.. grid-item-card:: API Reference
:link: autoapi/oas2mcp/index
:link-type: doc
Browse the full generated API reference for the package.
.. grid-item-card:: Design Notes
:link: architecture
:link-type: doc
Understand the conversion pipeline, models, and adapters.
.. grid-item-card:: Examples
:link: examples
:link-type: doc
See example specs, generated artifacts, and usage patterns.
.. note::
``oas2mcp`` is built for typed, inspectable OpenAPI-to-MCP workflows with deterministic parsing, focused AI agent refinement, FastMCP bootstrap artifacts, and LangGraph deployment wrappers.
Highlights
----------
.. grid:: 1 1 3 3
:gutter: 2
.. grid-item-card:: Typed models
:class-card: sd-shadow-sm
Clear Python types and structured models across the pipeline.
.. grid-item-card:: Source links
:class-card: sd-shadow-sm
Every documented symbol links back to repository source where possible.
.. grid-item-card:: Rich API docs
:class-card: sd-shadow-sm
Auto-generated reference docs powered by AutoAPI.
.. grid-item-card:: Agent system
:class-card: sd-shadow-sm
Catalog summarizer, operation enhancer, and shared surface planner stages.
.. grid-item-card:: Deployable outputs
:class-card: sd-shadow-sm
FastMCP bootstrap config, LangGraph entrypoints, and LangSmith tracing hooks.
.. toctree::
:maxdepth: 2
:hidden:
autoapi/oas2mcp/index
quickstart
architecture
examples
Indices and tables
==================
* :ref:`genindex`
* :ref:`modindex`
* :ref:`search`