Tracking: setup, scaffold, family and the shared helpers, creating projects and families

Lane: openreposhape-3 (openRepoShape-3)

The parent issue for one area of openRepoShape: creating a project or a family (setup.sh, scaffold-project.py, family.py) and the helpers and templates every tool shares (shape_materialize.py, repo_shape.py, the shipped validators). Brett Heap ruled on 2026-10-12 that the open issues are grouped under one parent per area, with a release milestone. Every open issue in this area is a sub-issue here.

v1.0.0. The first release is "core flows, all OSes": every procedure in AGENTS.md works end to end for ordinary repositories on Linux, macOS and Windows. The tag waits for every in-scope issue to close and for a full test run to pass on all three OSes on the release commit. In this area, the sub-issues in the v1.0.0 milestone are:

family.py init exits 0 and pushes a holder its own validate-family.py refuses for --id null, --created-by ' ', --created-on 20261011 or --name 'Contoso Group', and puts --tracking-branch true on a branch its manifest reads back as True #307 (P2): family.py init --name 'Contoso Group' or --created-on 20261011 exits 0 and pushes a holder its own validator refuses; init cannot be re-run.
family.py bump (and add, remove) take no --dry-run, though the file's exit-code paragraph says a dry run prints and exits 0 and bump-leg.py has one, so the only way to see what a bump would pin is to run it: fetch, check the member out at --to and rewrite family.yaml #326 (P2): bump and remove got --dry-run (family.py: bump --dry-run fetches the member, resolves --to and prints the old -> new row and its trailers without checking the member out or rewriting family.yaml, and remove --dry-run prints the row, gitlink, .gitmodules section and directory it would remove and changes nothing, so the module's EXIT CODES line is true of both (#326) #345, landed); only add has none, so its mount cannot be previewed. family.py: add --dry-run makes every refusal add makes before it mounts, reads the member from throwaway clones outside the holder for the row's id and digest, and prints the row, gitlink, .gitmodules section and trailers without mounting, writing or committing anything, and add takes --trailer as bump does, so every subcommand has the dry run the module's EXIT CODES line promises (#326) #400 adds it.
setup.sh: the shim self-bootstraps before it hands over and its pre-scan never looks for --help, so setup.sh --help run from anything but a checkout of the standard (a downloaded copy, curl … | bash -s -- --help, a git archive extraction) clones github.com/opensoft/openRepoShape before printing a word, prints the clone's usage rather than its own, dies with git's exit 128 offline, and makes scripts/render-cli-reference.py in such a tree reach GitHub, though setup-project.py --help answers offline from the same place #341 (P2): setup.sh --help from a download or `curl .
family.py bump and remove have no rollback: when a git step after write_members fails (a holder pre-commit hook that exits 1, a held index.lock), they exit 2 with REFUSED a git or gh command failed and leave family.yaml rewritten, the member checked out at --to (or deinitialised and removed), and the change staged or half-staged, without a word about any of it, where add puts the holder back and names what it put back #365 (P2): bump/remove refused by a pre-commit hook or held index.lock exit 2 with family.yaml rewritten and staged; nothing says so. Reproduced on main.
shape_materialize.py force_stage's docstring says it has TWO CALLERS and that only update-shape.py apply --add calls it alone, and stage_written's says the same, though since #242 family.py's _commit_one (every add, bump and remove commit) and cmd_bump call it alone too, so the next change to its contract is reasoned about with half of its alone-callers missing #369 (P2): Maintainer-only docstring miscounts force_stage callers; no behaviour wrong, no user hit. PR already open and docstring-only, so merging costs nothing.
repo_shape.py's Refusal gives every refusal built without a remediation the assembly root's own (run git submodule update --init --recursive, then python3 scripts/validate-pins.py ... contracts/<leg>-pin.yaml), so in a family holder family.py add over an unparsable .gitmodules (git-failed), bump on a family.yaml of the wrong kind, and the holder's own validate-family.py on a short pin.commit each tell the operator to run a file no holder carries and to move a leg pin a holder does not have #370 (P2): A family user with a conflict-marked .gitmodules or family.yaml is told to run validate-pins.py and move a leg pin the holder lacks.
family.py bump --to an annotated tag's object id, which is 40 hex and passes COMMIT_RE, writes the TAG's id into the member's pin.commit while git checkout puts the commit the tag points at in the gitlink, so bump exits 0 and commits a holder whose own validate-family.py reports member-gitlink-mismatch, exit 1, where add --at pins the commit the same tag names #405 (P1): family.py bump --to <annotated tag's id> (what git rev-parse v2 prints) exits 0 and commits a holder whose own validate-family.py is red.

Known limits. Every other sub-issue is a known limit: only exotic input triggers it (bytes that are not UTF-8, the C locale, control characters, hand-edited YAML, symlink tricks, names that differ only by case). The release notes list them, and they are worked when a release takes them in. A new edge case that review finds goes on this checklist, not into a new issue, unless it is in a release's scope or would hit a real user (Brett Heap, 2026-10-12).

Known limits measured by lane openRepoShape-3 on main 45e01f4 and kept here rather than filed:

→ family.py bump --to: make the checks add makes (a duplicate id, an annotated tag's own id), and say to push first for an unpushed commit #541 F1: family.py bump never asks whether the id the member's project.yaml records at --to is one another row already records, so it commits a holder its own validate-family.py refuses as member-duplicate-id. Direction: ask bump (and --dry-run) the question add asks since family.py: add refuses member-duplicate-id before it mounts anything when the member's project.yaml records an id a row already pins, so adding a repository renamed at the forge no longer commits a second row the holder's own validate-family.py refuses (#330) #381.

Edge cases found in review of lane openRepoShape-3's pull requests, kept here rather than filed:

→ rides with family.py bump --to: make the checks add makes (a duplicate id, an annotated tag's own id), and say to push first for an unpushed commit #541. Ruled 2026-10-14: bump --to peels an annotated tag's own id, as add --at does. family.py add --at <an annotated tag's own id> peels it to the commit it tags (family.py: add asks every refusal of the holder and of a throwaway clone of the member before it mounts anything, and unwinds and names whatever fails after the mount, so a refused add (a leg, a short or unknown --at, an empty member) no longer leaves a staged gitlink and .gitmodules section that the next add commits under its own title (#203) #236), while bump --to refuses the same id once family.py: bump refuses a --to that names an annotated tag, a tree or a blob as member-to-not-a-commit before the checkout, naming the commit a tag names, and its dry run refuses alike, so a release tag's own id no longer exits 0 with a commit pinning the tag over a gitlink at its commit, which the holder's validate-family.py refused as member-gitlink-mismatch (#405) #466 lands, and the two word a tag of a tree differently ("names a tree", "is an annotated tag of a tree"). Decide one rule: add refuses too, or the peel is documented as deliberate. From the review of family.py: bump refuses a --to that names an annotated tag, a tree or a blob as member-to-not-a-commit before the checkout, naming the commit a tag names, and its dry run refuses alike, so a release tag's own id no longer exits 0 with a commit pinning the tag over a gitlink at its commit, which the holder's validate-family.py refused as member-gitlink-mismatch (#405) #466.
→ scaffold-project.py re-run: a leg that gh sees and git cannot read gets a generic failure, and the reuse run's "two new legs" and private-leg text are wrong #542 scaffold-project.py: when git cannot read a leg that gh can see, the refusal is the generic a git or gh command failed instead of scaffold-remote-exists (scaffold-project.py, family.py init: after a ruleset refuses the first push, neither exit RULESET_HINT offers can finish: the pull request has no base in the repository gh repo create left empty, and the scaffold's re-run is refused scaffold-remote-exists for its own two empty legs, so a ruleset-protected organisation cannot seed a new project; --reuse-empty-repo is to reuse empty legs, and the hint is to offer only that re-run #462's must-keep), because _held_seed runs for every non-empty leg (scaffold-project.py:431 at scaffold-project.py and family.py init: --reuse-empty-repo reuses a leg that is empty or whose one commit is the scaffold's own seed, and the ruleset hint offers that re-run, by an operator holding the bypass right, as its one exit, so a project a ruleset refused part-way is finished by one re-run, not refused scaffold-remote-exists for its own legs or sent to a pull request with no base (#462) #506's head c6761db) and only_commit (:357-365) runs git ls-remote and git fetch --depth=1; run() does not set GIT_TERMINAL_PROMPT=0, so on a terminal git can stop for a username; and the depth-1 fetch downloads the leg's whole tip tree to read one commit. Direction: catch CommandFailed in _held_seed and return None; optionally --filter=tree:0. From the review of scaffold-project.py and family.py init: --reuse-empty-repo reuses a leg that is empty or whose one commit is the scaffold's own seed, and the ruleset hint offers that re-run, by an operator holding the bypass right, as its one exit, so a project a ruleset refused part-way is finished by one re-run, not refused scaffold-remote-exists for its own legs or sent to a pull request with no base (#462) #506 (N1).
→ scaffold-project.py re-run: a leg that gh sees and git cannot read gets a generic failure, and the reuse run's "two new legs" and private-leg text are wrong #542 scaffold-project.py's visibility messages assume both legs were just created: "the two new legs inherit…" and "The two new legs are created…" (:918, :924 at c6761db), and _private_leg_note (:1078-1085) goes by the run's --visibility, so a re-run that reuses legs can print a NOTE that contradicts its own per-leg is reused and is public WARNING, and private reused legs under a public run get no private-leg note while project.yaml records the run's value. From the review of scaffold-project.py and family.py init: --reuse-empty-repo reuses a leg that is empty or whose one commit is the scaffold's own seed, and the ruleset hint offers that re-run, by an operator holding the bypass right, as its one exit, so a project a ruleset refused part-way is finished by one re-run, not refused scaffold-remote-exists for its own legs or sent to a pull request with no base (#462) #506 (N2).
→ adopt-project.py: what execute and AGENTS.md say around a run (the plan's follow-up, a failed verification's open PR, an answer with no resolution:, the ruleset hint) #523 adopt-project.py's ruleset hint still offers "The scaffold's other exit, a seed branch and a pull request…" (adopt-project.py:5248, comment at :5230, on main at f391e58), an exit scaffold-project.py and family.py init: --reuse-empty-repo reuses a leg that is empty or whose one commit is the scaffold's own seed, and the ruleset hint offers that re-run, by an operator holding the bypass right, as its one exit, so a project a ruleset refused part-way is finished by one re-run, not refused scaffold-remote-exists for its own legs or sent to a pull request with no base (#462) #506 removed from the scaffold; tests/test_ruleset_hint_tracking_branch.py:115 pins that sentence word for word, so the fix changes both. From the review of scaffold-project.py and family.py init: --reuse-empty-repo reuses a leg that is empty or whose one commit is the scaffold's own seed, and the ruleset hint offers that re-run, by an operator holding the bypass right, as its one exit, so a project a ruleset refused part-way is finished by one re-run, not refused scaffold-remote-exists for its own legs or sent to a pull request with no base (#462) #506 (N3).
-> AGENTS.md, README.md, the handbook, the release notes and scaffold-project.py: the --family refusal path, "nothing is created", three ruled bump limits and a seed/scaffold re-run are unsaid or slightly wrong #921 A seed found only on seed/scaffold (the branch the old hint made) is accepted, and the re-run creates main at it without touching seed/scaffold; on GitHub the leg's default branch likely stays seed/scaffold, and the run says nothing about it (tests/test_scaffold_reuse_seeded_legs.py:166; not verified on github.com). From the review of scaffold-project.py and family.py init: --reuse-empty-repo reuses a leg that is empty or whose one commit is the scaffold's own seed, and the ruleset hint offers that re-run, by an operator holding the bypass right, as its one exit, so a project a ruleset refused part-way is finished by one re-run, not refused scaffold-remote-exists for its own legs or sent to a pull request with no base (#462) #506 (N4).
-> tests/test_ruleset_hint_tracking_branch.py: the "kept at its commit" check cannot catch a new seed commit made in the same second #922 tests/test_ruleset_hint_tracking_branch.py:429-430's "kept at its commit" assertion cannot catch a new seed commit made in the same second (an identical commit): with _take_held_seed disabled, [scaffold-code-main] still passes while five other tests fail. The dedicated reuse test pins a later commit date, so the guard as a whole holds. From the review of scaffold-project.py and family.py init: --reuse-empty-repo reuses a leg that is empty or whose one commit is the scaffold's own seed, and the ruleset hint offers that re-run, by an operator holding the bypass right, as its one exit, so a project a ruleset refused part-way is finished by one re-run, not refused scaffold-remote-exists for its own legs or sent to a pull request with no base (#462) #506 (N5).

Known limits found by lane openRepoShape-3's acceptance passes (standard 36538fd, each re-checked on main 7a8b981) and kept here rather than filed:

README.md, the handbook, AGENTS.md and the family-root docs: `init`'s…

family.py, scaffold-project.py and bump-leg.py: `init`'s plan rows, s…

setup-project.py: a project name typed at the prompt is checked as `-…

(lane openRepoShape-4, 2026-10-18T04:05:33Z) One line for this checklist, from #868 (#855). It is RV868's SHOULD-2 (#868 issuecomment-6245096782). It was ruled disclosure only, and #868's body names it under "What it cannot tell". It is not a regression: on main the refusal agreed with every reader (validate-manifest.py said shape is missing); since #868 the readers read the block, and only the rewriters do not. (Corrected on RV868b's NIT-2, #868 issuecomment-6245862602: this line first said that nothing reached these rewriters on main, which was false, because update-shape.py apply refused there too, with the same update-manifest-no-shape line.) No tool of this standard writes such a file: every template's first line is schema_version: 1.

The three line rewriters refuse a file whose FIRST line is the key they look for, saved with a byte-order mark, in words that say the key is not there, while every reader reads it.
- (1) What happens. Since bootstrap.py and repo_shape.py: a .shape-upstream-tip or a YAML file a Windows editor saved with a byte-order mark reads as the text it holds, so the upstream notice names the tip, a marked project.yaml is no longer schema_version is None, and every reader of a marked pin, the doctor's included, reads it alike (#855) #868, parse_yaml takes off the leading mark, so the validators, load_yaml and the readers of a pin at a commit read the file. The three finders match their key at column 0 of the raw text, where line 1 starts with U+FEFF, so they do not find it:
  - update-shape.py:1532 (rewrite_shape_pin, commit:);
  - update-shape.py:1683 (_shape_block_bounds, the shape: header, refused at :1698);
  - scripts/bump-leg.py:292 (rewrite_pin, commit:).
- (2) Measured on bootstrap.py and repo_shape.py: a .shape-upstream-tip or a YAML file a Windows editor saved with a byte-order mark reads as the text it holds, so the upstream notice names the tip, a marked project.yaml is no longer schema_version is None, and every reader of a marked pin, the doctor's included, reads it alike (#855) #868's head 62074123, in process, with the marks written as bytes. The same files without the mark, or with schema_version: first, are written.
  - update-shape.py: REFUSED update-pin-unrecognised: contracts/shape-pin.yaml has 0 commit line(s) where exactly one was expected, so this tool will not rewrite it, for a shape pin whose first line is commit:.
  - update-shape.py: REFUSED update-manifest-no-shape: project.yaml has no shape: block, so there is nothing to mirror the pin into, for a project.yaml whose first line is shape:. On that file the project's own validate-manifest.py says manifest ok and load_yaml reads shape.commit.
  - scripts/bump-leg.py: REFUSED bump-leg-pin-unrecognised: the pin file has 0 commit line(s) where exactly one was expected, so this tool will not rewrite it, for a leg pin whose first line is commit:, with one mark or two (RV868 measured both).
- (3) Direction (RV868's).
  - In all three finders, match the key or the header after one leading mark, on line 1 only.
  - The parser takes off every leading mark, so taking off all of them on line 1 would follow it to two marks as well.
  - Write the mark back as it was, as each rewriter already keeps every other line's bytes.
  - A test per finder: the marked first-key file is written, its first bytes are the mark, and its value is the target.

tests: four tests, `directory_named` and the file twin ask the one he…

repo_shape.py strips only YAML's space and tab, so a value a Unicode …

(lane openRepoShape-4, 2026-10-18T04:47:49Z) One line for this checklist, from #870 (#864). It is RV870's NIT-3 (#870 issuecomment-6245132847), left out of #870 as outside #864. It is not a regression: main says the same.

family.py add refuses a holder whose family.yaml says members: null, with the null written as a word, in words that say the file has no members: line.
- (1) What happens.
  - add refuses with family-members-unrecognised. Its text says the file has 0 members: line(s), which is untrue: the line is there, and the readers read it as null.
  - add's line finder (rewritten() in scripts/family.py) takes members: and members: [] and no other spelling.
- (2) Measured by RV870 on shape-doctor.py, family.py and the family root's bootstrap.py and validate-family.py: a holder's members: that is not a list, or holds an entry that is not a mapping, is said in the gate's words, so the doctor's members row waits on the family row instead of calling the family empty, add, bump, remove, make bootstrap and make siblings refuse it by name where each ended in a TypeError, read no members or committed the entry away, and --no-members finds the entry (#864) #870's head 03fe51e6. It is the same on main.
- (3) Direction.
  - Have the finder take each spelling of a YAML null that the parser reads (null, Null, NULL and ~) as it takes members:, so that add writes the first member there too.
  - Otherwise, have the refusal name the null.
  - Add a test per spelling.

(lane openRepoShape-3, 2026-10-18T14:55:27Z) Triage of #459's unchecked items at the v1.1.1 release commit 5017830. This is lane 3's TRIAGE-459, read only, before this filing.

How to read it. U001 to U188 number #459's 188 unchecked lines in posting order: U001 to U015 are the body's lines, and U016 to U188 are the comments' lines. A line posted twice is ruled once, as "A = B" (lane 4's copy of 2026-10-16T14:10:49Z repeats 26 earlier lines). V-rows are the v1.0.0 Known limits under #459 that have no checklist line. Every fix commit named below is on the first-parent history of 5017830, and every code line is at 5017830. Probes ran with a stub gh that exits 97, GH_TOKEN=bogus, an empty GH_CONFIG_DIR, GIT_CONFIG_GLOBAL=/dev/null and local remotes.

Counts. Of the 188 lines: 115 FIXED; 3 partly fixed (U041, and U119 = U151), whose open half is filed; 1 COVERED (U180, by #430); 1 STALE (U045); 68 OPEN, which is 64 items once the duplicates are counted once. Nothing is UNSETTLED. The 17 V-rows are all FIXED.

Filed today, milestone v1.2.0 (21 issues):

Issue	Group	#459 lines
#905	`family.py bump`'s fetch	U182, U183, U184, U185
#907	setup step (4) and the naming gate's link remedies	U061 = U169, U170, U171, U173, U174
#909	chain pin sources	U172, U113 = U145, U114 = U146
#911	empty flags and id remedies in the scaffold and `family.py init`	U156, U157, U158, U159, U160, U162
#913	plan and NOTEs about visibility	U163, U164, U166, U167, U168, U010, U060; U039 (needs a ruling)
#915	the scaffold's git prompts on reuse	U036, U037, U038
#917	the YAML reader in `repo_shape.py`	U021, U181; U123 = U155 (documented limit, needs a ruling)
#919	`validate-pins.py` and `repo_shape.py` over odd pins and raw text	U176, U177, U178, U179
#920	`family.py add`	U049, U188
#921	docs	U033, U088, U066, U068, U071, U006; U097 (needs a ruling)
#922	test gap	U007
#923	small text hygiene	U062, U063, U064
#924	alone	U034
#925	alone	U022
#926	alone	U041 (the open autocrlf half)
#927	alone	U074
#928	alone	U018
#929	alone	U119 = U151 (the open family half)
#930	alone	U129
#932	alone	U186
#933	alone	U187

The body. Lane 3 ticked its own body lines that are FIXED (U001 to U005, U008, U009, U011 to U015), and marked the three OPEN ones with their new issue: U006 -> #921, U007 -> #922, U010 -> #913. Comment lines left unticked stay as their authors wrote them: this comment is their ruling, and each author may tick or annotate their own lines from the table below.

The table

fam is scripts/family.py, vf is templates/family-root/scripts/validate-family.py, vp is templates/assembly-root/scripts/validate-pins.py, sm is scripts/shape_materialize.py and rs is scripts/repo_shape.py.

Line	Ruling	Evidence at `5017830`
U001 bump: duplicate id	FIXED	`8890f57` (#568, #541); fam:7992 `_checked_bumped_id`
U002 bump --to peels an annotated tag	FIXED	`8890f57` (#568, #541); fam:8322
U003 leg gh sees and git cannot read	FIXED	`26c1600` (#565, #542); scaffold-project.py:533-535, `PROBE_ENV` :469
U004 "two new legs" text	FIXED	`26c1600` (#565, #542); scaffold:1525-1540. Its last clause is U039
U005 adopt ruleset hint "other exit"	FIXED	`526cd2c` (#582, #523); adopt-project.py:10773; test_ruleset_hint_tracking_branch.py:166
U006 seed only on `seed/scaffold`	OPEN -> #921	scaffold:1583-1599; documented at v1.0.0.md:482
U007 "kept at its commit" test gap	OPEN -> #922	test_ruleset_hint_tracking_branch.py:429-432
U008 scaffold NEXT clone lacks protocol.file	FIXED	`bcfa364` (#569, #538); scaffold:2078
U009 step (4) ignores `-- --pin`	FIXED	`bcfa364` (#569, #538); setup-project.py:2417
U010 plan names no branch, no local visibility	OPEN -> #913	scaffold:1321-1361, :1387-1399; documented at v1.0.0.md:470
U011 scaffold NEXT STEPS inside setup	FIXED	`bcfa364` (#569); setup:2699 `--no-next-steps`
U012 `digest_source` holds a path	FIXED	`396c0ff` (#570, #540); sm:3617
U013 LOCKSTEP RULE on every finding	FIXED	`54041cc` (#575, #539); vp:107, :1864
U014 init --dry-run skips the reuse check	FIXED	`2989b4c` (#573, #537) and `d4a7fab` (#622, #609); fam:3406-3411
U015 bump --to an unpushed commit	FIXED	`8890f57` (#568, #541); fam:8403-8407
U016 bootstrap with no git on PATH	FIXED	`24a977c` (#745, #694); family bootstrap.py:295-310, assembly :278
U017 = U130 naming gate Unicode errors	FIXED	`c15a49e` (#826, #714); validate-repository-naming.py:990, :1008-1010
U018 two `git-not-runnable` rules	OPEN -> #928	vf:289-292, :1075-1078; vp:570-616. #430 covers the `.exists()` guard
U019 `member-gitlink-unnamed` repeats	FIXED	`6626a3c` (#747, #681); vf:2100-2105
U020 shape-pin row with no `path:`	FIXED	`6626a3c` (#747, #681); vp:1161-1170; vf:1175-1184
U021 duplicate YAML key	OPEN -> #917	rs:1052, :1061, :1063, :724
U022 manifest helpers exist twice	OPEN -> #925	fam:3127-3350; sm:4781-5203
U023 `--org OpenSoft`	FIXED	`bcfa364` (#569, #538); setup:1495 casefold
U024 vf remedies name no branch step	FIXED	`54041cc` (#575, #539); vf:237 `_switch_line`
U025 `--into` sentence	FIXED	`e62b5e1` (#738, #695); README.md:912; handbook.html:1372
U026 step (7) hand clone, protocol.file	FIXED	`bcfa364` (#569); setup:2758
U027 scaffold --dry-run skips remotes	FIXED	`2989b4c` (#573, #537); scaffold:2022
U028 prompt-typed name not checked	FIXED	`4dea85e` (#756, #729); setup:1734-1737
U029 file where the family folder goes	FIXED	`bcfa364` (#569); setup:2212
U030 `-- --dry-run` "WERE created"	FIXED	`bcfa364` (#569); setup:2385
U031 = U131 CONTROL_RE C0 only (L4-24)	OPEN, not filed	setup:442; documented at v1.1.0.md:746-748; posting needs a ruling
U032 unquoted `cd` and backstop paths	FIXED	`bcfa364` (#569); setup:3021, :2759-2763
U033 docs: `<into>/<Project>`	OPEN -> #921	AGENTS.md:96; handbook.html:1997
U034 L2-15 two entry spellings	OPEN -> #924	scaffold:1806-1826; adopt-project.py:13393-13405
U035 L2-16 pin template header	FIXED	`396c0ff` (#570); neutral-product-pin.yaml:10-11
U036 L2-20 `Username for`	OPEN -> #915	scaffold:469, :1597, :1643, :1761
U037 L2-21 ssh, helper, askpass prompts	OPEN -> #915	scaffold:469
U038 L2-22 ssh-rewrite reason and remedy	OPEN -> #915	scaffold:551-552, :631-633
U039 L2-23 `visibility:` (and its family addition)	OPEN -> #913 (needs a ruling)	scaffold:1964; templates/assembly-root/project.yaml:57-58; fam:2665
U040 L2-28 `--no` ambiguous	OPEN, not filed	scaffold:367, :420; its own direction is "none needed"
U041 L2-29	quoting FIXED by `3d81259` (#821, #711), setup:2910; autocrlf OPEN -> #926	setup:2758-2763 against :2841-2851
U042 L2-32 path in `digest_source:`	FIXED (documented)	`190133d` (#630, #563); v1.1.0.md:696-699
U043 L2-38 `member-gitlink-absent` remedy	FIXED	`6626a3c` (#747, #681); vf:560-571
U044 bump pins a mount-only commit	FIXED	`8f453d0` (#625, #613); fam:8248, :8283
U045 tag on no branch, `member-to-unknown`	STALE	that refusal is gone; the asymmetry left is ruled D2 option 1 (fam:8370-8393), and its notes half is U068
U046 `family.yaml` committed as the work tree has it	FIXED	`fa04d2c` (#816, #684); fam:1423 `_checked_manifest_committed`
U047 sparse rules leave `.gitmodules` out	FIXED	`6f760b7` (#626, #612); fam:1634
U048 merge, cherry-pick or rebase in progress	FIXED	`6f760b7` (#626); fam:1286
U049 add: check the commit after the fact	OPEN -> #920	fam:1757
U050 docs: one spelling of `members/`	FIXED	`e62b5e1` (#738, #695); README.md:967; AGENTS.md:395
U051 `members` file, sparse `members/`	FIXED	`6f760b7` (#626); fam:5760, :1634-1660
U052 after `git mv members Members`	FIXED	`98f1497` (#767, #685); fam:3585
U053 pulled `Members/README`	FIXED	`98f1497` (#767); test_family_add_spellings_remedies.py:719
U054 `git mv --sparse` needs git 2.34	FIXED	`98f1497` (#767); fam:3665
U055 family name Windows cannot hold	FIXED	`42aed78` (#809, #688); fam:752
U056 `tracking_branch` names a tag	FIXED	`40e678e` (#805, #690); siblings.py:812
U057 staging name exists	FIXED	`40e678e` (#805); siblings.py:1175
U058 dangling link named after the family	FIXED	`40e678e` (#805); siblings.py:1273
U059 README and handbook `make siblings`	FIXED	`e62b5e1` (#738, #695); README.md:932-935
U060 L2-61 unseeable private repo reads absent	OPEN -> #913	sm:1724-1742
U061 = U169 L2-62 = L2-98 step (4) reads a sibling checkout	OPEN -> #907	setup:2499-2501; validate-repository-naming.py:1194; rs:3133
U062 L2-63 `passed_values` docstring	OPEN -> #923	setup `passed_values`
U063 L2-64 two folder rules	OPEN -> #923	sm:1568, :1665; setup:2274-2324
U064 L2-65 pyflakes baseline	OPEN -> #923	six warnings, quoted in #923
U065 narrowed fetch refspec	FIXED	`6f64eea` (#795, #686); fam:8125
U066 fork `origin` (known limit)	OPEN -> #921 (notes)	fam:8095; in no release's notes
U067 commit held only through a tag	FIXED	`6f64eea` (#795); fam:8210
U068 `add --at` tag-only pin, notes	OPEN -> #921 (notes)	fam:8370-8393; v1.0.0.md:260
U069 stale tracking refs	FIXED (#459's half)	`6f64eea` (#795); fam:8740 `--prune`. bump-leg's half is #898
U070 no remote named `origin`	FIXED	`6f64eea` (#795); fam:8281-8287
U071 pre-#625 mount-only pin (known limit)	OPEN -> #921 (notes)	in no release's notes
U072 CONIN$ on Windows	OPEN, not filed	Brett Heap's ruled known limit; setup:1633; documented at v1.1.0.md:708-710
U073 Git Bash gets PowerShell quoting	FIXED	`3d81259` (#821, #711); rs:3436; residual at v1.1.1.md:983
U074 L2-70 remedies lack `--family-root`	OPEN -> #927	fam:6147-6148, :6263-6264; bump-leg.py:254-260
U075 (d) remedy un-registers a member	FIXED	`1eeef3b` (#670, #646); fam:3586-3587
U076 (a) `family bootstrap ok`	FIXED	`dc7d3fb` (#682, #648); family bootstrap.py:33-34
U077 (b) order of member checks	FIXED	`dc7d3fb` (#682); vf:53, :1058
U078 (c) `--no-members` command	FIXED	`dc7d3fb` (#682); vf:594
U079 deleted `members` file	FIXED	`fa04d2c` (#816, #684); fam:5670
U080 git bisect	FIXED	`fa04d2c` (#816); fam:1054
U081 empty sparse-checkout file	FIXED	`fa04d2c` (#816); fam:1617-1618, :1711
U082 reftable marks	FIXED	`d8b8eb3` (#846, #684); fam:1119, :1190
U083 empty `rebase-merge/`	FIXED	`fa04d2c` (#816); fam:1131-1135
U084 merge conflict in `family.yaml`	FIXED	`fa04d2c` (#816); fam:316-322
U085 `merge --squash`, `cherry-pick -n`	OPEN, not filed	benign by its own text; fam:1285-1320
U086 newline path in the refusal	FIXED	`d66a905` (#731, #692); fam:1381
U087 L2-77 raw folder paths	FIXED	`d66a905` (#731, #692); sm:1585, :1617, :1633
U088 L2-78 "nothing is created"	OPEN -> #921	README.md:484; setup:22, :2309, :2324; scaffold:362
U089 = U132 PinSources casefold	FIXED	`9f235da` (#827, #712); vp:1273-1300
U090 = U133 unsearchable `contracts/`	FIXED	`9f235da` (#827); vp:351 `_is_file`
U091 = U134 absolute path twice	FIXED	`9f235da` (#827); vp:282-300
U092 = U135 advice not cause-aware	FIXED	`9f235da` (#827); vp:172, :336-342
U093 = U136 workflow names raw	FIXED	`9f235da` (#827); vp:193
U094 surrogate or NUL in `project:`	FIXED	`40e678e` (#805, #690); siblings.py:698, :753
U095 `parent_folder_warning` raw	FIXED	`40e678e` (#805); siblings.py:1229-1234
U096 `family_folder_here` display name	FIXED	`6794a9f` (#703, #696); setup:1908
U097 old holder README `cd`	OPEN -> #921 (needs a ruling)	templates/family-root/README.md:10; update-shape.py:1832-1850
U098 `family_name_note` with no `name:`	OPEN, not filed	R454: not worth a change
U099 creator asked from git config only	FIXED	`42aed78` (#809, #688); fam:2454
U100 whitespace-only `GIT_AUTHOR_NAME`	FIXED	`42aed78` (#809); fam:2280-2282
U101 NEXT push forced for `+x`	FIXED	`f6d17bc` (#813, #689); fam:1750
U102 branch of 251-255 characters	FIXED	`f6d17bc` (#813); sm:1342
U103 branch named like a tag	FIXED	`f6d17bc` (#813); fam:1750
U104 = U137 fs encoding lacks a character	FIXED	`0dee22b` (#824, #710); assembly bootstrap.py:372, :426
U105 = U138 non-UTF-8 branch	FIXED	`0dee22b` (#824); bootstrap.py:273
U106 = U139 step (a) carries a surrogate	FIXED	`0dee22b` (#824); bootstrap.py:273 (R455's option (b))
U107 `submodule_urls` splitlines	FIXED	`40e678e` (#805); siblings.py:313
U108 = U140 `legs:` not a list	FIXED	`c15a49e` (#826, #714); validate-repository-naming.py:907-908; vp:1800-1810
U109 = U141 chain link not a name	FIXED	`c15a49e` (#826); validate-repository-naming.py:143
U110 = U142 note not pointing at the finding	FIXED	`9f235da` (#827); vp:1744-1748
U111 = U143 ESC in a pin entry's path	FIXED	`9f235da` (#827); vp:1459
U112 = U144 test gap M6b	FIXED	`c15a49e` (#826); tests/test_legs_not_a_list.py:476
U113 = U145 env guard per chain	OPEN -> #909	rs:1760, :3274
U114 = U146 Kelvin sign	OPEN -> #909	scaffold:1071-1075
U115 = U147 empty block key	FIXED	`75d7050` (#823, #715); rs:863
U116 = U148 `k: {?: 1}`	FIXED	`75d7050` (#823); rs:636
U117 = U149 Unicode spaces	FIXED	`75d7050` (#823); rs:248
U118 = U150 no `line N:`	FIXED	`64ec47c` (#817, #708); rs:1164-1172
U119 = U151 whole-manifest YamlError words	adopt and scaffold FIXED by `64ec47c` (#817, #708), sm:5051, :5167; family OPEN -> #929	fam:3280-3282, :3327-3329
U120 = U152 `_read_back` guard	FIXED	`75d7050` (#823); sm:4491-4494
U121 = U153 `_neutral_pin_entry` docstring	FIXED	`75d7050` (#823); scaffold:1813
U122 = U154 `[a: b: c]` words	FIXED	`75d7050` (#823); rs:703
U123 = U155 YAML subset (L4-156)	OPEN -> #917 (documented limit section)	documented at v1.1.0.md:749-751
U124 README clone line unquoted	FIXED	`cbff780` (#812, #693); fam:2669; scaffold:1984
U125 add with a non-UTF-8 remote dir	FIXED	`cbff780` (#812); fam:2582, :7024
U126 symlink-target remedy	FIXED	`cbff780` (#812); scaffold:1020; fam:2502
U127 lone surrogate in the remedy	FIXED	`cbff780` (#812); scaffold:1027; fam:2520
U128 `directory_named` skips on Windows	FIXED	`d6071c7` (#806, #757); tests/test_clone_url_bytes.py:166-174
U129 L2-74 "cannot be read" remedy	OPEN -> #930	sm:1661-1664, :1670-1683
U156 L2-79 empty `--id=`	OPEN -> #911	scaffold:801; fam:2226
U157 L2-80 empty elector or creator	OPEN -> #911	scaffold:762-763, :962; fam:2424-2426
U158 L2-81 two refusals for a passed id	OPEN -> #911	scaffold:834; fam:2096
U159 L2-82 `foo_bar` told "lowercase"	OPEN -> #911	scaffold:834; fam:2096; adopt-project.py:4512 (adopt's half is #902)
U160 L2-83 `if not given:` loops	OPEN -> #911	scaffold:747; fam:2353, :3345
U161 L2-84 topic over 50 characters	FIXED	`281df26` (#859, #838) and `a249910` (#871, #863); scaffold:918-920; fam:2202
U162 L2-85 `openGlass`	OPEN -> #911	fam:2097-2101
U163 L2-92 leg visibility read fails	OPEN -> #913	scaffold:1522
U164 L2-93 internal root, Free plan words	OPEN -> #913	sm:1795-1798, :1913-1915
U165 L2-94 internal holder reads "private"	OPEN, not filed	cannot happen on GitHub; rs:1308 is a pinned copy
U166 L2-95 no credential NOTE for a public holder	OPEN -> #913	fam:2956
U167 L2-96 root read twice	OPEN -> #913	scaffold:1302, :1309, :1484
U168 L2-97 test row gap	OPEN -> #913	tests/test_free_plan_hint_reads.py:312
U170 L2-99 link holding `=`	OPEN -> #907	validate-repository-naming.py:178; setup:2483
U171 L2-100 one `--pin-source` two ways	OPEN -> #907	setup:2414; sm:3129
U172 L2-101 validate-manifest asks no reasons	OPEN -> #909	validate-manifest.py:197-199
U173 L2-102 `--link-remedy` padding, no NOTE	OPEN -> #907	validate-repository-naming.py:184, :484-510
U174 L2-103 label on a keyed source	OPEN -> #907	validate-repository-naming.py:393
U175 L2-104 paste commands raw in quotes	OPEN, not filed	fam:898-901, :1955, :7622; documented at v1.1.1.md:972-978; posting needs a ruling
U176 `not-a-git-repo` prints `--root` raw	OPEN -> #919	rs:3303
U177 `gh-unreadable` stderr raw	OPEN -> #919	rs:1279-1284; vp:1585-1590
U178 non-file in a pin's place	OPEN -> #919	vp:528-533, :1135-1136, :1181-1183
U179 gitfile at mode 000	OPEN -> #919	vp:1016, :373-387
U180 bump-leg and vf `.git` `.exists()`	COVERED by #430 (all)	bump-leg.py:1338; vf:946, :1073
U181 block scalar drops text	OPEN -> #917	rs:910-914
U182 `--refmap=`	OPEN -> #905	fam:8740-8741 (bump-leg's half is #898)
U183 tags-into-`origin/` clash	OPEN -> #905	fam:8159-8166, :8742-8749
U184 `git config --list` splits values	OPEN -> #905	fam:8175-8178
U185 path named `origin`	OPEN -> #905	fam:8740, :8742-8745
U186 `merge --ff-only` line, non-UTF-8	OPEN -> #932	assembly bootstrap.py:955-959
U187 BOM line rewriters	OPEN -> #933	update-shape.py:1535, :1478-1481; bump-leg.py:345
U188 `members: null`	OPEN -> #920	fam:423, :430-437
V362, V363, V364	FIXED	`62ef256` (#574); fam:4293, :4312, :4315, :5174
V392, V393	FIXED	`da48256` (#635); refused by probe
V407	FIXED	`40b476d` (#571); fam:8975
V421	FIXED	`f1d71aa` (#643); scaffold:1633
V426	FIXED	`42f6411` (#641); fam:2238
V433	FIXED	`a524542` (#640); scaffold:1901
V434	FIXED	`f1c73e2` (#633); vp:257, :417
V441, V442	FIXED	`66267f0` (#576); validate-repository-naming.py:57, :256
V443	FIXED	`5b29ee1` (#631); validate-manifest.py:120
V444	FIXED	`1babe42` (#638); rs:1680, :1721
V450	FIXED	`396c0ff` (#570); sm:3595
V454	FIXED	`0ed1b84` (#642); siblings.py:1474
V455	FIXED	`cd5ffa2` (#644); bootstrap.py:308

Not filed, with the reason

U165 (L2-94): the state needs an internal repository on a Free plan, which GitHub does not have.
U040 (L2-28): its own direction is "none needed"; argparse refuses before anything is created.
U085: benign by its own text; the verb commits only its own paths.
U098: R454 says it is not worth a change; only an invalid manifest reaches it.
U175 (L2-104) and U031 = U131 (L4-24): documented known limits (v1.1.1.md:972-978 and v1.1.0.md:746-748); posting either needs a ruling. Their lines stay unchecked, with this note.
U072 (CONIN$): Brett Heap's ruled known limit, documented at v1.1.0.md:708-710.

Not on #459's checklist: candidates the closing pull requests name

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 left two states: a file named <Family> stops the printed mkdir -p, and an existing <Family>/<Family> receives the holder inside it.
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's "Checklist candidates": init's landing and folder plan rows, and seven refusals, print paths raw. This overlaps U175.
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: a branch part named CON or NUL is unassigned.
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 body lists eight "For Tracking: setup, scaffold, family and the shared helpers, creating projects and families #459's checklist" lines from lane 1's review that are not posted here, such as splitlines shifting line N.

Lane: openreposhape-3 (openRepoShape-3)

family.py add: a commit that records the member's gitlink anywhere bu…

family.py init: where the holder's own YAML reader refuses the manife…

family.py: `bump`'s fetch no longer moves the mount's own branches, w…

the family root's siblings.py and bootstrap.py: a member's path this …

family.py and bump-leg.py: every remedy that names `family.py bump` n…

setup-project.py: the self-bootstrap refuses an `$OPENREPOSHAPE_REPO`…

(lane openRepoShape-5, 2026-10-23T12:05:18Z)

Lane 5 filed three follow-ups that belong to this tracker. They come from its independent review of #1166 (#1139), findings 1, 4 and 8 (#1166 issuecomment-6352173084), and each was re-measured at main 961640c6.

repo_shape.py: NamingPolicy reads a name with .match against patterns that end in $, so a name that ends in a line feed is classified as the name without it, and validate-repository-naming.py exits 0 over such a leg repository (#1166 follow-up) #1202: repo_shape.py's NamingPolicy reads names with .match against patterns that end in $. A name ending in a line feed is classified as the name without it, so validate-repository-naming.py exits 0 over such a leg repository.
validate-family.py: no pattern is asked of a member row's id:, so an id that ends in a line feed or holds a space is family ok under --no-members, and a row that main's family.py add wrote stays family ok (#1166 follow-up) #1204: validate-family.py asks no pattern of a member row's id:. Under --no-members, a row id that ends in a line feed or holds a space is family ok, and a row that main's family.py add wrote stays family ok.
validate-manifest.py and validate-family.py: values no pattern reads (name, reference, elected_by, a leg's path, org, tracking_branch, digest_algorithm) are manifest ok or family ok with a trailing line feed (#1166 follow-up) #1205: values that no pattern reads (name, reference, elected_by, a leg's path, org, tracking_branch, digest_algorithm) are manifest ok or family ok with a trailing line feed. No tool reads digest_algorithm at all.

All three are post-tag follow-ups, with no milestone.

Lane: openreposhape-5 (openRepoShape-5)

Navigation Menu

Repository navigation

Tracking: setup, scaffold, family and the shared helpers, creating projects and families #459

143 remaining items

The table

Not filed, with the reason

Not on #459's checklist: candidates the closing pull requests name

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Tracking: setup, scaffold, family and the shared helpers, creating projects and families #459

Description

Activity

143 remaining items

brettheap commented on Oct 18, 2026

brettheap commented on Oct 18, 2026

brettheap commented on Oct 18, 2026

The table

Not filed, with the reason

Not on #459's checklist: candidates the closing pull requests name

brettheap commented on Oct 23, 2026

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions