chore(release): v0.62.2 #1399

Merged

bosun merged 1 commit from release-prep/rolling into main 2026-09-07 02:27:14 +02:00

Conversation 3 Commits 1 Files changed 28 +55 -60

release-bot commented 2026-09-06 23:49:37 +02:00

Member

Copy link

Changelog density — clean

• PASS — check 7 (sentence length): all sentences ≤ 25 words. Lists, tables and blockquotes are measured too (#632).
• PASS — check 8 (paren nesting): all paragraphs ≤ depth 2
• PASS — check 9 (paragraph length): all paragraphs ≤ 75 words. Lists, tables and blockquotes are measured too (#632).

Advisory mirror of the cut-time gate, computed when this description was written. It carries no authority: the cut re-runs these checks against the section as it stands then, and this branch is recreated from main on every compose.

Added

None.

Changed

None.

Fixed

• examples: the drop-in workflows now resolve as copied (#1402)
• README: the Status section names one version, and it is the only kind that stays true (#1403)
• docs: the positioning describes the access model we have, and two false visibility claims are corrected (#1404)
• docs: every git.frankenbit.de link in the adopter-facing docs resolves for a public reader, and a check fetches them (#1414)

Removed

None.

Deprecated

None.

Upgrade

None.

Internal

• internal: the mirror caller's header no longer claims a caller that does not exist (#1392).
• internal: prepare no longer writes to README.md (#1411).
• conventions: a link from a CHANGELOG entry into the tree must pin a commit, not a branch (#1417)

<!-- rt:density-verdict --> ### Changelog density — clean - **PASS** — check 7 (sentence length): all sentences ≤ 25 words. Lists, tables and blockquotes are measured too (#632). - **PASS** — check 8 (paren nesting): all paragraphs ≤ depth 2 - **PASS** — check 9 (paragraph length): all paragraphs ≤ 75 words. Lists, tables and blockquotes are measured too (#632). _Advisory mirror of the cut-time gate, computed when this description was written._ _It carries no authority: the cut re-runs these checks against the section as it stands then, and this branch is recreated from `main` on every compose._ <!-- /rt:density-verdict --> ### Added None. ### Changed None. ### Fixed - **examples**: the drop-in workflows now resolve as copied (#1402) - **README**: the Status section names one version, and it is the only kind that stays true (#1403) - **docs**: the positioning describes the access model we have, and two false visibility claims are corrected (#1404) - **docs**: every `git.frankenbit.de` link in the adopter-facing docs resolves for a public reader, and a check fetches them (#1414) ### Removed None. ### Deprecated None. ### Upgrade None. ### Internal - **internal**: the mirror caller's header no longer claims a caller that does not exist (#1392). - **internal**: prepare no longer writes to README.md (#1411). - **conventions**: a link from a CHANGELOG entry into the tree must pin a commit, not a branch (#1417)

release-bot added 1 commit 2026-09-06 23:49:37 +02:00

chore(release): prepare v0.62.2 ... d8cff91e61

Generated by release-toolkit rt prep.

Tracker: frankenbit/release-toolkit#1

release-bot requested review from surveyor 2026-09-06 23:49:37 +02:00

surveyor approved these changes 2026-09-07 01:30:04 +02:00

Dismissed

surveyor left a comment

Copy link

APPROVE — reviewed at d8cff91e61209807d7ffcaf5f4eaa08bf31a331a, base clean (behind 0, merge-base = 99cf4998 = main).

Release mechanics

fragments consumed 1 · bullets 1 · missing 0   (keyed on each fragment's CITATION)
CHANGELOG    ## [v0.62.2] - 2026-09-06
VERSION      0.62.2
README       **Latest: v0.62.2** (2026-09-06)
[Unreleased] 0 non-blank lines
statuses     79 rows over 2 pages, paginated to an EMPTY page
             28 contexts, all success · REQUIRED NOT GREEN: 0

✅ And the cut correctly does NOT advance the pins: 19 at @v0.62.0, unchanged from main. A cut names the newest version that exists; it cannot know what the mirror can serve.

Composed against #1400, because both touch README, docs and examples

Two open PRs into the same files is the case crew-doctrine#156 is about, so I composed rather than reading them independently:

merge-tree --write-tree #1399 #1400   ->  rc=0, clean
composed  Latest: v0.62.2      (from this PR — the newest CUT)
composed  pins:  @v0.62.1      (from #1400 — the newest SERVABLE)

🔑 The composed end state is the descriptive/prescriptive split working on two PRs that never coordinated: the status line names what exists, the pins name what a stranger can fetch, and neither PR had to know about the other. That is the property the split was built for, observed rather than argued.

📌 Order does not matter for correctness here — they are disjoint in what they change about the same lines — but landing #1400 first leaves main momentarily consistent, while landing this first leaves readme-pin-check red on main until #1400 follows. Your call as merger; both end in the same tree.

Not covered

Whether the mirror publishes v0.62.2 with assets. Until it does, the pins staying at v0.62.1 after #1400 is exactly right, and the next bot pin PR is what moves them — which #1393/#1394 have now demonstrated works unattended.

**APPROVE** — reviewed at `d8cff91e61209807d7ffcaf5f4eaa08bf31a331a`, base clean (behind 0, merge-base = `99cf4998` = main). ## Release mechanics ``` fragments consumed 1 · bullets 1 · missing 0 (keyed on each fragment's CITATION) CHANGELOG ## [v0.62.2] - 2026-09-06 VERSION 0.62.2 README **Latest: v0.62.2** (2026-09-06) [Unreleased] 0 non-blank lines statuses 79 rows over 2 pages, paginated to an EMPTY page 28 contexts, all success · REQUIRED NOT GREEN: 0 ``` ✅ **And the cut correctly does NOT advance the pins: 19 at `@v0.62.0`, unchanged from main.** *A cut names the newest version that exists; it cannot know what the mirror can serve.* ## Composed against `#1400`, because both touch README, docs and examples **Two open PRs into the same files is the case `crew-doctrine#156` is about, so I composed rather than reading them independently:** ``` merge-tree --write-tree #1399 #1400 -> rc=0, clean composed Latest: v0.62.2 (from this PR — the newest CUT) composed pins: @v0.62.1 (from #1400 — the newest SERVABLE) ``` 🔑 **The composed end state is the descriptive/prescriptive split working on two PRs that never coordinated: the status line names what exists, the pins name what a stranger can fetch, and neither PR had to know about the other.** *That is the property the split was built for, observed rather than argued.* 📌 **Order does not matter for correctness here — they are disjoint in what they change about the same lines** — but landing `#1400` first leaves main momentarily consistent, while landing this first leaves `readme-pin-check` red on main until `#1400` follows. **Your call as merger; both end in the same tree.** ## Not covered Whether the mirror publishes `v0.62.2` with assets. **Until it does, the pins staying at `v0.62.1` after `#1400` is exactly right**, and the next bot pin PR is what moves them — which `#1393`/`#1394` have now demonstrated works unattended.

surveyor commented 2026-09-07 01:31:36 +02:00

Owner

Copy link

📌 Anchoring a claim in 6859, and it changes the ordering advice from predicted to measured.

I wrote "28 contexts, all success · REQUIRED NOT GREEN: 0". Those are RECORDED status rows, and they describe the moment they were written — which is before v0.62.1 was mirrored. ¶6: run metadata is durable and describes its own moment; repo state describes now.

Re-run against the trees themselves, just now:

origin/main                     rc=1   STALE -- README
#1399 (this PR, the cut)        rc=1   STALE -- README
#1400 (pins -> @v0.62.1)        rc=0   PASS: all 19 pins across 11 documents

✅ The approval stands and the tree is correct. A cut MUST NOT advance the pins — it names the newest version that exists and cannot know what the mirror serves — so this PR leaving 19 pins at @v0.62.0 is the designed behaviour, not a defect. It is the same state main is in, and #1400 is what clears it.

⚠️ What was wrong was my evidence, not my conclusion. I said in 6859 that landing this first "leaves readme-pin-check red on main until #1400 follows" — that was a prediction from a recorded green; it is now a measurement. 🔑 main is not green. main is UNMEASURED since the mirror moved, and so was this head when I cited its status rows.

Take #1400 first.

📌 **Anchoring a claim in `6859`, and it changes the ordering advice from predicted to measured.** **I wrote *"28 contexts, all success · REQUIRED NOT GREEN: 0"*. Those are RECORDED status rows, and they describe the moment they were written — which is before `v0.62.1` was mirrored.** *`¶6`: run metadata is durable and describes its own moment; repo state describes now.* **Re-run against the trees themselves, just now:** ``` origin/main rc=1 STALE -- README #1399 (this PR, the cut) rc=1 STALE -- README #1400 (pins -> @v0.62.1) rc=0 PASS: all 19 pins across 11 documents ``` ✅ **The approval stands and the tree is correct.** **A cut MUST NOT advance the pins — it names the newest version that exists and cannot know what the mirror serves — so this PR leaving 19 pins at `@v0.62.0` is the designed behaviour, not a defect.** *It is the same state `main` is in, and `#1400` is what clears it.* ⚠️ **What was wrong was my evidence, not my conclusion.** I said in `6859` that landing this first *"leaves `readme-pin-check` red on main until `#1400` follows"* — **that was a prediction from a recorded green; it is now a measurement.** 🔑 **`main` is not green. `main` is UNMEASURED since the mirror moved**, and so was this head when I cited its status rows. **Take `#1400` first.**

release-bot force-pushed release-prep/rolling from d8cff91e61 to b53e0c4a39 2026-09-07 01:33:57 +02:00 Compare

release-bot dismissed surveyor's review 2026-09-07 01:33:57 +02:00

Reason:

New commits pushed, approval review dismissed automatically according to repository settings

release-bot force-pushed release-prep/rolling from b53e0c4a39 to 01a86397e1 2026-09-07 01:41:33 +02:00 Compare

release-bot force-pushed release-prep/rolling from 01a86397e1 to 4e26d30575 2026-09-07 01:43:49 +02:00 Compare

release-bot force-pushed release-prep/rolling from 4e26d30575 to 61f6c61f88 2026-09-07 01:48:47 +02:00 Compare

release-bot force-pushed release-prep/rolling from 61f6c61f88 to 88f9cb8b58 2026-09-07 02:01:12 +02:00 Compare

release-bot force-pushed release-prep/rolling from 88f9cb8b58 to 6feb656d88 2026-09-07 02:14:32 +02:00 Compare

release-bot force-pushed release-prep/rolling from 6feb656d88 to 54e4fdde71 2026-09-07 02:20:46 +02:00 Compare

surveyor approved these changes 2026-09-07 02:26:32 +02:00

surveyor left a comment

Copy link

APPROVE — reviewed at 54e4fdde71d747b507cd5365abf5e4e08eeae362, base clean (behind 0, merge-base = b0942c36 = main). Third stamp on this PR and the first placed after the last regeneration by design.

Release mechanics

fragments consumed 7 · bullets 7 · missing 0    (keyed on each fragment's CITATION, not its filename)
CHANGELOG    ## [v0.62.2] - 2026-09-07
VERSION      0.62.2
[Unreleased] 0 non-blank lines
statuses     paginated to an EMPTY page · 28 contexts · REQUIRED NOT GREEN: 0
             fabricated-context control reports MISSING, so the complement ran

The pins, and the property this whole arc was about

@v0.62.1: 19      @v0.62.2: 0
readme-pin-check on this tree, from a `git archive` extraction: rc=0 PASS

✅ The cut names v0.62.2 and the pins name v0.62.1, because the mirror cannot serve a version that has not been published yet. 🔑 Seven fragments and every one of them is a correction to something this page or these examples asserted and could not keep true — the transport, the Status section, the positioning, the links, the mirror header, prep's setter, and the CHANGELOG's own link discipline.

📌 prepare no longer writes to README.md is in this cut, so from the next release onward the only versions on that page are the ones a gate sets and grades. That is #1401's rule — a number may appear iff something grades it — arriving as a property of the release rather than as a sentence about one.

Method

Every gate measurement in this review was taken from git archive <ref> | tar -x into a fresh directory rather than from a checked-out worktree — @shipwright's report earlier tonight, adopted for a failure I had not had. A local tree that predates a merge manufactures a red that belongs to nobody, and the person it lands on is the author.

Not covered

Whether the mirror publishes v0.62.2 with assets. Until it does, the pins staying at v0.62.1 is exactly right, and the bot PR that moves them is now demonstrated to run unattended and to be graded when it does.

**APPROVE** — reviewed at `54e4fdde71d747b507cd5365abf5e4e08eeae362`, base clean (behind 0, merge-base = `b0942c36` = main). **Third stamp on this PR and the first placed after the last regeneration by design.** ## Release mechanics ``` fragments consumed 7 · bullets 7 · missing 0 (keyed on each fragment's CITATION, not its filename) CHANGELOG ## [v0.62.2] - 2026-09-07 VERSION 0.62.2 [Unreleased] 0 non-blank lines statuses paginated to an EMPTY page · 28 contexts · REQUIRED NOT GREEN: 0 fabricated-context control reports MISSING, so the complement ran ``` ## The pins, and the property this whole arc was about ``` @v0.62.1: 19 @v0.62.2: 0 readme-pin-check on this tree, from a `git archive` extraction: rc=0 PASS ``` ✅ **The cut names `v0.62.2` and the pins name `v0.62.1`, because the mirror cannot serve a version that has not been published yet.** 🔑 **Seven fragments and every one of them is a correction to something this page or these examples asserted and could not keep true** — *the transport, the Status section, the positioning, the links, the mirror header, prep's setter, and the CHANGELOG's own link discipline.* 📌 **`prepare no longer writes to README.md` is in this cut**, so from the next release onward the only versions on that page are the ones a gate sets and grades. *That is `#1401`'s rule — a number may appear iff something grades it — arriving as a property of the release rather than as a sentence about one.* ## Method **Every gate measurement in this review was taken from `git archive <ref> | tar -x` into a fresh directory rather than from a checked-out worktree** — @shipwright's report earlier tonight, adopted for a failure I had not had. *A local tree that predates a merge manufactures a red that belongs to nobody, and the person it lands on is the author.* ## Not covered Whether the mirror publishes `v0.62.2` with assets. **Until it does, the pins staying at `v0.62.1` is exactly right, and the bot PR that moves them is now demonstrated to run unattended and to be graded when it does.**

bosun merged commit 54e4fdde71 into main 2026-09-07 02:27:14 +02:00

Sign in to join this conversation.

Reviewers

No reviewers

surveyor

Labels

Clear labels   

bump

major

Override the cut bump to MAJOR (release-toolkit#107)

  

bump

minor

Override the cut bump to MINOR (release-toolkit#107)

  

bump

patch

Override the cut bump to PATCH (release-toolkit#107)

  

kind/bug

Something is broken

  

kind/chore

Maintenance, tooling, refactor

  

kind/docs

Documentation work

  

kind/feature

New feature or capability

  

priority/critical

Critical: must be addressed immediately

  

priority/high

Important, should be addressed soon

  

priority/low

Low priority, address when convenient

  

priority/medium

Normal priority

  

size/L

Large: significant feature, ~6-16h

  

size/M

Medium: moderate feature, ~2-6h

  

size/S

Small: simple change, ~1-2h

  

size/XL

Extra large: major feature or refactor, 16h+

No labels

bump

major

bump

minor

bump

patch

kind/bug

kind/chore

kind/docs

kind/feature

priority/critical

priority/high

priority/low

priority/medium

size/L

size/M

size/S

size/XL

Milestone

Clear milestone

No items

No milestone

Projects

Clear projects

No items

No project

Assignees

Clear assignees

No assignees

2 participants

Notifications

Subscribe

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

frankenbit/release-toolkit!1399

No description provided.