Report-backed review
The release candidate should have an immutable subject identity and dimensioned report the team can export, share, invalidate, and revisit when evidence changes.
pixie.codes
Create checks GS1 data, the QR itself, and an optional predictive screen. Print quality, checkout performance, and resolver behavior require their own observations. No single result approves release.
Example software-preflight state · six independent areas
These are real verifier status codes for one bounded example state. They do not combine into an overall PASS, certification, print approval, POS readiness, or universal-scanning claim.
data_conformance
Status: Data-conformant
Scope: Canonical Digital Link syntax and AI relationships checked by the recorded source set.
Provenance/time: Data-oracle versions and run time belong in the report.
Next: Re-run when the canonical payload changes.
qr_correctness
Status: QR-correct
Scope: The subject grid decodes to the encoder-declared exact UTF-8 bytes and passes structural checks.
Provenance/time: Encoder contract, grid hash, QR-oracle versions, and run time stay bound.
Next: Invalidate and re-run after any payload or grid change.
predictive_screen
Status: Indeterminate—out of scope
Scope: Preflight alone does not supply a synthetic design-screen observation.
Provenance/time: No compatible simulation observation is attached in this state.
Next: Run simulation against the exact same subject.
physical_print_quality
Status: Indeterminate—physical verification required
Scope: Software cannot observe the final printer, ink, substrate, size, finish, or production setup.
Provenance/time: No matching physical specimen observation is attached.
Next: Test the bound production specimen and record the method.
pos_performance
Status: Indeterminate—out of scope
Scope: Digital preflight does not establish target-device checkout behavior.
Provenance/time: No device, firmware, configuration, or acceptance record is attached.
Next: Run the agreed target-POS protocol on the matching specimen.
resolver_behavior
Status: Indeterminate—missing evidence
Scope: QR correctness does not prove current redirects, linksets, or resolver policy.
Provenance/time: No time-bounded resolver observation is attached in this state.
Next: Probe the resolver and record scope, response, and expiry.
Review each result independently: every status remains tied to its own method, subject, provenance, and observation time. Missing, stale, unavailable, or conflicting evidence remains indeterminate.
The release candidate should have an immutable subject identity and dimensioned report the team can export, share, invalidate, and revisit when evidence changes.
If the asset is weak, the workflow should point back to the design or print variable that needs attention rather than leaving the team with only a generic “not ready” result.
Verification is most valuable when it narrows uncertainty. The team should know which exact asset each observation covers; release approval remains a separate owner decision.
The report should support packaging, operations, and partner handoffs without forcing every team to rerun the entire decision context from scratch.
Verification should run against the specific asset under consideration, not an approximate version from earlier in the design cycle.
If a dimension fails, becomes stale, or disagrees with another observation, record what changed next. The evidence history matters more than a synthetic score.
Export a scoped report that keeps physical/POS evidence and time-scoped resolver observations explicit instead of collapsing them into a release verdict.
When the team can point to an immutable subject, scoped observations, and explicit indeterminate states, launch decisions do not depend on a single pass/fail badge.