crml-lang

Language/spec package for CRML.

• Pydantic models for CRML documents
• Bundled JSON Schema + structured validator
• YAML load/dump helpers (CRScenario, CRPortfolio)

Quickstart

Validate a scenario document:

from crml_lang import validate

report = validate("examples/scenarios/data-breach-simple.yaml", source_kind="path")
print(report.ok)

Load and work with typed models:

from crml_lang import CRScenario

scenario = CRScenario.load_from_yaml("examples/scenarios/data-breach-simple.yaml")
print(scenario.meta.name)

Bundle a portfolio (inline referenced scenarios/packs into a self-contained artifact):

from crml_lang import bundle_portfolio

report = bundle_portfolio("examples/portfolios/portfolio.yaml", source_kind="path")
print(report.ok)

bundle = report.bundle
assert bundle is not None

Bundle from in-memory models (no filesystem access required):

from crml_lang import CRPortfolio, CRScenario, bundle_portfolio

portfolio = CRPortfolio.load_from_yaml("examples/portfolios/portfolio.yaml")
scenario = CRScenario.load_from_yaml("examples/scenarios/data-breach-simple.yaml")

report = bundle_portfolio(
	portfolio,
	source_kind="model",
	scenarios={"s1": scenario},
)
print(report.ok)

Plan a portfolio (deterministic engine pipeline step):

from crml_engine.pipeline import plan_portfolio

plan = plan_portfolio("examples/portfolios/portfolio.yaml", source_kind="path")
print(plan.ok)