family.py and validate-family.py: the git switch -c family/<act>-<member> remedy fails when an earlier act left that branch, and the docs spell the branch word two ways

The git switch -c family/<act>-<member> remedy should say what to do when the branch exists, and the docs and NEXT lines should use one placeholder.

From the checklist triage of 2026-10-15; Brett Heap approved its recommendations at 2026-10-15T16:49:38Z. One pull request covers the lines below.

Lands after: #575.

Lines

L2-46: the git switch -c family/<act>-<member> remedy fails "already exists" when an earlier act left that branch. Where: scripts/family.py _checked_branch (:688-693), _switch_command (:673-676); validate-family.py _switch_line (on validate-pins.py and validate-family.py: THE LOCKSTEP RULE closes the count only where a leg or member is out of step, a lone digest or workflow-reference mismatch names its bump-leg.py line, and holder remedies branch first (#539) #575). Effort S; all. On main 26c1600: Probe p3: git switch -c family/add-atlas with that branch already present gives fatal: a branch named 'family/add-atlas' already exists (exit 128); _checked_branch's remedy at 26c1600 prints the line whatever exists. The validator half appears only on validate-pins.py and validate-family.py: THE LOCKSTEP RULE closes the count only where a leg or member is out of step, a lone digest or workflow-reference mismatch names its bump-leg.py line, and holder remedies branch first (#539) #575's head 0680799 (main's validate-family.py has no switch -c), so the review's "fix both or neither" ties it to validate-pins.py and validate-family.py: THE LOCKSTEP RULE closes the count only where a leg or member is out of step, a lone digest or workflow-reference mismatch names its bump-leg.py line, and holder remedies branch first (#539) #575 (ct461x459-D2). (source)

Rides along (backlog; tick it here if the same change covers it)

L2-45: family docs and family.py NEXT spell the branch family/add-<project> beside --member <org>/<Project>. Where: AGENTS.md:372; scripts/family.py:1328; templates/family-root/README.md:111; templates/family-root/AGENTS-shape.md:27. Effort S; all. On main 26c1600: Cosmetic, and true at all four places at 26c1600; 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 (3e3c833) does not touch these lines, and validate-pins.py and validate-family.py: THE LOCKSTEP RULE closes the count only where a leg or member is out of step, a lone digest or workflow-reference mismatch names its bump-leg.py line, and holder remedies branch first (#539) #575 adds a third spelling, family/<verb>-<Project>, in validate-family.py, so it rides with L2-46 (AGENTS-shape.md is pinned, scripts/shape_materialize.py:536). (source)

Each line above is marked → this issue on its tracking issue. Tick it there too when this closes.

The lines as written on the tracking issues

L2-46 (nit; templates/family-root/scripts/validate-family.py and scripts/family.py, #575 review): git -C <holder> switch -c family/<act>-<member> fails with "already exists" when an earlier act left a local branch of that name, and the remedy says nothing of it.

_switch_line (templates/family-root/scripts/validate-family.py:119-134 at 93b68f6) prints the switch -c line for the add and remove remedies. family.py words its own family.py add, bump and remove commit straight onto the holder's tracking branch, and add's NEXT pushes a <branch> that does not exist, though AGENTS.md says to land each as a pull request #487 refusal the same way: _switch_command (scripts/family.py:673-676) in the remedy of _checked_branch (:688-693).

rv539 ran it: with family/add-<member> still local from the first try, the switch -c line fails and the person has to know git switch <branch> (review575-93b68f6/tools/progress.md, the 23:54:29Z entry). It happens whenever an earlier act left that branch behind.

Direction: the review's rule is to fix both or neither. Say what to do when the branch exists (git switch <branch>, or another name) in the validator's lines and in family.py's remedy. The validator is a pinned copy, so bundle that half with the next change to it.

— #459, comment

L2-45 (nit; templates/family-root/AGENTS-shape.md, templates/family-root/README.md, AGENTS.md and scripts/family.py, #575 writer): the family docs and family.py's NEXT STEPS spell the branch family/add-<project> beside --member <org>/<Project>, two spellings for one value.

At 0680799: templates/family-root/README.md:111, templates/family-root/AGENTS-shape.md:27, AGENTS.md:372 and scripts/family.py:1328. Pinned validators: THE LOCKSTEP RULE only where it applies, and remedies that name the git switch -c step (one shape bump) #539's fix round gave the holder's validator one placeholder, PROJECT_PLACEHOLDER = "<Project>", for the branch word and --member alike (templates/family-root/scripts/validate-family.py:118-121, _switch_line at :124-140). The four places above still show the pair.

The real branch word is the project's name lowercased (_suggested_branch, scripts/family.py:661-670). <project> hints at that and nothing says it. rv539b confirmed the four lines and calls them docs outside Pinned validators: THE LOCKSTEP RULE only where it applies, and remedies that name the git switch -c step (one shape bump) #539's two validators (review575-93b68f6/reverify-0680799.md, "Not findings").

Direction: one placeholder in all four places, or one sentence saying the branch word is the name in lowercase. AGENTS-shape.md is a pinned copy (scripts/shape_materialize.py:536), so each holder takes that edit as one upstream-changed row; bundle it with the next change to the file. The holder's README is rendered once and is not pinned.

— #459, comment

CLAIMED — lane openRepoShape-5, session 50cede29-ab82-4ca3-9f67-d98e1a58c8d2@Eagle, 2026-10-15T22:26:38Z, for #607

Logged in lanes/log/openRepoShape-5.md at 1ea2ca44d413c7b4c1a0f105e9544dfd0f85c35b.

The three reads (lane-collision-protocol Rule 1):

1. existing `CLAIMED —` comments on opensoft/openRepoShape#607:
   none
2. `gh pr list --repo opensoft/openRepoShape --state all --search "#607"`, kept where the row or its body names `#607`:
   none
3. `git ls-remote --heads git@github.com:opensoft/openRepoShape` branches naming `607` as a whole token:
   none

family.py and validate-family.py: where an earlier act left the branc…

family.py, validate-family.py and test_family_switch_branch_exists.py…

Merge main 95780dd (#637, #607) into fix/611-validate-family-head-git…

validate-family.py and family.py: a gitlink no section of HEAD's `.gi…

Merge main 95780dd (#638, #635, #640, #623, #637) into fix/610-pinned…

bump-leg.py and setup-project.py: the `git switch -c` line says under…

Pinned shape copies: one bump, so AGENTS-shape.md is true of every ro…

Navigation Menu

Repository navigation

family.py and validate-family.py: the `git switch -c family/<act>-<member>` remedy fails when an earlier act left that branch, and the docs spell the branch word two ways #607

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

family.py and validate-family.py: the git switch -c family/<act>-<member> remedy fails when an earlier act left that branch, and the docs spell the branch word two ways #607

Description

Activity

brettheap commented on Oct 15, 2026

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions

family.py and validate-family.py: the `git switch -c family/<act>-<member>` remedy fails when an earlier act left that branch, and the docs spell the branch word two ways #607