Skip to content

Latest commit

 

History

History
70 lines (47 loc) · 1.63 KB

File metadata and controls

70 lines (47 loc) · 1.63 KB

Public Artifact Publication (No AWS Access Needed)

This guide publishes benchmark evidence so readers can verify paper claims without any AWS credentials.

Why

  • s3://... links usually require IAM access
  • Public release assets are easier for reviewers and readers
  • Checksums make integrity verifiable

1) Prepare a safe public bundle

./scripts/prepare_public_artifact.sh \
  --input-dir path/to/generated/kms_validation_<timestamp> \
  --name kms_validation_<timestamp>.tar.gz

Outputs in artifacts/public/:

  • kms_validation_<timestamp>.tar.gz
  • kms_validation_<timestamp>.tar.gz.sha256
  • kms_validation_<timestamp>.tar.gz.manifest.txt

The script scans for common sensitive markers (ARNs, instance IDs, access keys, IPv4, account-ID contexts) and fails by default if found.

2) Upload to GitHub Release

Authenticate once:

gh auth login -h github.com

Upload to an existing tag (example v1.0.0):

./scripts/publish_public_artifact.sh \
  --tag v1.0.0 \
  --artifact artifacts/public/kms_validation_<timestamp>.tar.gz

3) What to place in the paper

  • Release page URL (stable): https://github.com/cyntrisec/EphemeralML/releases/tag/v1.0.0
  • Artifact file name: kms_validation_<timestamp>.tar.gz
  • SHA-256 checksum: <sha256-from-generated-artifact>

4) Reader verification

After download:

sha256sum -c kms_validation_<timestamp>.tar.gz.sha256

Expected output:

  • ...: OK

Optional: Zenodo DOI

If you need archival citation quality:

  1. Connect the GitHub repo to Zenodo.
  2. Archive the release containing the artifact.
  3. Add DOI to paper provenance and README.