Skip to content

v0.1.0-rc1 — OpenAPI→MCP Problem-Semantics Preservation Profile

Pre-release
Pre-release

Choose a tag to compare

@moon10021017 moon10021017 released this 13 Aug 02:10
· 1 commit to main since this release
07bcc4d

OpenAPI→MCP Problem-Semantics Preservation Profile v0.1.0-rc1

This release candidate publishes a bounded, executable conformance profile for testing whether RFC 9457 Problem Details semantics remain machine-addressable across OpenAPI→MCP translation.

Headline result

Using the same frozen 12-case corpus and oracle:

Target / control Preserved Loss Invented N/A
FastMCP 3.4.7 0 11 0 1
CNOE openapi-mcp-codegen 0 11 0 1
nihal1294/openapi-to-mcp 0 11 0 1
Constructive reference over official MCP 2026-07-28 wire 11 0 0 1
RC1 fresh-install reproduction of constructive reference 11 0 0 1

The strongest supported comparative claim is:

For the frozen profile tested here, semantic loss is not forced by the tested MCP carrier capacity.

The artifact does not establish ecosystem-wide prevalence, complete RFC 9457 coverage, or a unique required MCP representation.

What is included

  • normative experimental preservation profile
  • frozen 12-case conformance corpus
  • carrier-neutral adapter and oracle
  • empirical evidence from three tested translation architectures
  • constructive preserving reference mapping
  • official MCP 2026-07-28 wire evidence
  • fresh-install clean-room reproduction
  • claim ledger, reviewer attack surface, and release verifier

Verification

python3 verify_release.py

Release artifact

Public-safe RC1 ZIP SHA-256:

f1a7dbddbeecb1cee1cdc84f34ff89b99c653267a10cfb51ebc8bfdf4df9b158

Status

Experimental release candidate. This is not an accepted MCP, IETF, or industry standard.