Artifact and Build Integrity
Build once, identify immutably, record provenance, verify before promotion and never rebuild “the same” release in a more privileged environment.
Frame the problem
Security starts with a concrete asset, attacker capability and trust crossing.
Why the system fails
Mutable tags, unauthenticated artifacts or separate environment rebuilds make it impossible to prove what was tested is what ran.
The important question is not “what is Artifact and Build Integrity?” but “which assumption let untrusted data or an over-scoped identity cross build output → registry → deployment?” Trace the decision at the boundary, then constrain what can happen after the first control fails.
Design the control in layers
Start with the control closest to the interpretation or privilege boundary: Use content-addressed immutable artifacts Then add a control that reduces blast radius and telemetry that proves the decision was enforced.
The resulting design is not labelled secure. Record the identified controls, the known failure paths, the remaining exposure, and the evidence you would need during an incident.
| Prevent | Detect | Recover |
|---|---|---|
| Use content-addressed immutable artifacts · Sign or attest provenance and verify at deployment · Separate builders from deployers | Digest/signature mismatches and provenance gaps | Contain the affected identity or component, scope impact from audit evidence, and preserve a regression test. |
Key points
- Asset: The exact bytes deployed to production.
- Boundary: Build output → registry → deployment
- Primary control: Use content-addressed immutable artifacts
- Detection signal: Digest/signature mismatches and provenance gaps
- Always ask what limits damage when the primary control fails.
Boundary control exercise
This lesson uses the shared boundary-control exercise.
Follow the attack
Safe conceptual simulation: capability → missing control → crossed boundary → asset impact.
- 1Attacker starts with: Someone who can modify a build, registry tag or deployment input.
- 2Mutable tags, unauthenticated artifacts or separate environment rebuilds make it impossible to prove what was tested is what ran.
- 3The weak or missing boundary control is crossed: Build output → registry → deployment
- 4Impact: Unreviewed code reaches production or incident scope becomes unknowable.
- Unreviewed code reaches production or incident scope becomes unknowable.
Defend, detect, recover
One prevention is a single point of security failure. Layer it and make failure observable.
- • Use content-addressed immutable artifacts
- • Sign or attest provenance and verify at deployment
- • Separate builders from deployers
- • Digest/signature mismatches and provenance gaps
- • Contain the affected identity or component.
- • Scope access from audit evidence.
- • Fix the boundary and add a regression test.
- • Misconfiguration and new access paths can bypass the intended control.
- • A privileged insider or compromised control plane may still reach the asset.