On this page
0002 Gemara Supersedes OSCAL Input
Status: accepted
Date: 2026-02-01 (retroactive)
Deciders: complytime-dev team
Context#
The original compliance pipeline (complyscribe) used OSCAL as both the input authoring format and the runtime data model. This required compliance engineers to work directly with OSCAL’s verbose JSON/XML structures and created tight coupling between authoring tooling and the assessment runtime.
Gemara provides a CUE-based schema framework purpose-built for compliance content — simpler authoring, schema validation at the type level, and OCI-native distribution. OSCAL remains valuable as an output/interchange format but is not the right authoring surface.
Decision#
Gemara is the input model for compliance content authoring and policy definition. complyctl consumes Gemara artifacts (#Policy, #ControlCatalog) natively.
OSCAL is supported as an output format only — complyctl scan --format oscal produces OSCAL Assessment Results for downstream consumers that require it.
The OSCAL-input pipeline (complyscribe) is archived.
Consequences#
- Compliance engineers author in Gemara’s CUE-validated YAML — simpler than raw OSCAL.
- Schema evolution is managed via CUE module versioning, not OSCAL’s monolithic release cycle.
- Organizations with existing OSCAL tooling can still consume scan results in OSCAL format.
- The compliance-trestle dependency is eliminated from the runtime path.
- complyscribe is archived; its CaC-to-OSCAL sync capabilities are no longer maintained.
Source: complyctl PR #381, complyscribe PR #854