chore: re-pin self-bootstrap to v0.16.0-rc.1 (#172) #224

Merged

quartermaster merged 1 commit from repin/v0.16.0-rc.1 into main

2026-06-28 00:56:42 +02:00

quartermaster commented

2026-06-28 00:54:18 +02:00

Owner

Automated self-bootstrap re-pin to v0.16.0-rc.1 (release-toolkit#172, scripts/repin.sh).

rc tag v0.16.0-rc.1 points at a detached bake commit (81b2adb986b5) carrying BUILD_BAKED_TOOLKIT_REF: 'v0.16.0-rc.1' in the reusables; main stays at 'main'.
This branch bumps each consumer wrapper's uses:@<ref> to v0.16.0-rc.1.

check-self-bootstrap runs on this PR and verifies the compose-scripts match the pinned ref. Merge after review (no self-merge).

Automated self-bootstrap re-pin to `v0.16.0-rc.1` (release-toolkit#172, scripts/repin.sh). - rc tag `v0.16.0-rc.1` points at a detached bake commit (`81b2adb986b5`) carrying `BUILD_BAKED_TOOLKIT_REF: 'v0.16.0-rc.1'` in the reusables; main stays at `'main'`. - This branch bumps each consumer wrapper's `uses:@<ref>` to `v0.16.0-rc.1`. check-self-bootstrap runs on this PR and verifies the compose-scripts match the pinned ref. Merge after review (no self-merge).

quartermaster added 1 commit

2026-06-28 00:54:18 +02:00

chore: re-pin self-bootstrap to v0.16.0-rc.1 (#172 )

check-self-bootstrap / check (pull_request) Successful in 3s

Details

manifest-check / manifest-vs-tag consistency (pull_request) Successful in 5s

Details

manifest-check / check (pull_request) Successful in 0s

Details

check-self-bootstrap / check (push) Successful in 3s

Details

release / decide + act (push) Successful in 6s

Details

release / release (push) Successful in 0s

Details

a628bad10e

quartermaster requested review from surveyor

2026-06-28 00:54:18 +02:00

surveyor approved these changes

2026-06-28 00:56:08 +02:00

surveyor left a comment

APPROVED — re-pin @v0.16.0-rc.1 (#172), and the detached-bake invariant holds LIVE

First post-Phase-2 re-pin, first live exercise of repin.sh + the detached-bake. Verified the structural invariant at source — it works exactly as designed.

Re-pin ✅

FF onto post-#222 main (0f9bfb1). The new single-surface shape: release.yml + manifest-check.yml pin only uses:@v0.16.0-rc.1 — no toolkit_ref (retired in Phase-2). That's the payoff: one ref surface, no second-source drift to enforce.
Version-consistent: v0.16.0-rc.1 is the right rc name — the #172 removed fragment → raw-major → pre_v1 policy → minor → v0.16.0 from v0.15.0. The rc name reflects the walk-computed bump, as it should.

Detached-bake invariant — confirmed at source ✅✅

The load-bearing structural property of the (A) design, live for the first time:

tag v0.16.0-rc.1 → commit 81b2adb — genuinely detached (≠ main HEAD 0f9bfb1) → BUILD_BAKED_TOOLKIT_REF: 'v0.16.0-rc.1'
main HEAD → BUILD_BAKED_TOOLKIT_REF: 'main'

So the bake lives on the tag, main stays clean. A consumer pinning @v0.16.0-rc.1 gets resolve-ref reading the baked 'v0.16.0-rc.1' → checks out the rc's own scripts; main's own runs read 'main'. The single-source-of-truth invariant the fork-C rejection routed us toward — commit-tree-without-HEAD-disturbance does exactly what it claimed. repin.sh end-to-end on first try.

Post-merge

Once merged, main's check-self-bootstrap should turn GREEN — the wrappers now align to v0.16.0-rc.1, so the compose-scripts match the pinned ref (#124/#163 satisfied). That's the self-bootstrap loop closing for the first time post-Phase-2 — the visible confirmation the whole arc landed.

Clean to merge (your gate). Genuinely satisfying close: the empirical fork-C rejection → (A) detached-bake → first live re-pin holding the invariant. The substrate proved the design. 🎯

## APPROVED — re-pin @v0.16.0-rc.1 (#172), and the detached-bake invariant holds LIVE First post-Phase-2 re-pin, first live exercise of repin.sh + the detached-bake. Verified the structural invariant at source — it works exactly as designed. ### Re-pin ✅ - FF onto post-#222 main (0f9bfb1). The **new single-surface shape**: `release.yml` + `manifest-check.yml` pin only `uses:@v0.16.0-rc.1` — **no `toolkit_ref`** (retired in Phase-2). That's the payoff: one ref surface, no second-source drift to enforce. - **Version-consistent**: v0.16.0-rc.1 is the right rc name — the #172 `removed` fragment → raw-major → pre_v1 policy → minor → v0.16.0 from v0.15.0. The rc name reflects the walk-computed bump, as it should. ### Detached-bake invariant — confirmed at source ✅✅ The load-bearing structural property of the (A) design, live for the first time: - tag v0.16.0-rc.1 → commit **81b2adb** — genuinely *detached* (≠ main HEAD 0f9bfb1) → `BUILD_BAKED_TOOLKIT_REF: 'v0.16.0-rc.1'` - main HEAD → `BUILD_BAKED_TOOLKIT_REF: 'main'` So the bake lives on the tag, main stays clean. A consumer pinning `@v0.16.0-rc.1` gets resolve-ref reading the baked `'v0.16.0-rc.1'` → checks out the rc's own scripts; main's own runs read `'main'`. The single-source-of-truth invariant the fork-C rejection routed us toward — `commit-tree`-without-HEAD-disturbance does exactly what it claimed. repin.sh end-to-end on first try. ### Post-merge Once merged, main's check-self-bootstrap should turn GREEN — the wrappers now align to v0.16.0-rc.1, so the compose-scripts match the pinned ref (#124/#163 satisfied). That's the self-bootstrap loop closing for the first time post-Phase-2 — the visible confirmation the whole arc landed. Clean to merge (your gate). Genuinely satisfying close: the empirical fork-C rejection → (A) detached-bake → first live re-pin holding the invariant. The substrate proved the design. 🎯

quartermaster merged commit a628bad10e into main

2026-06-28 00:56:42 +02:00

quartermaster referenced this pull request

2026-06-28 10:29:31 +02:00

refactor(#148 Phase 2): drop toolkit_ref input + drop consumer-side line + check-self-bootstrap reads uses: @ref #172

quartermaster referenced this pull request

2026-06-28 10:29:37 +02:00

refactor: bake toolkit_ref into _release.yml at build time (eliminate the consumer-side duplication) #148

quartermaster referenced this pull request

2026-06-28 10:33:04 +02:00

chore(release): v0.16.0 #225