vouch-mlflow
Sign MLflow model artifacts with Vouch Credentials, so a registered model carries verifiable lineage: who registered it, on whose authority, and a content digest that breaks if the weights are tampered with.
This complements OpenSSF Model Signing. OMS proves an artifact is intact and signed by a key; Vouch adds the agent and delegation dimension, which principal or pipeline registered the model, traceable back to an accountable human.
Install
pip install vouch-mlflow
It has no hard dependency on MLflow. The helpers work on any file or directory path, so they fit any artifact store.
Sign at registration time
from vouch import Signer
from vouch_mlflow import sign_model
signer = Signer(private_key=PRIV_JWK, did="did:web:ml.acme.com")
# After mlflow.log_model(...) wrote the model to a local path:
credential = sign_model(signer, "runs:/abc/model_local_path", name="fraud-detector")
# Attach to the run so it travels with the model:
import mlflow, json
mlflow.set_tag("vouch_credential", json.dumps(credential, separators=(",", ":")))
Verify on load
from vouch_mlflow import verify_model
ok, passport = verify_model(model_path, credential, public_key=registrant_pubkey)
if not ok:
raise RuntimeError("Model signature invalid or weights changed since signing")
verify_model checks both the signature and that the on-disk content still
matches the digest the credential was bound to.
License
Apache-2.0.
Metadata
Release files for vouch-mlflow 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| vouch_mlflow-0.1.0.tar.gz | 3.1 kB | Without provenance | Details for vouch_mlflow-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| vouch_mlflow-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for vouch_mlflow-0.1.0-py3-none-any.whl |
Total release size: 5.7 kB
Release files / vouch_mlflow-0.1.0.tar.gz
| Download URL | vouch_mlflow-0.1.0.tar.gz |
|---|---|
| Size | 3.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
588672b0592a2d1d7d568eb160f766dc172fcc4b08602d5c1d4fa9d25e3851f2
|
|
BLAKE2b-256 checksum How to use checksums |
8b5e7070b49c46d26bd4dc1241f6f4e85a18b5b8335a5df6ced07b79bff7ae5b
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.9
|
Release files / vouch_mlflow-0.1.0-py3-none-any.whl
| Download URL | vouch_mlflow-0.1.0-py3-none-any.whl |
|---|---|
| Size | 2.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
94e6c142deb5a24a7b73e1559eef43692deb72bffb799683df66b87c7bb51eed
|
|
BLAKE2b-256 checksum How to use checksums |
892c0d2907bd46937d5e6e2f5031060159adf3150a77ea722eca562274c1ca25
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.9
|
Security for vouch-mlflow 0.1.0
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.
No provenance found
None of the files in this release were uploaded with provenance metadata.
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.