Skip to content
Briefcase AI
Docs for briefcase-ai v3.3.0 — see what’s new.

Artifact Graph Quick Start

Push your first tracked image in under five minutes. All you need is Docker (or crane) and a running oci-bai stack.

Prerequisites

  • Stack running — make up (gateway on :8080)
  • Docker or crane for pushing images
  • The oci-bai CLI on your PATH

Steps

  1. Push an image through the gateway

    Tag any image for the gateway and push. Derivation from shared layers is inferred automatically.

    Terminal window
    docker tag my-image:latest localhost:8080/my-repo:v1
    docker push localhost:8080/my-repo:v1

    Using crane? crane push my-image.tar localhost:8080/my-repo:v1 --insecure

  2. Confirm the commit was recorded

    Terminal window
    oci-bai --repo my-repo log v1

    You should see the commit id, manifest digest, and who pushed it.

  3. Push a derived image

    Push a second image built from the first. oci-bai links them automatically.

    Terminal window
    docker push localhost:8080/my-repo:v2
    oci-bai --repo my-repo provenance v2

  4. Browse in the dashboard

    Open the dashboard and select my-repo.

    • Versions — the full commit graph
    • Provenance — derivation tree for any version
    • Compare — diff any two versions

    The hosted dashboard is in early preview. Contact support@briefcaseai.org for access.

  5. Terminal window
    oci-bai search "cuda>=12.4"
    oci-bai search "format==safetensors"
    oci-bai search "numpy==1.26.4 arch==arm64"

Next steps

Search Full query syntax including model format filters.
Provenance & lineage Derivation trees and the weight-sharing metric.
Full CLI reference Every oci-bai command and its flags.