[brettheap]

Lane: openreposhape-3 (openRepoShape-3)
Claim: #563 (comment)

Refs #563. This is #563's third pull request, the docs, and it does not close the issue.

Head aef568f0, on main 6f64eea2. It lands LAST, before the tag, because pipx install openreposhape works only once 1.1.0 is on PyPI. Rounds A, A3, A4, B1 and both parts of B2 of the final refresh are in; every other v1.1.0 pull request has landed. Lane 2's #627 (README and the handbook) has landed and is merged here. #563's other two pull requests have landed: #590, the package (fe237ec), and #599, the release workflow (32c22b0). The base is now main. The diff against it is this pull request's twelve files: README.md, AGENTS.md, docs/handbook.html, docs/cli.md, scripts/render-cli-reference.py, adopt-project.py, docs/release-notes/v1.1.0.md, tests/readme-cap.d/563.txt, tests/readme-cap.d/460.txt, tests/readme-cap.d/683.txt, tests/test_repo_hygiene.py and tests/test_agents_md_codes.py. The commit list also shows #590's and #599's branch commits, because those branches were merged here before they were squashed on main.

What

Decision 6 of #563's design (issuecomment-6191672077), approved by Brett Heap at 2026-10-15T13:32:51Z ("approve all six for #563"): the one-line install stays first, and pipx install openreposhape comes second, and first on native Windows.

• README.md (1474 lines, at its cap: main's 1459, + 12 from tests/readme-cap.d/563.txt, + 2 from tests/readme-cap.d/460.txt for Tracking: shape-doctor.py, reporting a repository's compliance #460's L1 below, + 1 from tests/readme-cap.d/683.txt for the doctor's exit 3, rounds A3, A4 and B2):
  • Step 3 has a second way, from PyPI: pipx install openreposhape, where pipx comes from, what it places, how it upgrades, and to use only one installer.
  • The native Windows block leads with pipx. The two-command download follows it, as the way that installs nothing, and tests: a tests-git-bash job runs the launcher's and install.sh's suites in Git for Windows' own bash, so README and the handbook say Git Bash runs the one-line install, tested in CI but for a typed yes, which it cannot take there (#585) #596's Git Bash sentence is kept.
  • § Adopting: step 3's one line installs git-filter-repo or prints the command for this machine, and the package brings it.
  • It says the bash openRepoShape --preflight exits 2, and that the installed doctor KEEPS "its clone" (no longer "its temporary clone").
  • Layout gets a src/openreposhape/ row, and the standard-library sentence says the package is a second way in, never a requirement.
• AGENTS.md (494 lines, at its cap of 494, raised from main's 492 in test_agents_md_is_short_enough_to_be_read):
  • The first rule: the one-line install.sh and pipx install openreposhape are the human's to run.
  • Section 2: the installed command may be pipx's, native Windows included, and pipx adds the four tool commands.
• docs/handbook.html carries the same facts in the page's own voice, with every README marker recomputed. Step 3 on the Windows tab:
  • the heading names three ways: "Install with pipx, fetch the flow, or use Git Bash's one line" (:788);
  • for Git Bash it shows the one-line install, "In Git Bash, the one-line install runs too:" (:829-830), with the caveat that its preflight cannot take a typed yes there (:831-832). This settles Tracking: the test suite, CI and the docs #461's line "The handbook's step 3 has no Git Bash path".
  • test_every_handbook_copy_of_the_one_line_install_is_the_same (tests/test_repo_hygiene.py:1578) holds all three handbook copies of the one line to ONE_LINE_INSTALL.
• adopt-project.py: filter-repo-missing names git-filter-repo's install in install.sh's order: the pipx or uv already there, then brew, the apt/dnf pipx route, then pacman. It says pipx install openreposhape brings it for its adopt-project.
• docs/cli.md, from scripts/render-cli-reference.py: the checkout group says what pipx install openreposhape places.
• docs/release-notes/v1.1.0.md, a draft:
  • What 1.1.0 adds; install and upgrade.
  • What changed since v1.0.1, by area, to main fc86720 (121 pull requests).
  • Known limits: the package's and Git Bash's own, then lane openRepoShape-4's ten lines and lane openRepoShape-1's twenty-six D9 lines, as two lists.
  • How this release is cut: the environments, the tag ruleset, the rehearsal run and the trusted publishers, all done.
  • Before tagging.

Also #460's L1 (5b140bd)

#630 also carries one line of #460, which Brett Heap gave to lane 3 at 2026-10-16T12:24:41Z. Lane 3's triage (T460a) put it here: every file it needs is already in #630, and #630 lands after #627 anyway.

• The line. README.md, docs/handbook.html and the comments in bump-leg.py and update-shape.py say when a project's own gate reaches a network: validate-pins.py asks gh api for a neutral product's pin with the operator's gh credentials, and make bootstrap, bump-leg.py and update-shape.py apply run it in the operator's environment, so none of them says any more that a project runs its own gate offline or that apply is otherwise offline (#411) #465 removed a sentence from README's validate-pins.py paragraph because its other half was false: a declared pin file that is not there is the refusal neutral-pin-missing (exit 2), not drift. The removal also took the true half, that a digest that recomputes differently is the finding neutral-pin-digest-mismatch (exit 1). v1.0.0's notes list that as a known limit (docs/release-notes/v1.0.0.md:507).
• The words. In README (:739-744), after the SKIP sentence: "A digest that recomputes differently is the finding neutral-pin-digest-mismatch (exit 1), and a checkout that cannot answer is the finding neutral-pin-source-unreadable; a declared pin file that is not there is the refusal neutral-pin-missing (exit 2)." The handbook's twin in legs (:1239-1242) says the same.
• The codes, checked on this tree in templates/assembly-root/scripts/validate-pins.py. neutral-pin-digest-mismatch is a report.finding, so exit 1. neutral-pin-missing is a raise Refusal, which main turns into exit 2.
• Cost and markers. README grows 2 lines, with tests/readme-cap.d/460.txt (+2). The three The three legs markers move: legs changes text and marker; the masthead and ci change the marker only. Lane 2's README.md and the handbook: say that a source's own single-repository.yaml record arrives as leg: drop, that the adopted Makefile's CONTRACTS_DIR follows the spec leg's mount, and that a second setup run is refused at the plan, so the three sentences #566, #572 and #573 left false or incomplete are true of main (#606) #627 moved the same three markers; the final refresh's merge of main resolved them by recomputing every marker (see below).
• The notes gain one line under "Install, CI and the docs" saying v1.0.0's known limit is closed.

Rounds

• Round 1 (adb4313..82f9403): the docs, the hint and the notes draft.

• RV630 round 1: no MUST. SHOULD 1-6 and nits 1, 3-9 were taken (b347943, c3e9a3d, 0470497, b006f16):

  • README's raise went from +28 to +12, and AGENTS.md's cap from 495 to 494;
  • the hint follows install.sh's whole order;
  • "temporary clone" is gone, and the exit 2 is said to be the bash command's;
  • the notes' release step 1 and references were fixed.

• Merges:

  • 2a7865e merged main 4cf5fb07 (tests: a tests-git-bash job runs the launcher's and install.sh's suites in Git for Windows' own bash, so README and the handbook say Git Bash runs the one-line install, tested in CI but for a typed yes, which it cannot take there (#585) #596's Git Bash job);
  • ca5c49f merged The release workflow: a dispatch on the annotated tag v<version> builds the openreposhape wheel twice from git archive of a commit whose own tree passed the tests job, uploads it to TestPyPI, installs it with pipx on three systems and Python 3.9, and uploads the same bytes to PyPI on approval (#563) #599's 750366bc, which carries The openreposhape package: openRepoShape and four tool commands, run from a clone of the standard at the stamped commit and checked against its digest; --install and install.sh refuse a symlinked target (#563) #590's Git Bash test spellings;
  • 493c6ad merged main d4a7fab9, after The openreposhape package: openRepoShape and four tool commands, run from a clone of the standard at the stamped commit and checked against its digest; --install and install.sh refuse a symlinked target (#563) #590 and The release workflow: a dispatch on the annotated tag v<version> builds the openreposhape wheel twice from git archive of a commit whose own tree passed the tests job, uploads it to TestPyPI, installs it with pipx on three systems and Python 3.9, and uploads the same bytes to PyPI on approval (#563) #599 landed;
  • 9e504c0 merged main 879d8047 (adopt-project.py: adopt-bad-id names an id derived from --project as derived, so --project North_Wind no longer says --id 'north_wind' must match over a flag nobody passed (#601) #620);
  • 7b7e657 merged main de926142 (adopt-project.py: the read-across follow-up names legs.code_path and legs.spec_path and no path, and the contracts/ rule's reason names none, so the hand edit of a leg's mount the tool asks for leaves no plan-time path in check, the split commit or the pull request (#600) #628, validate-family.py and family.py: a gitlink no section of HEAD's .gitmodules gives is member-gitmodules-absent, remove refuses a holder with no .gitmodules before deinit, and two gitlink remedies name what they missed (#611) #639, adopt-project.py and scaffold-project.py: the Free-plan secrets hint goes by the assembly root's own visibility on GitHub and names the root as its reason, so a public source adopted with private legs is no longer told its organisation secrets resolve to EMPTY (#602) #621, README.md and the handbook: say that a source's own single-repository.yaml record arrives as leg: drop, that the adopted Makefile's CONTRACTS_DIR follows the spec leg's mount, and that a second setup run is refused at the plan, so the three sentences #566, #572 and #573 left false or incomplete are true of main (#606) #627, bump-leg.py and setup-project.py: the git switch -c line says under it what to do where an earlier act left that branch, so already exists is no dead end (#607's sentence, in the two tools it left) #657, openRepoShape in Git Bash: the machine: row finds python, a drive-letter $OPENREPOSHAPE_REPO is a path, a Windows-spelled $OPENREPOSHAPE_BIN_DIR is respelled or refused, and tests-git-bash runs #598's installer cases (#649) #656, tests/test_doctor_next_is_a_command.py: the comment-tail guard reads each string expression whole, across + and an f-string's pieces, and flags a # led by two or more blanks or a tab with or without a blank after it and a # that opens text after a part that is not a constant, so cmd #x, cmd + " " + "# x" and f"{cmd}{pad}# x" no longer reach a NEXT unseen (#679) #698). The handbook conflicted at eight marker hunks only: the seven § Starting markers, the three The three legs markers that share their hunks, and the one § Adopting marker. Each took this branch's side, and then every marker was recomputed from the merged README.

  Each merge was clean or resolved to the sides RV630 predicted.

• Refresh round (61b3907):

  • step 3's Git Bash route on the Windows tab, with the guard test;
  • notes entries for the fifteen pull requests after 4cf5fb0;
  • the environments, the ruleset and the rehearsal.

• RV630 round 3: no MUST. SHOULD 1-2 and nits 1-5 were taken (ddd3c5f):

  • the handbook says only that the one line "runs too" in Git Bash, defines Git Bash once, and the Windows heading names the third way;
  • the notes' Git Bash known limit gives tests: a tests-git-bash job runs the launcher's and install.sh's suites in Git for Windows' own bash, so README and the handbook say Git Bash runs the one-line install, tested in CI but for a typed yes, which it cannot take there (#585) #596's ln -s fact: a copy unless MSYS=winsymlinks:nativestrict is set, which follows main for want of openRepoShape.ref;
  • the area headings fit their entries: "update-shape.py, bump-leg.py and a project's own scripts" and "scaffold, setup, family and the shared readers";
  • the YAML reader is the standard's own, in repo_shape.py;
  • the release acts say that lane 3 made the environments and the ruleset as brettheap on Brett Heap's word;
  • adopt-project.py: adopt-bad-id names an id derived from --project as derived, so --project North_Wind no longer says --id 'north_wind' must match over a flag nobody passed (#601) #620's entry.

• Final refresh, round A (7b7e657, 6fd33a9):

  • L4-47 (Tracking: adopt-project.py, converting an existing repository in place #457): README.md:771-772 and the handbook's § Adopting (:1276-1277) say "it KEEPS ITS NAME (which --project must be), ITS IDENTITY AND ITS FULL HISTORY, and becomes the assembly root", with no line added. Its Tracking: adopt-project.py, converting an existing repository in place #457 tick waits for README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630 to land.
  • Notes entries for adopt-project.py: the read-across follow-up names legs.code_path and legs.spec_path and no path, and the contracts/ rule's reason names none, so the hand edit of a leg's mount the tool asks for leaves no plan-time path in check, the split commit or the pull request (#600) #628 (with lane 2's re-run-plan line for a plan written before v1.1.0), adopt-project.py and scaffold-project.py: the Free-plan secrets hint goes by the assembly root's own visibility on GitHub and names the root as its reason, so a public source adopted with private legs is no longer told its organisation secrets resolve to EMPTY (#602) #621, bump-leg.py and setup-project.py: the git switch -c line says under it what to do where an earlier act left that branch, so already exists is no dead end (#607's sentence, in the two tools it left) #657, validate-family.py and family.py: a gitlink no section of HEAD's .gitmodules gives is member-gitmodules-absent, remove refuses a holder with no .gitmodules before deinit, and two gitlink remedies name what they missed (#611) #639, openRepoShape in Git Bash: the machine: row finds python, a drive-letter $OPENREPOSHAPE_REPO is a path, a Windows-spelled $OPENREPOSHAPE_BIN_DIR is respelled or refused, and tests-git-bash runs #598's installer cases (#649) #656, tests/test_doctor_next_is_a_command.py: the comment-tail guard reads each string expression whole, across + and an f-string's pieces, and flags a # led by two or more blanks or a tab with or without a blank after it and a # that opens text after a part that is not a constant, so cmd #x, cmd + " " + "# x" and f"{cmd}{pad}# x" no longer reach a NEXT unseen (#679) #698 and README.md and the handbook: say that a source's own single-repository.yaml record arrives as leg: drop, that the adopted Makefile's CONTRACTS_DIR follows the spec leg's mount, and that a second setup run is refused at the plan, so the three sentences #566, #572 and #573 left false or incomplete are true of main (#606) #627. Every pull request since bfed143 is now cited.
  • Known limits: lane 4's ten lines and lane 1's twenty-six D9 lines, and Developer Mode for a Git Bash symlink, as tests: a tests-git-bash job runs the launcher's and install.sh's suites in Git for Windows' own bash, so README and the handbook say Git Bash runs the one-line install, tested in CI but for a typed yes, which it cannot take there (#585) #596 says.
  • The trusted publishers, from Brett Heap's 13:04:43Z and 13:07:56Z words.

• Round A3 (interim, before the tag) (a86a422, d2d3d76, 2026-10-17):

  • Merged main bf1f099b: 25 pull requests, no conflict. README.md and the handbook were untouched by them, and every marker stayed current.
  • The doctor's exit 3 (W683, RV-f's N4 on shape-doctor.py and update-shape.py: run from a partial clone of the standard, the shape currency row asks without fetching whether the clone holds what the comparison reads, and is CANNOT ANSWER with a fetch for a person to run where it does not, and a read that fails is refused, never a copy called both (#683) #742): README.md:1208-1220 and the handbook's doctor twin now name every reason the checkout cannot answer, as shape-doctor.py's EXIT CODES does. The list covers a pin's commit it lacks, a partial clone missing what is compared (shape-doctor.py: run from a partial clone of the standard, the shape currency row fetches what the clone lacks, and with its remote out of reach calls each copy the standard changed since the pin both, so a doctor that fetches nothing fetches, and offline says DRIFTED of copies nobody edited #683), a read that fails, a crashed validator of its own (shape-doctor.py: validator_row reads a validator's exit 1 as "reports a finding" whatever caused it, so a validator that CRASHED (a : in the standard's path splits the PYTHONPATH the template fallback runs under, and validate-manifest.py dies on import repo_shape) is a FINDING row quoting ModuleNotFoundError, and a root whose manifest is valid goes from COMPLIANT, SHAPE BEHIND to INVALID (manifest) #278, shape-doctor.py's two naming rows read only the naming script's exit code, so a crash of the STANDARD's own validate-repository-naming.py is a FINDING about the project (INVALID (naming), exit 1, where #357 makes a crashed template CANNOT ANSWER), and a directory whose name is not UTF-8 gets naming FINDING Atl?as: UnicodeEncodeError: ..., the script's traceback quoted as a fact about the name #410), and its own naming policy not loading (shape-doctor.py: a checkout of the standard whose own contracts/repository-naming.yaml does not load makes a compliant project INVALID (naming), exit 1, "the naming policy is not satisfied (exit 2)", so a person is failed for the checkout's fault, and the module's EXIT CODES list leaves out an unfinished merge #678). This costs +1 line, recorded in tests/readme-cap.d/683.txt. The two Keeping a project's shape current markers move.
  • Notes entries for all 25: adopt-project.py: plan refuses --id '' and --elected-by '' by name, a derived id's refusal says to change --project, or in place to rename the source first, and an org read from a url is named as not given (#658) #699, adopt-project.py: a --work-dir that is a file or cannot be made is refused by name in plan, check and execute, and --help says what --work-dir is and where --org comes from (#650) #719, adopt-project.py: --source is read by its spelling, an org/repo in any spelling is cloned by its name, the plan records source.cloned, and a clone no plan records is removed, so a recorded clone that is gone or emptied is refused or cloned again for what it is (#652) #749, adopt-project.py: under LC_ALL=C PYTHONUTF8=0, plan, check and execute hand git and gh each name, url and branch and the leg description as the UTF-8 bytes it spells, so none of them ends in a UnicodeEncodeError before any leg exists (#653) #727, adopt-project.py: _default_branch keeps a branch name's byte that is not UTF-8, so a source whose HEAD or origin/HEAD is on caf<e9> is refused or adopted past it, and no longer ends plan, check and execute in a traceback (#425) #717, scaffold-project.py, adopt-project.py and family.py: the Free-plan secrets hint goes by what GitHub says of the root or holder and of its organisation's plan, and names the answer GitHub did not give, so init --reuse-empty-repo no longer calls a public holder private, an unread root is not taken as the run's visibility, and an unread plan is not silent #728, adopt-project.py: check asks this machine's git what the split stores for each file the plan keeps or moves and each pinned copy, so one it would store as other bytes no longer passes check and fails after both legs exist (#654) #743, adopt-project.py: a source symlink whose target begins with a NUL byte is check's plan-empty-symlink, which execute refuses before any leg, so \0spec.md no longer fails the first clone after both legs exist, or verifies over a spec leg no clone can check out (#388) #720, adopt-project.py: check takes --local-remote-dir and resolves each relative submodule url the plan keeps in a leg from the bare repositories execute --local-remote-dir would create there, so under that test path check no longer refuses an adoption execute verifies, or passes one execute refuses, over such a url (#417) #718, adopt-project.py: _verify fails a split whose root project.yaml is not the shape's manifest (adopt-manifest-not-shape), so a plan that gets past check's plan-root-blocks-manifest no longer ends adoption verified over an assembly the doctor calls NOT A SHAPE ROOT (#389) #722, adopt-project.py, shape-doctor.py: a plan's answer against a rule is recorded and an empty .gitmodules asks nothing, so docs/ answered code is not MISPLACED after adoption verified, nor a .gitmodules emptied by git rm asked on every run (#655) #721 (adopt); validate-pins.py and shape-doctor.py: one leg's gitlink off its pin beside another leg's digest or workflow reference names the step that comes first, that gitlink put back by hand, and then the other leg's bump-leg.py line, so neither finding says "First clear that finding" over a finding no line can clear first (#677) #702, Pinned shape copies: one bump, so AGENTS-shape.md is true of every root, the scaffold's chain WARNING offers only its own sources and spells --pin-source, the root's not-used note spells it too, and THE LOCKSTEP RULE's bump runs (#610) #634 (update-shape, bump-leg and a project's own scripts); adopt-project.py and scaffold-project.py: a --pin naming an annotated tag's own id pins the commit the tag names, and one naming a tree or a blob is refused, so a pin file no longer records a tag's id under "A COMMIT, NEVER A TAG" (#651) #725, scaffold-project.py and family.py: scaffold-bad-id and family-bad-id name an id derived from --project or --family as derived and say to change that flag, so --project North_Wind no longer says --id 'north_wind' must match and sends the person to an --id the naming policy then refuses (#659) #671, validate-manifest.py, the naming gate and setup-project.py: a chain link no tree answers for is offered what each caller reads, so the root's validator and setup's step (4) no longer offer --link-source, which neither takes, and the gate spells every --link-source value and path it prints #744 (scaffold, setup, family and the shared readers); shape-doctor.py and update-shape.py: run from a partial clone of the standard, the shape currency row asks without fetching whether the clone holds what the comparison reads, and is CANNOT ANSWER with a fetch for a person to run where it does not, and a read that fails is refused, never a copy called both (#683) #742, shape-doctor.py: a checkout whose own contracts/repository-naming.yaml does not load is CANNOT ANSWER, exit 3, for a project, where it read INVALID (naming), exit 1; any other naming exit 2 "refuses"; and the EXIT CODES list names an unfinished merge (#678) #701, shape-doctor.py: an own-copy validator row whose crash is a root's lost scripts/repo_shape.py names git -C <root> checkout HEAD -- scripts/repo_shape.py as its NEXT (none where HEAD no longer holds it) and the validator in its WHY (#697) #737, shape-doctor.py: the pins row keeps its gate where update-shape.py check says REFUSAL AHEAD beside a merged-away row, so it no longer says that the apply check prints drops the row when that apply is refused, and the shape currency counts name every state check counts (#676) #751, shape-doctor.py: a conflicted shape pin's NEXT reads both sides' pins, so a copy conflicted beside it that only the side not taken has a files: row for is named in its why and left_for_a_person for a person to merge, and never taken (#687) #736, shape-doctor.py: the placement row asks a leg's top-level single-repository.yaml the staying-single record's question only where kind: makes it the record, so one declaring release-config is asked the general question as plan asks it, and an answered entry names that rule (#691) #734, shape-doctor.py: submodule_config and gitmodules_entry read a holder's .gitmodules through one git config -z --list, so a section name holding a space or a byte that is not UTF-8, or a url holding NEL or U+2028, gives the doctor the mount url git and make siblings use, under any locale (#668) #700, shape-doctor.py: the members row says, member by member, what make siblings does with the directory beside the holder, in its own word and in place's order, so it no longer says it places a member it leaves exactly as it is, and it spells every value it quotes, so an ESC or a CR in a url no longer reaches the terminal (#669, #673) #705, openRepoShape --doctor: a signal sent to the command's shell alone waits for the doctor, so its clone of the standard is never removed while shape-doctor.py runs from it, and a doctor ended by a signal leaves nothing (#680) #740 (the doctor). There is also one docs line for the exit-3 sentence.
  • Counts, the open list (26) and the "current to main" sha are refreshed to bf1f099.
  • Waits for round B:
    • lane 5's eight lines (#630 issuecomment-6227129879), whose pull requests install.sh and openRepoShape: a Windows-spelled $OPENREPOSHAPE_BIN_DIR is respelled or refused before install.sh's fetch, and the PATH note quotes the directory it prints and refuses one holding : or a newline (#667) #726, validate-family.py and validate-pins.py: a staged gitlink removal says to commit it, member-gitlink-absent puts the gitlink back by hand, the variant mv is guarded, and a pin row with no path: is shape-pin-row (#681) #747, bootstrap.py and validate-family.py: a member no credential would check out fails the family bootstrap, every row's gitlink is asked before any checkout, and --no-members prints the case-variant move (#648) #682, render-cli-reference.py and tests/: the CLI reference's children write no bytecode under templates/, the launcher guard reads family.py's siblings child, the quoting-sentence inventory follows an aliased quoter, a suffixless tool and each tool's calling functions, and an S5778 pytest.raises builds its argument first (#665) #724, bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746, bootstrap.py and AGENTS-shape.md: a leg with an operation in progress or an ignored file in the way is refused, P9 names the branch it left, and no git is git-not-runnable, so make bootstrap loses nothing (#704) (#694) #745 and README.md, the handbook, AGENTS.md and the family-root docs: init's --into sentence names the family-landing-in-standard refusal, make siblings is described as it stops before cloning, and members/ has its one-spelling rule (#695) #738 are open drafts;
    • shape-doctor.py: the pins row keeps its gate where update-shape.py check says REFUSAL AHEAD beside a merged-away row, so it no longer says that the apply check prints drops the row when that apply is refused, and the shape currency counts name every state check counts (#676) #751's tick of Tracking: update-shape.py and bump-leg.py, keeping a project's shape copies and leg pins current #458's line (issuecomment-6158353941, "→ shape-doctor.py: the pins row says check's apply drops merged-away rows where check says REFUSAL AHEAD (a pinned copy kept as a link), and the shape currency counts leave out five of check's states #676"), a GitHub write this round does not make;
    • L4-47's tick on Tracking: adopt-project.py, converting an existing repository in place #457, when README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630 lands;
    • the counts, the open list and v1.0.0's known limits, re-read at tag time.

• Round A4 (lane 2's review) + B1 (landed since bf1f099) (d6a8556, b2a62ec, 2026-10-17):

  • Merged main 20004887, with no conflict. bootstrap.py and validate-family.py: a member no credential would check out fails the family bootstrap, every row's gitlink is asked before any checkout, and --no-members prints the case-variant move (#648) #682's README sentence and its handbook twin merged as text with their own Families marker.
  • A4, lane 2's review of A3 (issuecomment-6233033833, READY, three SHOULDs and four nits, all taken):
    • SHOULD 1: the exit-3 list says "a validator of its own that crashed or could not be run", in README.md:1208-1222, the handbook's doctor and the notes' docs line. This covers the case where the validator cannot be given its PYTHONPATH (shape-doctor.py: validator_row reads a validator's exit 1 as "reports a finding" whatever caused it, so a validator that CRASHED (a : in the standard's path splits the PYTHONPATH the template fallback runs under, and validate-manifest.py dies on import repo_shape) is a FINDING row quoting ModuleNotFoundError, and a root whose manifest is valid goes from COMPLIANT, SHAPE BEHIND to INVALID (manifest) #278). tests/readme-cap.d/683.txt is now +2, and its text says which causes come from shape-doctor.py and update-shape.py: run from a partial clone of the standard, the shape currency row asks without fetching whether the clone holds what the comparison reads, and is CANNOT ANSWER with a fetch for a person to run where it does not, and a read that fails is refused, never a copy called both (#683) #742 and which from shape-doctor.py: a checkout whose own contracts/repository-naming.yaml does not load is CANNOT ANSWER, exit 3, for a project, where it read INVALID (naming), exit 1; any other naming exit 2 "refuses"; and the EXIT CODES list names an unfinished merge (#678) #701. Both Keeping a project's shape current markers are re-hashed.
    • SHOULD 2: adopt-project.py: under LC_ALL=C PYTHONUTF8=0, plan, check and execute hand git and gh each name, url and branch and the leg description as the UTF-8 bytes it spells, so none of them ends in a UnicodeEncodeError before any leg exists (#653) #727's entry names the leg description and its em dash.
    • SHOULD 3: adopt-project.py: --source is read by its spelling, an org/repo in any spelling is cloned by its name, the plan records source.cloned, and a clone no plan records is removed, so a recorded clone that is gone or emptied is refused or cloned again for what it is (#652) #749's entry names source-ambiguous.
    • Nits:
      • the doctor's three near-duplicate pairs are merged;
      • adopt-project.py: check takes --local-remote-dir and resolves each relative submodule url the plan keeps in a leg from the bare repositories execute --local-remote-dir would create there, so under that test path check no longer refuses an adoption execute verifies, or passes one execute refuses, over such a url (#417) #718 has its own bullet;
      • Pinned shape copies: one bump, so AGENTS-shape.md is true of every root, the scaffold's chain WARNING offers only its own sources and spells --pin-source, the root's not-used note spells it too, and THE LOCKSTEP RULE's bump runs (#610) #634's entry gains git_out's spelling, and validate-manifest.py, the naming gate and setup-project.py: a chain link no tree answers for is offered what each caller reads, so the root's validator and setup's step (4) no longer offer --link-source, which neither takes, and the gate spells every --link-source value and path it prints #744's the gate's --link-source-label and --link-remedy;
      • the header comment is current.
  • B1: notes entries for tests: validate-pins.py's workflow scan skips a directory, a dangling symlink and a looping symlink named .yml, and both of siblings.py's headers print the holder's root and folder spelled, so the S_ISREG child check made lambda mode: True, or either path printed raw, now fails a test (#664) #706 (tests: two mutants survive review: the S_ISREG child check of validate-pins.py's workflow scan (a directory named sub.yml, a dangling and a looping symlink, #434) and siblings.py's holder-root spelling in both headers (#454) #664), setup-project.py: a family folder is named after the holder's REPOSITORY, so the Quill/Quill that family.py init --name QuillWorks makes is detected with no flag, and the QuillWorks/QuillWorks that make siblings refuses is not (#696) #703 (setup-project.py: family_folder_here reads the family's DISPLAY name, so the layout family.py init makes for a family created with --name is not detected as a family folder and the old display-name layout is #696), family.py: add's refusal of a member's own gitlink held only under another spelling of its row's path moves it back to that path in the index alone, so the printed remedy no longer takes the member's gitlink away from its row (#646) #670 (family.py add: where another member's own gitlink sits at a case variant (Members/<P> for its row members/<P>), the refusal's remedy prints git rm -r --cached or git mv, which takes that member's gitlink away from its row #646), bootstrap.py and validate-family.py: a member no credential would check out fails the family bootstrap, every row's gitlink is asked before any checkout, and --no-members prints the case-variant move (#648) #682 (bootstrap.py and validate-family.py: family bootstrap ok with no member checked out, an earlier row's member-uninitialized hiding a case-variant refusal, and member-gitlink-unnamed with no command under --no-members #648), install.sh and openRepoShape: a Windows-spelled $OPENREPOSHAPE_BIN_DIR is respelled or refused before install.sh's fetch, and the PATH note quotes the directory it prints and refuses one holding : or a newline (#667) #726 (install.sh and openRepoShape: a Windows-spelled $OPENREPOSHAPE_BIN_DIR is respelled in install.sh's own lines and refused before its fetch, and the PATH note quotes the directory it prints (#649's two leftovers) #667), adopt-project.py, shape-doctor.py and tests/test_shape_doctor.py: the verification row's colon, run_validator's "costs no child an import" and the decoy test's sitecustomize claim say what main does (#666) #730 (Comments that say what main no longer does: submodule_config's reason, run_validator's "costs no child an import", the decoy test's sitecustomize claim under -S, and the verification row's colon in adopt-project.py's header #666) and render-cli-reference.py and tests/: the CLI reference's children write no bytecode under templates/, the launcher guard reads family.py's siblings child, the quoting-sentence inventory follows an aliased quoter, a suffixless tool and each tool's calling functions, and an S5778 pytest.raises builds its argument first (#665) #724 (tests: four guards that do not hold what they say: a .pyc the CLI reference's child writes under templates/, a launcher guard blind to family.py, a quoting-sentence inventory with three holes, and an S5778 pytest.raises #665).
    • install.sh and openRepoShape: a Windows-spelled $OPENREPOSHAPE_BIN_DIR is respelled or refused before install.sh's fetch, and the PATH note quotes the directory it prints and refuses one holding : or a newline (#667) #726's and bootstrap.py and validate-family.py: a member no credential would check out fails the family bootstrap, every row's gitlink is asked before any checkout, and --no-members prints the case-variant move (#648) #682's come from lane 5's lines. bootstrap.py and validate-family.py: a member no credential would check out fails the family bootstrap, every row's gitlink is asked before any checkout, and --no-members prints the case-variant move (#648) #682's adds the unmerged .gitmodules from its last round and says that ok became exit 1.
    • render-cli-reference.py and tests/: the CLI reference's children write no bytecode under templates/, the launcher guard reads family.py's siblings child, the quoting-sentence inventory follows an aliased quoter, a suffixless tool and each tool's calling functions, and an S5778 pytest.raises builds its argument first (#665) #724's says that v1.0.0's test-hygiene line on the generator's .pyc files closes.
    • The counts (106 pull requests; 116 of 135 issues closed at 11:53Z), the open list (19) and "current to main" are at 2000488.
  • Waits for B2:
    • lane 5's lines for validate-family.py and validate-pins.py: a staged gitlink removal says to commit it, member-gitlink-absent puts the gitlink back by hand, the variant mv is guarded, and a pin row with no path: is shape-pin-row (#681) #747 (validate-family.py: a staged git rm is named again and fails retyped, the member-gitlink-absent remedy ends in a refused add, a shape-pin row with no path: reads None, and #639's mv orphans a live checkout and is stated by OS #681; v1.0.0.md:466 and :467 close), bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746 (bump-leg.py: a root on +x is told push -u origin +x (forced), a branch named like a tag passes bump-leg-on-tracking-branch, bump-leg-root-dirty prints C1 controls raw, and committed <sha>: lists paths the commit lacks #672; :433 closes for bump-leg.py), bootstrap.py and AGENTS-shape.md: a leg with an operation in progress or an ignored file in the way is refused, P9 names the branch it left, and no git is git-not-runnable, so make bootstrap loses nothing (#704) (#694) #745 (bootstrap.py (the family root's and the assembly root's): make bootstrap with no git on PATH ends in a FileNotFoundError traceback, exit 1, where the validators refuse git-not-runnable, exit 2 #694, bootstrap.py: moving a leg off a branch overwrites an ignored file and ends a merge, P9 does not name the branch it leaves, a branch at its upstream's tip is left when HEAD is on a branch, a HEAD outside refs/heads/ is read two ways #704; :463 closes) and README.md, the handbook, AGENTS.md and the family-root docs: init's --into sentence names the family-landing-in-standard refusal, make siblings is described as it stops before cloning, and members/ has its one-spelling rule (#695) #738 (README.md, the handbook, AGENTS.md and the holder's AGENTS-shape.md: init's --into sentence omits the landing refusal, make siblings is described as it was before #578, and nothing names the one-spelling rule for members/ #695), each only if it lands;
    • every other pull request that lands after 2000488;
    • shape-doctor.py: the pins row keeps its gate where update-shape.py check says REFUSAL AHEAD beside a merged-away row, so it no longer says that the apply check prints drops the row when that apply is refused, and the shape currency counts name every state check counts (#676) #751's tick of Tracking: update-shape.py and bump-leg.py, keeping a project's shape copies and leg pins current #458's line and L4-47's tick on Tracking: adopt-project.py, converting an existing repository in place #457;
    • the counts, the open list and v1.0.0's known limits, re-read at tag time.

Round B2 (final), part 1: landed since b2a62ec

Head d697b366e2e2a25a62c5db60acbee992581eb351, on main fc867207, pushed as a fast-forward from b2a62ec7 on 2026-10-17. Part 2 waits only for lane 4's #795 (#686).

• Commits. Each merge of main was clean. None of the 15 pull requests touches README.md, the handbook, AGENTS.md, docs/ or the README cap's ledger, so every handbook marker reads as before.

  • 441dcb41 merges main 40e678e6; 6a6817cd takes RV630's four nits and five entries.
  • 8049456b merges main 64ec47c3; cd56ed78 takes RV630's pre-review of the stage, with validate-family.py and validate-pins.py: a staged gitlink removal says to commit it, member-gitlink-absent puts the gitlink back by hand, the variant mv is guarded, and a pin row with no path: is shape-pin-row (#681) #747's and repo_shape.py, shape_materialize.py and adopt-project.py: a value the YAML reader refuses for opening with an indicator is named by its line and says to double-quote it or reword it, plan --org @x names --org again under the reader's own remedy, and check names one bad tracking_branch once (#708) #817's entries.
  • 42d4df8c merges main bad05d21; 1aeed03c adds seven entries and the counts.
  • 4d14e4a7 merges main fc867207; d697b366 adds bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746's entry, makes two markers final, names the triage lines by their issues, and refreshes the counts.

• Inserted, each from its squash on main. For all 15, git merge-tree --write-tree <main before the squash> <the pull request's final head> exits 0 and equals the squash's tree:

  PR	closes	squash	tree	where in the notes
  tests/conftest.py: the session turns git's automatic maintenance off, in command scope and in a global config that includes the person's own, so a copytree of a fixture no longer meets a detached repack deleting its loose objects (#645) #713	tests: the holder fixtures copy the family fixture with shutil.copytree while git's background gc or maintenance can still be packing it, so a test errors at setup when a loose object vanishes mid-copy (#440's race, through a file copy) #645	741679e7	760892f6	the CI bullet
  family.py and scaffold-project.py: add --local-remote-dir refuses a member clone URL .gitmodules cannot hold as text, --dry-run alike, each clone-URL remedy names a link's target and a lone surrogate, and the READMEs quote the clone URL, so the line under a b/ pastes as one command (#693) #812	--local-remote-dir: family.py add takes a directory whose name is not UTF-8 into .gitmodules, the clone-URL refusal's remedy misses a link's target and a lone surrogate, and the READMEs' clone line is rendered unquoted #693	cbff780c	d8fc9c37	--local-remote-dir
  family.py add: a members/ renamed with git mv is renamed back in two steps through a free name, after what the disk lacks under it is put back, and a printed --sparse says the git it needs, so no remedy fails as typed (#685) #767	family.py add: #574's spelling remedy gives a generic <new name> after git mv members Members, git rm --cached for a pulled Members/README, and a git mv --sparse that needs git 2.34 or later #685	98f14976	2fd92cfb	a new bullet: a members/ renamed with git mv
  bootstrap.py and AGENTS-shape.md: a leg with an operation in progress or an ignored file in the way is refused, P9 names the branch it left, and no git is git-not-runnable, so make bootstrap loses nothing (#704) (#694) #745	bootstrap.py: moving a leg off a branch overwrites an ignored file and ends a merge, P9 does not name the branch it leaves, a branch at its upstream's tip is left when HEAD is on a branch, a HEAD outside refs/heads/ is read two ways #704, bootstrap.py (the family root's and the assembly root's): make bootstrap with no git on PATH ends in a FileNotFoundError traceback, exit 1, where the validators refuse git-not-runnable, exit 2 #694	24a977c0	3b3e8923	Bootstrap
  siblings.py: the printed move runs as typed past a taken staging name and names a dangling link to move aside first, a NUL or lone surrogate in project: or url is a finding, .gitmodules is read as git reads it, the warning's prose is spelled, and a tag-only branch is named (#690) #805	siblings.py (make siblings): the printed mv goes astray over an existing .holder/.folder or a dangling link, a lone surrogate or NUL in project: is a traceback, a url with NEL or U+2028 is cut, two warnings print paths raw #690	40e678e6	d8ac1bb9	a new bullet: make siblings's inputs
  validate-family.py and validate-pins.py: a staged gitlink removal says to commit it, member-gitlink-absent puts the gitlink back by hand, the variant mv is guarded, and a pin row with no path: is shape-pin-row (#681) #747	validate-family.py: a staged git rm is named again and fails retyped, the member-gitlink-absent remedy ends in a refused add, a shape-pin row with no path: reads None, and #639's mv orphans a live checkout and is stated by OS #681	6626a3c5	994e7168	a new bullet: validate-family.py's remedies, typed as printed
  repo_shape.py, shape_materialize.py and adopt-project.py: a value the YAML reader refuses for opening with an indicator is named by its line and says to double-quote it or reword it, plan --org @x names --org again under the reader's own remedy, and check names one bad tracking_branch once (#708) #817	v1.1.0 regressions: adopt-project.py check exits 2 with no line and no remedy on a hand-written resolution: that opens with a backtick (#635), plan --org @x no longer names --org (#635), and check names one bad tracking_branch twice (#644) #708	64ec47c3	53515a95	the YAML reader
  tests/conftest.py: one helper builds the url.<file URI>.insteadOf rewrite that clone_just_pushed, read_through_git and the adopt tests' pasted each built, and the empty-mount submodule update --init sites use it (#662) #748	tests/conftest.py: one helper builds the url.<base>.insteadOf rewrite that clone_just_pushed and read_through_git each build, and the plain-path git submodule update --init sites of #554's review use it (#440) #662	431c8cf8	bc67aa88	the CI bullet
  shape_materialize.py, scaffold-project.py and family.py: the folder refusals and family-operation-in-progress name a path as shown_source names it, so an ESC or a line break in a path no longer reaches the terminal (#692) #731	scaffold-project.py, family.py and shape_materialize.py: #622's folder refusals print a folder name raw, so an ESC or a line break reaches the terminal, and family-operation-in-progress prints a holder's path across two lines #692	d66a9054	43279472	Folders, asked before they are made
  update-shape.py: apply's rollback line says to make a shut directory writable first, committed on <branch>: names what the commit holds, a branch named like a tag is pushed and read by its full name, and a quoted files: or shape: header is found as the reader finds it (#674) #810	update-shape.py apply: a rollback line fails over a read-only contracts/ unsaid, committed on <branch>: lists a path the commit lacks, --branch v1.0.0 cannot push beside a tag v1.0.0, and a quoted "files": header is refused #674	d17564d9	c05b99b4	a new bullet: apply's lines, its branch and its headers
  family.py init: the creator is asked whether a file can hold it whichever of --created-by, GIT_AUTHOR_NAME or git config user.name gave it, git is asked about the first commit's author and committer before the remote exists, and a --family that is a Windows device name is refused, --dry-run alike (#688) #809	family.py init: --created-by and GIT_AUTHOR_NAME are never asked whether the manifest can hold them, a whitespace-only GIT_AUTHOR_NAME fails after the remote exists, and a --family a Windows disk cannot hold exactly is accepted #688	42aed788	56416599	a new bullet: what family.py init records and commits as
  family.py: add, bump and remove name a bisect, an empty rebase directory, sparse rules that hold no pattern, a merge conflicted in family.yaml and a members file HEAD holds and the work tree does not as what each is, with the way out that works there, and refuse a family.yaml holding a change HEAD does not (#684) #816	none; part of family.py add/bump/remove: a members file deleted unstaged, a git bisect, an empty sparse-checkout file, an empty rebase-merge/ and a merge conflicted in family.yaml meet a refusal or remedy that does not fit (#626's gaps) #684	fa04d2cd	409e6882	a new bullet: more holder states, named
  family.py and shape_materialize.py: the NEXT push line names the branch by its full refname on both sides, so a branch beginning with + or named like a tag pushes as itself, and a tracking branch whose last part is over 250 bytes, which git cannot lock, is refused by refuse_bad_branch before anything exists, --dry-run too (#689) #813	family.py: the NEXT push line is a forced refspec for a branch beginning with + and fails for a branch named like a tag, and a tracking branch of 251-255 characters passes --dry-run, dies at the first commit once every remote exists #689	f6d17bc5	786b3d6d	--tracking-branch
  tests: one skip helper for names and branches that are not UTF-8, Windows guards on make_file and the --branch a\udcffb test by emulation, and the UTF-8 promise and the leg finding's role pinned (#663) #723	tests: the non-UTF-8 name and branch tests: one skip helper for a filesystem that refuses probe\udcff, Windows guards on make_file and the --branch a\udcffb case, and the UTF-8 promise and the leg finding's role pinned #663	bad05d21	bf8dda2a	the CI bullet
  bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746	bump-leg.py: a root on +x is told push -u origin +x (forced), a branch named like a tag passes bump-leg-on-tracking-branch, bump-leg-root-dirty prints C1 controls raw, and committed <sha>: lists paths the commit lacks #672	fc867207	bb920ac1	a new bullet: bump-leg.py's branch and its lines

  • The last eight entries are the pending texts RV630 pre-reviewed, inserted as written. For seven of them, the head each text was written from gives the squash's tree too.
  • bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746's text was written from 4a014a79, which sat on tests: one skip helper for names and branches that are not UTF-8, Windows guards on make_file and the --branch a\udcffb test by emulation, and the UTF-8 promise and the leg finding's role pinned (#663) #723's head before tests: one skip helper for names and branches that are not UTF-8, Windows guards on make_file and the --branch a\udcffb test by emulation, and the UTF-8 promise and the leg finding's role pinned (#663) #723's squash and does not merge clean with main. scripts/bump-leg.py is one blob (226e7b0d) at 4a014a79, at a5139a32 and in the squash, and the squash's own lines in its other four files equal that head's, so the text stands.
  • README.md, the handbook, AGENTS.md and the family-root docs: init's --into sentence names the family-landing-in-standard refusal, make siblings is described as it stops before cloning, and members/ has its one-spelling rule (#695) #738 (README.md, the handbook, AGENTS.md and the holder's AGENTS-shape.md: init's --into sentence omits the landing refusal, make siblings is described as it was before #578, and nothing names the one-spelling rule for members/ #695) is v1.1.1. Its docs entry is not taken. Its upgrade note (cd Y → cd X, with the README's Y/Y layout) is, because it is true of main since siblings.py and the family-root README: make siblings, park and resume print the family's name and every member row spelled and never end in a traceback on name: "a\ud83db" or a dash under PYTHONIOENCODING=ascii, and the holder's README names its folder after the repository, so --family Quill --name QuillWorks lays out as its README says (#454, #629) #642.

• RV630's pre-review items taken:

  • on A4 + B1 (issuecomment-6233641684), four nits: README's one-word "finished." line (README back to 1474, 683.txt +1, both Keeping a project's shape current markers re-hashed); bootstrap.py and validate-family.py: a member no credential would check out fails the family bootstrap, every row's gitlink is asked before any checkout, and --no-members prints the case-variant move (#648) #682's lead; adopt-project.py, shape-doctor.py and tests/test_shape_doctor.py: the verification row's colon, run_validator's "costs no child an import" and the decoy test's sitecustomize claim say what main does (#666) #730's "Comments and docstrings"; the doctor's members row as one bullet;
  • on the stage: SHOULD-1, the CI bullet says "runs git clone on" a just-pushed local remote, which is what tests: every test that clones a just-pushed local remote by its plain path now clones it with clone_just_pushed, so the macOS failed to copy file race of #440 cannot reach it; load_without_bytecode keeps the doctor's and the Windows-path tests' loads of the family root's scripts from writing .pyc into the templates; the not-a-list doctor cases hold INVALID, exit 1; and GOOD_KEYS holds th #554 changed; NIT-1, family.py and scaffold-project.py: add --local-remote-dir refuses a member clone URL .gitmodules cannot hold as text, --dry-run alike, each clone-URL remedy names a link's target and a lone surrogate, and the READMEs quote the clone URL, so the line under a b/ pastes as one command (#693) #812's "that needs it"; NIT-2, bootstrap.py and validate-family.py: a member no credential would check out fails the family bootstrap, every row's gitlink is asked before any checkout, and --no-members prints the case-variant move (#648) #682's first sentence; and its corrections to the pending entries for repo_shape.py, shape_materialize.py and adopt-project.py: a value the YAML reader refuses for opening with an indicator is named by its line and says to double-quote it or reword it, plan --org @x names --org again under the reader's own remedy, and check names one bad tracking_branch once (#708) #817, tests/conftest.py: one helper builds the url.<file URI>.insteadOf rewrite that clone_just_pushed, read_through_git and the adopt tests' pasted each built, and the empty-mount submodule update --init sites use it (#662) #748 ("With tests: every test that clones a just-pushed local remote by its plain path now clones it with clone_just_pushed, so the macOS failed to copy file race of #440 cannot reach it; load_without_bytecode keeps the doctor's and the Windows-path tests' loads of the family root's scripts from writing .pyc into the templates; the not-a-list doctor cases hold INVALID, exit 1; and GOOD_KEYS holds th #554 and siblings.py and family.py: make siblings asks about the family folder before it clones, clones each member on its tracking branch, and family.py siblings runs the holder's own copy (#543) #578"), family.py: add, bump and remove name a bisect, an empty rebase directory, sparse rules that hold no pattern, a merge conflicted in family.yaml and a members file HEAD holds and the work tree does not as what each is, with the way out that works there, and refuse a family.yaml holding a change HEAD does not (#684) #816 ("non-cone") and README.md, the handbook, AGENTS.md and the family-root docs: init's --into sentence names the family-landing-in-standard refusal, make siblings is described as it stops before cloning, and members/ has its one-spelling rule (#695) #738 (v1.1.1).

• v1.0.0's known limits. Said closed: :463 (bootstrap.py and AGENTS-shape.md: a leg with an operation in progress or an ignored file in the way is refused, P9 names the branch it left, and no git is git-not-runnable, so make bootstrap loses nothing (#704) (#694) #745), :466 and :467 (validate-family.py and validate-pins.py: a staged gitlink removal says to commit it, member-gitlink-absent puts the gitlink back by hand, the variant mv is guarded, and a pin row with no path: is shape-pin-row (#681) #747), :439 (update-shape.py: apply's rollback line says to make a shut directory writable first, committed on <branch>: names what the commit holds, a branch named like a tag is pushed and read by its full name, and a quoted files: or shape: header is found as the reader finds it (#674) #810), :433 as one clause in bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746's entry ("With update-shape.py: apply's rollback line says to make a shut directory writable first, committed on <branch>: names what the commit holds, a branch named like a tag is pushed and read by its full name, and a quoted files: or shape: header is found as the reader finds it (#674) #810, that closes…"), and :530 (tests/conftest.py: one helper builds the url.<file URI>.insteadOf rewrite that clone_just_pushed, read_through_git and the adopt tests' pasted each built, and the empty-mount submodule update --init sites use it (#662) #748, with tests: every test that clones a just-pushed local remote by its plain path now clones it with clone_just_pushed, so the macOS failed to copy file race of #440 cannot reach it; load_without_bytecode keeps the doctor's and the Windows-path tests' loads of the family root's scripts from writing .pyc into the templates; the not-a-list doctor cases hold INVALID, exit 1; and GOOD_KEYS holds th #554 and siblings.py and family.py: make siblings asks about the family folder before it clones, clones each member on its tracking branch, and family.py siblings runs the holder's own copy (#543) #578). Still open: :492, listed (README.md, the handbook, AGENTS.md and the holder's AGENTS-shape.md: init's --into sentence omits the landing refusal, make siblings is described as it was before #578, and nothing names the one-spelling rule for members/ #695 is v1.1.1), and :440, which bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746's entry says the dry run now names.

• Markers made final in the file. "What changed" now reads "The list runs to main fc86720; family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795, if it lands next, is added before the tag." Known limits' re-read marker and "as read so far" are cut. The counts marker keeps the sentence the release-body tool fills.

• Public wording in Known limits. "lane openRepoShape-4's checklist triage" is now "the checklist triage of the L4 lines on Tracking: adopt-project.py, converting an existing repository in place #457 and Tracking: setup, scaffold, family and the shared helpers, creating projects and families #459", and "lane openRepoShape-1's checklist triage (D9)" is "the checklist triage of the L1 lines on Tracking: adopt-project.py, converting an existing repository in place #457". Each id reads "Tracking: adopt-project.py, converting an existing repository in place #457's L4-05", "Tracking: setup, scaffold, family and the shared helpers, creating projects and families #459's L4-24", "Tracking: adopt-project.py, converting an existing repository in place #457's L1-02" and so on.

• Counts, in the four places: main fc86720; 121 pull requests since bfed143; at 2026-10-17T19:08Z, 131 of the v1.1.0 milestone's 133 issues closed. Open: Publish openRepoShape to TestPyPI and PyPI as openreposhape, so one pipx install puts the tools and git-filter-repo on any machine (design first) #563 (README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630) and family.py bump: a mount with a narrowed fetch refspec is refused member-to-local-only on a false detail, a commit only a tag holds is told "a pin the rest of the world cannot fetch", a mount with no origin is told to fix the network #686 (family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795). family.py add/bump/remove: a members file deleted unstaged, a git bisect, an empty sparse-checkout file, an empty rebase-merge/ and a merge conflicted in family.yaml meet a refusal or remedy that does not fit (#626's gaps) #684 (family.py: add, bump and remove name a bisect, an empty rebase directory, sparse rules that hold no pattern, a merge conflicted in family.yaml and a members file HEAD holds and the work tree does not as what each is, with the way out that works there, and refuse a family.yaml holding a change HEAD does not (#684) #816 fixed all but its reftable case) and README.md, the handbook, AGENTS.md and the holder's AGENTS-shape.md: init's --into sentence omits the landing refusal, make siblings is described as it was before #578, and nothing names the one-spelling rule for members/ #695 (README.md, the handbook, AGENTS.md and the family-root docs: init's --into sentence names the family-landing-in-standard refusal, make siblings is described as it stops before cloning, and members/ has its one-spelling rule (#695) #738) are on v1.1.1.

• Release paths. git diff --stat 20004887..fc867207 -- pyproject.toml src/openreposhape/ install.sh .github/workflows/ is empty. bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746's squash touches scripts/bump-leg.py and four test files, and no workflow.

• Tests, under a stub gh exiting 97, GH_TOKEN=bogus and an empty GH_CONFIG_DIR: the eight docs files gave 300 passed in 37.04s at d697b366 (and 300 passed in 34.82s at 1aeed03c). render-cli-reference.py --check gave [ok], and git diff --check origin/main was clean. README.md is 1474 lines, its cap: 1422 + 2 (460.txt) + 5 (550.txt) + 12 (563.txt) + 22 (564.txt) + 10 (606.txt) + 1 (683.txt) = 1474. AGENTS.md is 494 lines, its cap.

• Part 2, once family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795 lands: merge main; family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795's entry before "family.py refuses before it writes."; "What changed" becomes "The list is complete to the release commit."; the counts; the tests; the push.

• Not done here: the ticks on Tracking: update-shape.py and bump-leg.py, keeping a project's shape copies and leg pins current #458 (shape-doctor.py: the pins row keeps its gate where update-shape.py check says REFUSAL AHEAD beside a merged-away row, so it no longer says that the apply check prints drops the row when that apply is refused, and the shape currency counts name every state check counts (#676) #751's line) and Tracking: adopt-project.py, converting an existing repository in place #457 (L4-47), and any milestone move. Those are lane 3's coordinator's.

Round B2 (final), part 2: #795

Head aef568f02a9c626c9ae93f67e1a508c14684312d, on main 6f64eea2, pushed as a fast-forward from d697b366 on 2026-10-17. RV630 reviewed part 1 (d697b366) READY; this part makes the changes it gave, and nothing else.

• Commits. 016599b0 merges main 6f64eea2, clean: family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795 touches no README.md, handbook, AGENTS.md, docs/ or README cap ledger file, so every handbook marker reads as before. aef568f0 changes only docs/release-notes/v1.1.0.md (+22 -10).
• family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795 (family.py bump: a mount with a narrowed fetch refspec is refused member-to-local-only on a false detail, a commit only a tag holds is told "a pin the rest of the world cannot fetch", a mount with no origin is told to fix the network #686), squash 6f64eea2, checked against the squash itself. git merge-tree --write-tree fc867207 <head> exits 0 and equals the squash's tree 0f2d2ddd for the final head cf143c69 and for 17bbbaba. For cf5e64b3, the head the entry was written from, the merged tree differs from the squash only in tests/test_family_bump_reachability_words.py (+94 -21). scripts/family.py is one blob (a7660e9c) in the squash and in both later heads, so the entry stands as written.
• The three text changes:

  1. family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795's entry, a new bullet before "family.py refuses before it writes.":

     bump --to, reachable or not. family.py bump fetches every branch of the member's origin, whatever fetch refspec the mount has, and prunes the remote-tracking branches origin has deleted, so a commit a branch holds is no longer refused for a refspec narrowed by hand, and one only a deleted branch held is no longer pinned. A commit only a tag of the remote reaches is told that a tag is not a branch, and a mount with no remote named origin is told so, where each was told the pin cannot be fetched, or to fix the network (family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795, family.py bump: a mount with a narrowed fetch refspec is refused member-to-local-only on a false detail, a commit only a tag holds is told "a pin the rest of the world cannot fetch", a mount with no origin is told to fix the network #686).

  2. "What changed" (line 120): "The list runs to main fc86720; family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795, if it lands next, is added before the tag." is now "The list runs to main 6f64eea."

  3. Known limits, RV630's lead and list, exactly:

     v1.0.0's known limits that this release does not close still hold; this list is not all of them. Two that work on this release named and left open:

     • The README's and the handbook's --into sentence for family.py init gives its default, the directory you are standing in, but not that init refuses that default inside a checkout of the standard (family-landing-in-standard) (README.md, the handbook, AGENTS.md and the holder's AGENTS-shape.md: init's --into sentence omits the landing refusal, make siblings is described as it was before #578, and nothing names the one-spelling rule for members/ #695).
     • bump-leg.py --dry-run over a pin its own validators already refuse, such as one with a stray tab line, exits 0 and prints the old -> new line, where the real run exits 2 bump-leg-validators-red and rolls back. The dry run now says that it runs no validator (bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746, bump-leg.py: a root on +x is told push -u origin +x (forced), a branch named like a tag passes bump-leg-on-tracking-branch, bump-leg-root-dirty prints C1 controls raw, and committed <sha>: lists paths the commit lacks #672).

     Everything after the list in that section is as it was.

• Counts, in the four places: main 6f64eea; 122 pull requests since bfed143; at 2026-10-17T20:15Z, 132 of the v1.1.0 milestone's 133 issues closed. The one open is Publish openRepoShape to TestPyPI and PyPI as openreposhape, so one pipx install puts the tools and git-filter-repo on any machine (design first) #563 (README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630). family.py add/bump/remove: a members file deleted unstaged, a git bisect, an empty sparse-checkout file, an empty rebase-merge/ and a merge conflicted in family.yaml meet a refusal or remedy that does not fit (#626's gaps) #684 and README.md, the handbook, AGENTS.md and the holder's AGENTS-shape.md: init's --into sentence omits the landing refusal, make siblings is described as it was before #578, and nothing names the one-spelling rule for members/ #695 (README.md, the handbook, AGENTS.md and the family-root docs: init's --into sentence names the family-landing-in-standard refusal, make siblings is described as it stops before cloning, and members/ has its one-spelling rule (#695) #738) stay on v1.1.1. The counts marker keeps the sentence the release-body tool fills.
• Release paths. git diff --stat 20004887..6f64eea2 -- pyproject.toml src/openreposhape/ install.sh .github/workflows/ is empty.
• Tests, on the tree aef568f0 commits, under a stub gh exiting 97, GH_TOKEN=bogus and an empty GH_CONFIG_DIR: the eight docs files gave 300 passed in 37.45s. render-cli-reference.py --check gave [ok], and git diff --check origin/main was clean. README.md is 1474 lines, its cap: 1422 + 2 (460.txt) + 5 (550.txt) + 12 (563.txt) + 22 (564.txt) + 10 (606.txt) + 1 (683.txt) = 1474. AGENTS.md is 494 lines, its cap.

Tests

At 6fd33a9, with /usr/bin/python3 -m pytest (3.12.3), a stub gh exiting 97 first on PATH, GH_TOKEN=bogus, an empty GH_CONFIG_DIR, -o tmp_path_retention_policy=failed, and no GIT_ALLOW_PROTOCOL. The set is 14 files:

• test_repo_hygiene, test_cli_reference, test_agents_md_codes, test_openreposhape_package, test_release_workflow;
• test_install_sh, test_openreposhape_command, test_install_sh_filter_repo;
• test_setup_by_hand_commands, test_readme_sections, test_readme_windows_quoting_sentence, test_update_add_codes_documented;
• test_refusal_default_remediation, test_adopt_filter_repo_failure.

Result: 544 passed in 357.14s (517 at 5b140bd, before main's new tests).

• render-cli-reference.py --check: [ok].
• actionlint 1.7.12 over tests.yml, release.yml and both template validate.yml: no findings.
• git diff --check: clean.
• Earlier, at ca5c49f: the two Git Bash suites passed 117 plain and 117 with CRLF emulated.
• Round B2 part 1, at d697b366, ran the same eight files: 300 passed in 37.04s; render-cli-reference.py --check gave [ok], and git diff --check origin/main was clean.
• Round A4 + B1, at b2a62ec, ran the same eight files: 300 passed in 42.72s; render-cli-reference.py --check gave [ok], and git diff --check was clean.
• Round A3, at d2d3d76, ran the docs and release-notes set under the same rules: test_repo_hygiene, test_readme_sections, test_cli_reference, test_release_workflow, test_agents_md_codes, test_setup_by_hand_commands, test_readme_windows_quoting_sentence and test_update_add_codes_documented. Result: 293 passed in 45.45s. render-cli-reference.py --check gave [ok], and git diff --check was clean.
• The full suite was not run.

Open placeholders in the release notes

Each is marked **[placeholder: …]** or <release-commit>, or sits in a "remove at tag time" section:

1. The counts at tag time. At main fc86720: 121 pull requests since bfed143; at 2026-10-17T19:08Z, 131 of the milestone's 133 issues were closed. The marker and its sentence keep the form the release-body tool fills.
2. The open milestone list (2 at 19:08Z: Publish openRepoShape to TestPyPI and PyPI as openreposhape, so one pipx install puts the tools and git-filter-repo on any machine (design first) #563 (README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630) and family.py bump: a mount with a narrowed fetch refspec is refused member-to-local-only on a false detail, a commit only a tag holds is told "a pin the rest of the world cannot fetch", a mount with no origin is told to fix the network #686 (family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795)), in "Before tagging", to be re-read at tag time.
3. <release-commit> throughout.

No longer placeholders since B2 part 1: "What changed" runs to main fc86720 and says #795 is added if it lands next (part 2 makes it "The list is complete to the release commit."), and v1.0.0's known limits are re-read, with the marker and "as read so far" cut.

Not done here, for review

• The skip reasons in tests/ still say pip install git-filter-repo. There are 40 test files, and every hit is a skip reason (git grep -l 'pip install git-filter-repo' HEAD -- tests/). Changing them is a follow-up if wanted.
• The release workflow: a dispatch on the annotated tag v<version> builds the openreposhape wheel twice from git archive of a commit whose own tree passed the tests job, uploads it to TestPyPI, installs it with pipx on three systems and Python 3.9, and uploads the same bytes to PyPI on approval (#563) #599's docstring in tests/test_release_workflow.py says "No docs/release-notes/v1.1.0.md exists yet". Its assertions still hold.
• Windows is unproven for the package. No test scaffolds through the package's openRepoShape on Windows. The release's install check runs it there, and the rehearsal passed that check on windows-latest.

🤖 Generated with Claude Code

[brettheap]

(lane openRepoShape-3, 2026-10-17T19:23:12Z) RV630's final delta review of #630, part 1 (head d697b366, on main fc867207): READY.

• Merges: every main merge is exact. Each tree equals git merge-tree of its two parents (441dcb4, 8049456, 42d4df8, 4d14e4a), and the non-merge commits touch only docs/release-notes/v1.1.0.md.
• Entries: all ten inserted entries match their squashes. For validate-family.py and validate-pins.py: a staged gitlink removal says to commit it, member-gitlink-absent puts the gitlink back by hand, the variant mv is guarded, and a pin row with no path: is shape-pin-row (#681) #747, repo_shape.py, shape_materialize.py and adopt-project.py: a value the YAML reader refuses for opening with an indicator is named by its line and says to double-quote it or reword it, plan --org @x names --org again under the reader's own remedy, and check names one bad tracking_branch once (#708) #817, tests/conftest.py: one helper builds the url.<file URI>.insteadOf rewrite that clone_just_pushed, read_through_git and the adopt tests' pasted each built, and the empty-mount submodule update --init sites use it (#662) #748, shape_materialize.py, scaffold-project.py and family.py: the folder refusals and family-operation-in-progress name a path as shown_source names it, so an ESC or a line break in a path no longer reaches the terminal (#692) #731, update-shape.py: apply's rollback line says to make a shut directory writable first, committed on <branch>: names what the commit holds, a branch named like a tag is pushed and read by its full name, and a quoted files: or shape: header is found as the reader finds it (#674) #810, family.py init: the creator is asked whether a file can hold it whichever of --created-by, GIT_AUTHOR_NAME or git config user.name gave it, git is asked about the first commit's author and committer before the remote exists, and a --family that is a Windows device name is refused, --dry-run alike (#688) #809, family.py: add, bump and remove name a bisect, an empty rebase directory, sparse rules that hold no pattern, a merge conflicted in family.yaml and a members file HEAD holds and the work tree does not as what each is, with the way out that works there, and refuse a family.yaml holding a change HEAD does not (#684) #816, family.py and shape_materialize.py: the NEXT push line names the branch by its full refname on both sides, so a branch beginning with + or named like a tag pushes as itself, and a tracking branch whose last part is over 250 bytes, which git cannot lock, is refused by refuse_bad_branch before anything exists, --dry-run too (#689) #813, tests: one skip helper for names and branches that are not UTF-8, Windows guards on make_file and the --branch a\udcffb test by emulation, and the UTF-8 promise and the leg finding's role pinned (#663) #723 and bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746, the squash tree equals merge-tree(squash^, final head). bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746's squash diff was read in full, and its entry is accurate.
• The pre-review: the SHOULD and every nit are taken as written. repo_shape.py, shape_materialize.py and adopt-project.py: a value the YAML reader refuses for opening with an indicator is named by its line and says to double-quote it or reword it, plan --org @x names --org again under the reader's own remedy, and check names one bad tracking_branch once (#708) #817 uses the "still" framing, and tests/conftest.py: one helper builds the url.<file URI>.insteadOf rewrite that clone_just_pushed, read_through_git and the adopt tests' pasted each built, and the empty-mount submodule update --init sites use it (#662) #748 is RV630's text "With tests: every test that clones a just-pushed local remote by its plain path now clones it with clone_just_pushed, so the macOS failed to copy file race of #440 cannot reach it; load_without_bytecode keeps the doctor's and the Windows-path tests' loads of the family root's scripts from writing .pyc into the templates; the not-a-list doctor cases hold INVALID, exit 1; and GOOD_KEYS holds th #554 and siblings.py and family.py: make siblings asks about the family folder before it clones, clones each member on its tracking branch, and family.py siblings runs the holder's own copy (#543) #578", which is right for v1.0.0 :530.
• v1.0.0 limits: :433, :439, :463, :466, :467 and :530 close. :492 and :440 stay open.
• Counts: they match gh at 19:17Z: 121 PRs; 2 open, 131 closed. The counts marker's wording is unchanged, which the release-body tool needs.
• Public wording: no lane name is left in Known limits. All 36 L-ids name their issue, checked in full against lane 4's comment 6217406337 and lane 1's 6214768828.
• The rest: README, the handbook and AGENTS.md are unchanged since 6a6817c. README is 1474 of its 1474 cap and AGENTS.md 494, and all 23 handbook markers hold.
• Runs: the eight docs files give 300 passed, render-cli-reference.py --check gives [ok], and git diff --check is clean.
• fc867207 (bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746) touches none of pyproject.toml, src/openreposhape/, install.sh or the workflows. Lane 3 re-runs the release rehearsal on d697b366 anyway (tests run 39061666486), as the only Windows, macOS and Python 3.9 run since b2a62ec7.

Part 2, after #795 lands:

• family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795's entry;
• line 120 made final;
• the Known limits lead reworded, so it claims no complete list, with v1.0.0 :440 listed beside :492;
• the recount.

RV630 then re-checks that delta, and #630 lands under Brett Heap's ruling of 2026-10-17T18:57:19Z ("go with option 1 for #630").

Lane: openreposhape-3 (openRepoShape-3)

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[brettheap]

(lane openRepoShape-3, 2026-10-17T20:25:55Z) RV630b's final delta review of #630, part 2 (d697b366..aef568f0, on main 6f64eea2): READY. There are no MUST or SHOULD items.

• The three NITs are left for after the tag. NIT-1 and NIT-2 are wording in the notes and neither is false. NIT-3 is this body, which does not feed the squash message.
• README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630 lands now, LAST, under the standing word of 2026-10-16T12:58:08Z and Brett Heap's ruling of 2026-10-17T18:57:19Z, "go with option 1 for README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630". It lands on:
  • this review and RV630's part 1;
  • SonarCloud green on aef568f0;
  • lane 3's gate on the predicted squash (tree dd0f94e8, 414 passed, 1 skipped).
• Main's push run on the squash is the release commit's test run.

---

RV630b: final delta review of #630, d697b366..aef568f0 on l3-563-docs

Verdict: READY. There are no MUST or SHOULD items. There are three optional NITs, and none blocks the landing. The head was still aef568f02a9c626c9ae93f67e1a508c14684312d and main 6f64eea2 at 20:23:40Z (ls-remote). I made no GitHub write. The predicted squash commit 21fdc03a exists only in my scratch clone and was never pushed.

Checklist

1. The merge is exact.

   • 016599b0 has parents d697b366 and 6f64eea2. Its tree 679d7341 equals git merge-tree --write-tree d697b366 6f64eea2 (exit 0).
   • git diff --stat d697b366 aef568f0 touches 4 files:
     • docs/release-notes/v1.1.0.md (32 lines);
     • scripts/family.py;
     • tests/test_family_bump_reachability_words.py;
     • tests/test_family_bump_to_local_only.py.
   • The three code files' blobs at the head equal main's (a7660e9c, 6dfdb9ba, 7867db6a). aef568f0 alone changes only the notes (+22 −10).
   • The head's diff against origin/main is still the PR's twelve files. README, AGENTS.md, the handbook, docs/cli.md and readme-cap.d/ are unchanged since d697b366.

2. family.py: bump fetches every branch of the member's origin whatever the mount's refspec and drops the origin/ refs no ref of origin backs, says a tag of the remote reaches a commit no branch does instead of "a pin the rest of the world cannot fetch", and tells a mount with no origin it has none instead of to fix the network (#686) #795's entry is at v1.1.0.md:449-456, directly before "family.py refuses before it writes.". Each claim was checked against git show 6f64eea2 and scripts/family.py on main:

   Claim	Measured on main
   "fetches every branch … whatever fetch refspec the mount has"	:6568 is _ORIGIN_BRANCHES = "+refs/heads/*:refs/remotes/origin/*", used in the fetch at :7186. The old fetch at fc867207:6995 was a plain git fetch -q origin.
   "prunes the remote-tracking branches origin has deleted"	:7186 runs fetch -q --prune --no-prune-tags origin …. This is Brett's ruling relayed in the squash message ("prune the fetch the tool already makes").
   "a commit a branch holds is no longer refused for a narrowed refspec" / "one only a deleted branch held is no longer pinned"	The squash message and the code comment at :7165-7184 agree.
   "only a tag of the remote reaches is told that a tag is not a branch"	:6735-6743; the refusal still applies, only the words change.
   "no remote named origin is told so"	:6727 (member-to-local-only), :6858 (member-to-unknown) and the note at :7191.
   "where each was told the pin cannot be fetched, or to fix the network"	The old words are at fc867207:6589 ("fix the network or the credential") and :6597 ("A pin the rest of the world cannot fetch").

   • Style: it uses the same bold lead and (#PR, #issue) order as the neighbouring entries ((#625, #613), (#578, #543), (#626, #612)). It sits right after the family.py: bump --to refuses a commit only the member's mount holds, reachable from no branch of the member's remote, as bump-leg.py refuses its twin, so a holder is never pinned to a commit nobody else can fetch (#613) #625/family.py bump --to: pins a commit only the member's mount holds, which bump-leg.py refuses as local-only #613 reachability sentence, which is the logical place.
   • family.py bump: a mount with a narrowed fetch refspec is refused member-to-local-only on a false detail, a commit only a tag holds is told "a pin the rest of the world cannot fetch", a mount with no origin is told to fix the network #686 is right: the issue is CLOSED as COMPLETED, on milestone v1.1.0, closed by [795]. The squash says "Closes family.py bump: a mount with a narrowed fetch refspec is refused member-to-local-only on a false detail, a commit only a tag holds is told "a pin the rest of the world cannot fetch", a mount with no origin is told to fix the network #686", and the issue's title matches the entry's three states.

3. Line 120 reads runs to main \6f64eea`., ending with a plain period. The sentence spans :119-120: "The list" / "runs to main 6f64eea`.".

4. Known limits.

   • cmp of the brief's lines 24-35 against the head's :681-692 exits 0, so the text is byte for byte the brief's.
   • The 134 lines from "New in v1.1.0:" to the next ## match d697b366's :681 onward (cmp exits 0).
   • The README.md, the handbook, AGENTS.md and the holder's AGENTS-shape.md: init's --into sentence omits the landing refusal, make siblings is described as it was before #578, and nothing names the one-spelling rule for members/ #695 limit is true on main and on the head. README :906-907 (head) and handbook :1365 give only "(the directory you are standing in, by default)". family.py:1804-1840 on main refuses family-landing-in-standard. README.md, the handbook, AGENTS.md and the holder's AGENTS-shape.md: init's --into sentence omits the landing refusal, make siblings is described as it was before #578, and nothing names the one-spelling rule for members/ #695 is OPEN on v1.1.1, and its first line is exactly this. It matches v1.0.0 :492.
   • The bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746/bump-leg.py: a root on +x is told push -u origin +x (forced), a branch named like a tag passes bump-leg-on-tracking-branch, bump-leg-root-dirty prints C1 controls raw, and committed <sha>: lists paths the commit lacks #672 limit is true on main.
     • bump-leg.py:1700-1712: the dry run returns before the validators.
     • :1786-1794: the dry run now says "A dry run does not run the project's own validators … can still end bump-leg-validators-red, with every byte rolled back."
     • :1606-1609: the real run raises bump-leg-validators-red with ROLLED_BACK, and a refusal returns 2 (:1901).
     • bump-leg.py: a root on +x is told push -u origin +x (forced), a branch named like a tag passes bump-leg-on-tracking-branch, bump-leg-root-dirty prints C1 controls raw, and committed <sha>: lists paths the commit lacks #672 is CLOSED by bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746. Its dry-run line's chosen direction was "say so; no behaviour changes", so "left open" is right.
     • It matches v1.0.0 :440 and agrees with bump-leg.py: the branch is read in full and pushed as refs/heads/<b>:refs/heads/<b>, dirty names are spelled, committed lists what git holds, and the dry run says it runs no validator (#672) #746's own entry at :280 ("where its behaviour is unchanged").

5. Counts.

   • Measured: gh api …/milestones/2 gave open=1 closed=132 at 20:21:19Z; git rev-list --first-parent --count bfed143..origin/main gave 122; rev-parse --short=7 gave 6f64eea.
   • Line 4 reads 6f64eea. Lines 14-16 read 6f64eea, 122, 20:15Z, and 132 of 133.
   • With digits and shas masked, the counts marker and the header comment are the same as at d697b366 (diff empty).
   • The open-list paragraph is at :943-946 (the brief said 930-934): "(1 open, 132 closed)", the open list holds only "Publish openRepoShape to TestPyPI and PyPI as openreposhape, so one pipx install puts the tools and git-filter-repo on any machine (design first) #563 (README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630)", and the family.py add/bump/remove: a members file deleted unstaged, a git bisect, an empty sparse-checkout file, an empty rebase-merge/ and a merge conflicted in family.yaml meet a refusal or remedy that does not fit (#626's gaps) #684/README.md, the handbook, AGENTS.md and the holder's AGENTS-shape.md: init's --into sentence omits the landing refusal, make siblings is described as it was before #578, and nothing names the one-spelling rule for members/ #695 sentence is present. family.py add/bump/remove: a members file deleted unstaged, a git bisect, an empty sparse-checkout file, an empty rebase-merge/ and a merge conflicted in family.yaml meet a refusal or remedy that does not fit (#626's gaps) #684 is OPEN on v1.1.1 (gh issue view 684), and so is README.md, the handbook, AGENTS.md and the holder's AGENTS-shape.md: init's --into sentence omits the landing refusal, make siblings is described as it was before #578, and nothing names the one-spelling rule for members/ #695.

6. Release-body dry run.

   • Facts: main_sha 21fdc03 (the predicted squash), 2026-10-17 20:22, pr_count 123 (bfed143..21fdc03a), closed 132, open 1, with no fills (no fill marker is left).
   • Run on the notes as they stand in the predicted squash (byte-identical to the head's), with --release-commit 21fdc03a5f6c7110e054816ec6f850f871782773 --repo <clone> --dry-run, under the stub gh.
   • Result: tool-exit=0. It dropped the leading comment (:3-6), "How this release is cut (remove at tag time)" (:828-932) and "Before tagging (remove at tag time)" (:933-950).
   • It filled the counts; replaced <release-commit> ×1; rewrote 3 links and checked all 3 at 21fdc03a; and left no WARNING.
   • Grepping the body for placeholder / remove at tag time / <release-commit> / FIRST DRAFT / <!-- / illustrative finds nothing (exit 1). There are 0 CRs.
   • Body: 826 lines, 52202 bytes, sha256 3672feb3dd2ce64caec36fecbf8f4e2bbe7ccce97a20ab4f989790dbb45ef9d7.
   • Commit ids it lists: 21fdc03 (line 9), bfed143 (line 10), and 6f64eea (line 115, line 120's text, which is deliberate and not stale).

7. Tests, run in a detached worktree at aef568f0 with /usr/bin/python3, a stub gh first on PATH exiting 97, GH_TOKEN=bogus, an empty GH_CONFIG_DIR and no GIT_ALLOW_PROTOCOL:

   Command	Result
   pytest -q -p no:cacheprovider --basetemp=…/bt-RV630b-docs -o tmp_path_retention_policy=failed on the 8 docs files (test_repo_hygiene, test_readme_sections, test_cli_reference, test_release_workflow, test_agents_md_codes, test_setup_by_hand_commands, test_readme_windows_quoting_sentence, test_update_add_codes_documented)	300 passed in 39.63s
   scripts/render-cli-reference.py --check	[ok] docs/cli.md is what the generator produces (exit 0)
   git diff --check origin/main...aef568f0	exit 0, clean
   README cap	1474 lines = 1422 + 2 + 5 + 12 + 22 + 10 + 1 = 1474, at the cap
   AGENTS.md cap	494 lines; the cap is <= 494 (test_repo_hygiene.py:667)

   The worktree was clean after the runs.

8. Predicted squash. git merge-tree --write-tree origin/main aef568f0 exits 0, giving tree dd0f94e86d8c97bc756335ffb68d3a5fd38c3919, which is the same as the head's own tree.

   • My local predicted commit on origin/main is 21fdc03a5f6c7110e054816ec6f850f871782773, and it changes the same 12 files (+1232 −107).
   • The brief named no other lane-3 v1.1.1 PRs, so I merged against none.

SonarCloud and CI on the head

From gh pr checks 630 and the check runs for aef568f0:

• SonarCloud Code Analysis: success, "Quality Gate passed" (20:17:36Z, head_sha aef568f0).
• package and tests-git-bash passed.
• tests was still in progress at 20:23Z. Under Brett's 18:57:19Z ruling it is not awaited; main's push run on the release commit is still required before the tag.

MUST / SHOULD

None.

NITs (optional; READY either way)

• NIT-1, docs/release-notes/v1.1.0.md:119-120.
  • Scenario: the list cites README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630 itself, at :668 "(README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630, shape-doctor.py: run from a partial clone of the standard, the shape currency row fetches what the clone lacks, and with its remote out of reach calls each copy the standard changed since the pin both, so a doctor that fetches nothing fetches, and offline says DRIFTED of copies nobody edited #683)" and :677 "(README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630, Tracking: shape-doctor.py, reporting a repository's compliance #460)". README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630 lands after 6f64eea, as the release commit.
  • Measurement: at tag time the published body will say "At main <RC> … 123 pull requests" and also "The list runs to main 6f64eea".
  • This is not false. If you want it exact, the fix is: The list runs to main \6f64eea`, and to README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630, this release's docs, which lands last.`
• NIT-2, docs/release-notes/v1.1.0.md:450-453.
  • Scenario: per the squash message and :7175-7184, --prune also drops a ref put under refs/remotes/origin/ by hand, and does so in --dry-run too. Offline, nothing is pruned, so the deleted-branch case is refused only once the fetch works.
  • Fix: replace prunes the remote-tracking branches \origin` has deleted, so a commit a / branch holds is no longer refused for a refspec narrowed by hand, and one / only a deleted branch held is no longer pinned.withprunes, in `--dry-run` too, every `origin/` ref that `origin` no longer backs, so a commit a branch holds is no longer refused for a refspec narrowed by hand, and one only a deleted branch held is no longer pinned once the fetch works.` (rewrap it to the file's width).
• NIT-3: the PR body, not the file. Some of its text is stale against part 2:
  • "## What" still says "to main fc86720 (121 pull requests)" and "lane openRepoShape-4's ten lines and lane openRepoShape-1's twenty-six D9 lines".
  • "## Open placeholders" items 1-2 still say fc86720, 121, 131 of 133 at 19:08Z, and 2 open.
  • Part 1's plan line "'The list is complete to the release commit.'" is not what part 2 did.
  • Fix: 6f64eea, 122, 132 of 133 at 20:15Z, 1 open (Publish openRepoShape to TestPyPI and PyPI as openreposhape, so one pipx install puts the tools and git-filter-repo on any machine (design first) #563 (README, AGENTS.md, docs/cli.md and the handbook: pipx install openreposhape is the second way in and the first on native Windows, git-filter-repo is installed in install.sh's order and never with a bare pip install, and a first draft of the v1.1.0 release notes (#563) #630)); "the checklist triage of the L4 lines on Tracking: adopt-project.py, converting an existing repository in place #457 and Tracking: setup, scaffold, family and the shared helpers, creating projects and families #459 / the L1 lines on Tracking: adopt-project.py, converting an existing repository in place #457"; and "The list runs to main 6f64eea." This matters only if the body feeds the squash message.

Mutants

No pytest reads v1.1.0.md's content: tests/ cites it only in a docstring at test_release_workflow.py:580, and test_repo_hygiene.py:1567 names only v1.0.1.md. Text drift in the entry, line 120 or Known limits is therefore killed by no test, only by review.

The counts marker is guarded by the release-body tool. In a scratch copy I changed "had landed" to "landed", and the tool returned exit 1, an empty body, and REFUSED: line 14: the paragraph under **[placeholder: the counts at tag time.]** is not the sentence the tool fills.

Lane: openreposhape-3 (openRepoShape-3)

[sourcery-ai]

Sorry @brettheap, you've used your own review budget of 250,000 diff characters for the last 7 days.

You can request another review in 2 days and 4 hours by commenting @sourcery-ai review. Upgrade to get a review now.