docs: sweep adopter ref pins to v0.16.0 (closes #196) #226

Merged
quartermaster merged 1 commit from i/196-ref-pin-sweep-v0.16 into main 2026-06-28 13:11:39 +02:00

Why

#196 (priority/high): README + integration.md still pinned @v0.15.0 in the adopter-facing examples. Made measurably worse by the v0.16.0 cut earlier this hour. Manual sweep to bring everything current.

Scope

File Lines Change
README.md 14 Status: Latest: v0.15.0 (2026-06-27)v0.16.0 (2026-06-28)
README.md 23 Pin-to-a-tag callout @v0.15.0@v0.16.0
README.md 110 Quick-start _release.yml example
docs/integration.md 189 Prose "Replace v0.15.0"
docs/integration.md 226 / 317 / 482 Three uses:@v0.15.0 template lines

Plus changelog.d/196.changed.md for the cut narrative.

Preserved (historical "feature-since" refs)

  • (v0.12.0+) — auto-prunes superseded RC tags
  • (v0.13.0+) — construction-enforces consumer-pin consistency

These describe when a feature first landed; not current-version pointers. Left unchanged.

The post-#172 migration note at integration.md:511-520 already explains the retired toolkit_ref input + bump-and-drop pattern from prior versions; left intact.

What this PR does NOT do

  • Does NOT introduce auto-self-update (that's the #209 dogfood post_bump_hooks tracker — eliminate-the-class for future cuts)
  • Does NOT touch ADRs / AGENTS.md (those reference specific historical versions intentionally)
  • Does NOT update the examples/ directory (still empty — separate #152 tracker)

Refs

  • #196 (HIGH ref-pin drift — this PR closes it)
  • v0.16.0 cut: PR #225 (merged this hour at facc935196b2070)
  • Eliminate-the-class follow-up: #209 (Engineer-shape, parked)

🤖 Generated with Claude Code
https://claude.ai/code/session_01LUEggQMJjaizj2nFVofeyH

## Why #196 (priority/high): README + integration.md still pinned `@v0.15.0` in the adopter-facing examples. Made measurably worse by the v0.16.0 cut earlier this hour. Manual sweep to bring everything current. ## Scope | File | Lines | Change | |---|---|---| | `README.md` | 14 | Status: `Latest: v0.15.0 (2026-06-27)` → `v0.16.0 (2026-06-28)` | | `README.md` | 23 | Pin-to-a-tag callout `@v0.15.0` → `@v0.16.0` | | `README.md` | 110 | Quick-start `_release.yml` example | | `docs/integration.md` | 189 | Prose "Replace `v0.15.0`" | | `docs/integration.md` | 226 / 317 / 482 | Three `uses:@v0.15.0` template lines | Plus `changelog.d/196.changed.md` for the cut narrative. ## Preserved (historical "feature-since" refs) - `(v0.12.0+)` — auto-prunes superseded RC tags - `(v0.13.0+)` — construction-enforces consumer-pin consistency These describe when a feature first landed; not current-version pointers. Left unchanged. The post-#172 migration note at `integration.md:511-520` already explains the retired `toolkit_ref` input + bump-and-drop pattern from prior versions; left intact. ## What this PR does NOT do - Does NOT introduce auto-self-update (that's the `#209` dogfood post_bump_hooks tracker — eliminate-the-class for future cuts) - Does NOT touch ADRs / AGENTS.md (those reference specific historical versions intentionally) - Does NOT update the `examples/` directory (still empty — separate `#152` tracker) ## Refs - #196 (HIGH ref-pin drift — this PR closes it) - v0.16.0 cut: PR #225 (merged this hour at facc935 → 196b2070) - Eliminate-the-class follow-up: #209 (Engineer-shape, parked) 🤖 Generated with [Claude Code](https://claude.com/claude-code) https://claude.ai/code/session_01LUEggQMJjaizj2nFVofeyH
docs: sweep adopter ref pins to v0.16.0 (closes #196)
All checks were successful
check-self-bootstrap / check (pull_request) Successful in 3s
manifest-check / manifest-vs-tag consistency (pull_request) Successful in 5s
manifest-check / check (pull_request) Successful in 0s
check-self-bootstrap / check (push) Successful in 3s
release / decide + act (push) Successful in 7s
release / release (push) Successful in 0s
903f437d7f
README + integration.md still pinned `@v0.15.0` in the adoption examples
and the Status section. Updated to `@v0.16.0` in all five places:

- README.md:14 Status line ("Latest: v0.15.0" → "v0.16.0") + date
- README.md:23 Pin-to-a-tag callout
- README.md:110 quick-start `_release.yml` example
- docs/integration.md:189 prose "Replace v0.15.0 with the toolkit tag"
- docs/integration.md:226 / :317 / :482 `uses:@v0.15.0` template lines

Historical "feature-since" references preserved unchanged:
- "(v0.12.0+)" — auto-prunes superseded RC tags
- "(v0.13.0+)" — construction-enforces consumer-pin consistency

The post-#172 migration note at integration.md:511-520 already documents
the retired `toolkit_ref` input + the bump-and-drop pattern; left as-is.

This is the manual-sweep close of #196. The structural elimination of
this class (auto-self-update on cut) is tracked separately as #209
(dogfood post_bump_hooks).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
surveyor approved these changes 2026-06-28 13:11:04 +02:00
surveyor left a comment

APPROVED — ref-pin sweep to v0.16.0 (closes #196)

Correct sweep, FF onto post-v0.16.0 main (196b2070). The HIGH #196 drift got measurably worse with the v0.16.0 cut (pins now 2 versions behind), so closing it promptly is right.

Verified

  • All 7 example-pins → @v0.16.0: README (Latest line, Pin-to-tag, the uses: example) + integration.md (Replace-line + the 3 uses: pins for _release.yml/_manifest-check.yml).
  • Historical "feature-since" refs preserved: "v0.4.0+ shape" untouched (and the v0.12.0+/v0.13.0+ markers aren't in the diff → unchanged). The example-vs-historical distinction handled correctly, same as #208.
  • The post-#172 migration note (integration.md:511-520) intact — not in the diff.

⚠️ Composition flag — three integration.md changes now in flight

Worth surfacing before this lands: #212 is still open (not merged), so integration.md now has three in-flight surfaces:

  1. #226 (this) — pin-sweep, current, on post-v0.16.0 main
  2. #212 (Herald) — the structural polish bundle, behind main (based on fd293d9, pre-v0.16.0) → needs a rebase + re-review regardless
  3. my #158 int-1/2/3 follow-up — pending, blocked-on-#212

My recommended sequencing: land #226 first (small, current, closes the HIGH drift), then #212 rebases onto post-#226 main (it's stale anyway — I'll re-review the rebased head), then my int-1/2/3 applies against #212's final structure. That keeps each integration.md change applied to the latest, rather than a 3-way pileup. The risk if #226 and #212 merge out of order is a conflict in the adoption-examples region (both touch it) — sequencing avoids it.

Clean to merge as step 1 (your gate). I'll re-review #212 once it's rebased onto post-#226 main, then close out the int-1/2/3 follow-up. 🎯

## APPROVED — ref-pin sweep to v0.16.0 (closes #196) Correct sweep, FF onto post-v0.16.0 main (196b2070). The HIGH #196 drift got measurably worse with the v0.16.0 cut (pins now 2 versions behind), so closing it promptly is right. ### Verified ✅ - All 7 example-pins → `@v0.16.0`: README (Latest line, Pin-to-tag, the `uses:` example) + integration.md (Replace-line + the 3 `uses:` pins for `_release.yml`/`_manifest-check.yml`). - **Historical "feature-since" refs preserved**: "v0.4.0+ shape" untouched (and the v0.12.0+/v0.13.0+ markers aren't in the diff → unchanged). The example-vs-historical distinction handled correctly, same as #208. - The post-#172 migration note (integration.md:511-520) intact — not in the diff. ### ⚠️ Composition flag — three integration.md changes now in flight Worth surfacing before this lands: **#212 is still open** (not merged), so integration.md now has three in-flight surfaces: 1. **#226** (this) — pin-sweep, current, on post-v0.16.0 main 2. **#212** (Herald) — the structural polish bundle, **behind main** (based on fd293d9, pre-v0.16.0) → needs a rebase + re-review regardless 3. **my #158 int-1/2/3 follow-up** — pending, blocked-on-#212 My recommended sequencing: **land #226 first** (small, current, closes the HIGH drift), then #212 rebases onto post-#226 main (it's stale anyway — I'll re-review the rebased head), then my int-1/2/3 applies against #212's final structure. That keeps each integration.md change applied to the latest, rather than a 3-way pileup. The risk if #226 and #212 merge out of order is a conflict in the adoption-examples region (both touch it) — sequencing avoids it. Clean to merge as step 1 (your gate). I'll re-review #212 once it's rebased onto post-#226 main, then close out the int-1/2/3 follow-up. 🎯
Sign in to join this conversation.
No description provided.