Aurekai integration surface for DVC with reproducible artifact pipelines for model memory, manifests, SAE audits, semantic cache benchmarks, proof bundles, and release gates.
Status: active Type: data-ml
aurekai-dvc provides a complete dvc.yaml stage graph and executable scripts to produce verifiable artifact outputs in artifacts directory. It supports local script validation even when dvc is not installed.
- doctor-deep
- manifest-verify
- model-memory-pack
- sae-audit
- semantic-cache-bench
- proof-bundle-export
- release-gate
python3 -m pip install -r requirements.txt bash tests/validate-scripts.sh
If dvc is installed:
dvc repro
- Platform: https://github.com/aurekai/aurekai
- Native runtime: https://github.com/aurekai/native-runtime
- Integration registry: https://github.com/aurekai/aurekai/blob/main/registry/integrations.json
- Ecosystem map: https://github.com/aurekai/aurekai/blob/main/ECOSYSTEM_NAMES.md