Pre-release readiness

The binding complete checklist is the canonical release audit.

Pre-release-ready means all release preparation and targeted acceptance are complete: Core, Demo, Console, Node Connector, CLI/JSON/MCP/HTTP compatibility, developer journeys, EN/DE documentation, Pages candidate build/navigation, changelog, release notes, immutable ecosystem SHAs and the versioned gate and evidence plan. Core-only green does not qualify. Missing, failed, skipped or unverified required proofs block this status.

Release-ready additionally requires successful complete pre-release approval. Promotion, tagging, publication and post-release verification need their own authorization. Actual Pages deployment follows promotion; candidate builds and workflow compatibility are checked before pre-release without publishing.

Reuse green gates only when the existing evidence verifier proves unchanged inputs. Rerun affected gates only. Preparation does not authorize pre-release.

Parallel local checks

Run independent local checks together against an isolated copy of the current source, including staged, unstaged and untracked inputs:

python scripts/local-qualification.py --jobs 4 --race-package ./internal/application --race-package ./contracts

Select affected race packages by repeating --race-package; --go selects an installed Go executable. The runner shares compilation caches, bounds parallel workers and records the captured commit/tree, every check's status, duration and log in its printed result directory. A failed, cancelled or timed-out check fails the aggregate. Edits after capture do not change the tested snapshot. Go tests, vet, CLI build, script tests, docs audits, readability and strict EN/DE builds run independently. Runtime acceptance flags are excluded: native Docker/Podman gates remain separately qualified on their exact inputs. Local success is not ecosystem or pre-release readiness. Reuse unchanged qualified inputs through the existing evidence verifier rather than repeatedly running the full local suite.