Skip to main content

sysml2kit

CI PyPI Python License Docs

API-first Python tooling for building, querying, validating, and automating SysML v2 models.

Status: pre-alpha. The 0.4.x line covers the full loop: model, writer, interchange, queries, validation, diff, mermaid views, parse backend, MCP server, API client with a live-server harness, and multi-fidelity verification execution. The API may still move between minor versions; pin an exact version if you depend on it. Changes: CHANGELOG.md.

sysml2kit is the requirements/architecture/traceability layer for engineering automation stacks: build a system model in Python, emit standard SysML v2 textual notation and Systems Modeling API JSON, run traceability queries (which requirements are unsatisfied? unverified? allocated where?), validate, diff, and execute verification: analyses bound to registered engines run for real, and their metrics check the model's requirements. It targets the OMG SysML v2 standard, not any vendor tool. Docs: https://jman4162.github.io/sysml2kit/

What it does

  • Object model: a documented subset of SysML v2 (packages, parts, ports, attributes with units, requirements, satisfy/verify/derive/allocate, analysis cases) as pydantic models, with an opaque passthrough for elements outside the subset so richer models survive a round trip.
  • Textual notation writer: deterministic .sysml output. Parsing is delegated to a pluggable backend (pip install sysml2kit[parse] installs sysmlpy); the kit itself does not reimplement the grammar.
  • JSON interchange: read/write the Systems Modeling API serialization, the format the standard REST API speaks.
  • Traceability queries: unsatisfied/unverified requirements, allocation tables, requirement-to-part trace matrices.
  • Validation and diff: rule-based model checks and element-level diffs.
  • Verification execution: verificationBinding metadata binds an analysis case to an engine from the sysml2kit.engines entry-point group; sysml2kit verify runs it and checks each requirement with margins, and can write results back into the model with provenance. Sibling bindings labeled with fidelity/costSeconds form a fidelity ladder: --policy all reports the cross-rung spread as an error bar, --policy escalate --budget-s N spends a compute budget on the thinnest margins first, and every run records measured seconds per rung. The whole loop — bindings, verify links, policies — works from .sysml text as well as JSON.
  • Mermaid views: ownership-tree and requirement-trace diagrams.
  • MCP server: nine tools for agents (sysml2kit mcp serve).
  • API client: a thin HTTP client for the OMG Systems Modeling API and Services endpoints, plus a docker compose harness running the pilot implementation for live round-trip testing.

Install

pip install sysml2kit            # core: build, write, query, validate, diff
pip install "sysml2kit[parse]"   # + read .sysml files (sysmlpy backend)
pip install "sysml2kit[graph]"   # + NetworkX export
pip install "sysml2kit[mcp]"     # + MCP server for agents
pip install "sysml2kit[verify]"  # + YAML verification-binding configs

Quick start

from sysml2kit import Model, builder

model = Model()
pkg = builder.pkg(model, "Vehicle")
battery = builder.part(model, "battery", owner=pkg)
range_req = builder.req(
    model,
    "REQ-001",
    "Range",
    owner=pkg,
    text="The vehicle shall travel at least 400 km on one charge.",
)
builder.satisfy(model, source=battery, target=range_req)

from sysml2kit.text import write_model

print(write_model(model))  # standard SysML v2 textual notation

from sysml2kit.query import unverified_requirements

print(unverified_requirements(model))  # [REQ-001] — no verify link yet

Architecture

sysml2kit
├── model         # element classes, Model container, builder API
├── text          # SysML v2 textual notation writer
├── interchange   # Systems Modeling API JSON reader/writer
├── query         # traceability queries
├── validation    # rule-based checks (S2K001...)
├── diff          # element-level model diff
├── views         # mermaid diagrams (trace, tree)
├── verify        # verification bindings, engine registry, runner
├── api           # Systems Modeling API HTTP client
├── backends      # parser backends (sysmlpy behind the [parse] extra)
├── interop       # tool-agnostic requirement extraction
├── mcp           # MCP server (behind the [mcp] extra)
├── graph, units, workspace   # NetworkX export, pint helpers, path safety
└── cli           # `sysml2kit` command line

The spec pin, element subset, and known deviations are documented in SPEC.md. Reference spec release: OMG SysML-v2-Release tag 2026-05.

Domain content lives outside the kit. For antenna/RF systems engineering, see sysml2kit-rf-library, a SysML v2 model library consumed through this package. Downstream bridges are merged in phased-array-systems (interop.sysml: requirement sets and the phased-array-systems verification engine) and aedl (aedl.interop: bound-form requirements).

For agents

An MCP server ships behind the mcp extra with nine tools: model_show, model_validate, model_diff, model_export, model_diagram, requirements_trace, requirements_extract, requirements_verify, library_load. Artifacts are returned as file paths, not payloads.

pip install "sysml2kit[mcp,parse]"
sysml2kit mcp serve            # stdio; --transport http also supported
{"mcpServers": {"sysml2kit": {"command": "sysml2kit", "args": ["mcp", "serve"]}}}

The CLI covers the same operations for shell use: sysml2kit show | validate | diff | export | fmt | verify | api | mcp serve (export --to mermaid renders diagrams).

Development

uv sync --all-extras --group dev
uv run pytest                     # core suite (no extras needed)
uv run pytest -m parse            # round-trip tests against sysmlpy
uv run ruff check . && uv run ruff format --check .
uv run mypy
scripts/slopcheck.sh              # prose lint, advisory
uv run mkdocs serve               # docs preview

Citation

@software{hodge2026sysml2kit,
  author  = {Hodge, John},
  title   = {sysml2kit: API-first Python tooling for SysML v2 models},
  year    = {2026},
  url     = {https://github.com/jman4162/sysml2kit},
  license = {Apache-2.0}
}

License

Apache-2.0. See LICENSE and NOTICE. This project contains no code or model text from the EPL-2.0 OMG pilot implementation; conformance checks run it out of process only.

Metadata

Release files for sysml2kit 0.4.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for sysml2kit 0.4.0
File Size Uploaded Provenance
sysml2kit-0.4.0.tar.gz 229.3 kB With provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for sysml2kit 0.4.0
File Interpreter ABI Platform
sysml2kit-0.4.0-py3-none-any.whl Python 3 none any Details

Total release size: 303.5 kB

Release files / sysml2kit-0.4.0.tar.gz

Download URL sysml2kit-0.4.0.tar.gz
Size 229.3 kB
Tags Source
SHA-256 checksum
How to use checksums
26019265ae3461ea2ad94f3e1bf2f89c11e4af87971d834cb25bfa5a1e77123b
BLAKE2b-256 checksum
How to use checksums
fe3158ced041bd99997a252b3eb09bcd99ff0e7c4106c4a4660e86ecb15ddde5
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Aug 21, 2026.

Transparency log

Release files / sysml2kit-0.4.0-py3-none-any.whl

Download URL sysml2kit-0.4.0-py3-none-any.whl
Size 74.1 kB
Tags Python 3
SHA-256 checksum
How to use checksums
505c7f820b9aa67cab91ef3095c62001be5960719a3c6682cb41695d9fae49c1
BLAKE2b-256 checksum
How to use checksums
4d3bbb693dd9e439a6062b9a3361d1b097ea1ed5fb739c49f510919841b0e5bd
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Aug 21, 2026.

Transparency log

Security for sysml2kit 0.4.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

All 2 files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for sysml2kit-0.4.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Aug 21, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page