Move every release up one and put lilleman's budget and efficiency items in a new 0.2.0

This commit is contained in:
2026-10-16 04:07:04 +02:00
parent ac69f6da6d
commit 7570af6191
2 changed files with 120 additions and 118 deletions
+2
View File
@@ -132,6 +132,8 @@ Valid while the ladder orders its rungs by competence.
2026-10-16, lilleman, answering the review of PR #283. An item that restores a maintain goal the project breaks today is a defect: exempt as `defect`. Its Moves name that maintain goal and how much of it is broken, and the goal weighs as the heaviest achieve goal, since a broken maintain goal outranks any achieve goal. Every defect ranks by its score, so a trivial one waits behind valuable work. Only questions and then security findings rank first in their release: a question's answer may change other items, and a question waits for the human anyway, so ranking it first delays no chunk. An item names the most specific goal it moves, and a parent only for a move none of its sub-goals covers, so a sub-goal's move does not count twice; `rank_todo.py` refuses a goal named beside its own sub-goal. Each project reshapes its own `todo.md`; `CLAUDE.md` → "Working loop" makes the reshape the only work until done, as it makes writing missing premises. An item that makes this project easier to read moves goal 1.4 as well as 1.1, since reading cost is measured in tokens per chunk; a goal of its own for readability is left for later. Cost gains a fixed part, F in `Score = 100 * V / (2**S + F)`: every chunk carries a cost its size does not shrink, the reading before its first commit, about 130k tokens, and its review passes, so `2**S` alone overrated tiny items. F is 8, about one S 3 change, until `todo.md` item 158 measures it.
2026-10-16, lilleman, ruling on the plan in PR #283: release 0.2.0 takes plan items 146, 152, 73, 50, 65, 84, 153, 67, 115 and 140 ahead of their scores, to put budget and efficiency in the other projects first. A standing goal cannot capture that priority for now, so it is a one-time ruling: the later releases, and every release after 0.2.0, fill from the score again.
2026-10-15, lilleman. Serves goal 1.4: loop-claude routes a chunk's first cycle on the item the chunk takes (`routing: one ladder`). An item lists the items that must be done before it starts in a Blocked by column, and `rank_todo.py` ranks each blocker above every item it blocks. The table's order is `todo.md`'s only order: `rank_todo.py` fails on text beside the table under `## Items`, since a prose order there, "Release 0.3.0 takes items 125, 114 and 124 first", contradicted the table and loop-claude could not read it. A decision that wants an order sets the scores instead. An item waits while it is marked `question` or a blocker of it is still in the table, whatever the rows' order, so a release's first row that does not wait is the item a chunk takes. A blocker the table no longer holds was finished or deleted, and is dropped, so the cell shows what still blocks. A panel of three readers seated as lilleman picked the spelling: two ranked "Blocked by" first, as the cell lists only what still blocks, and all three chose bare IDs such as `12, 14` in a column right after Release, read before the tagline and kept apart from the scores. `rank_todo.py` adds the column to a table without it. The tools read cells by column name, so a later column leaves a reader on an older release working; `rank_todo.py`, which writes every column, fails on one it does not know rather than drop it.
## todo.md: only a session or subagent on the project's rung or above scores or reshapes the plan; one below it drafts new items and changes to filed ones for a subagent at or above that rung to apply and score, and removes finished ones
+118 -118
View File
@@ -13,68 +13,68 @@
| ID | Release | Blocked by | Exempt | Item | R | S | C | Moves | Score |
|---|---|---|---|---|---|---|---|---|---|
| 115 | 0.2.0 | | question | **Confirm or set how many finished chunks that each shipped one item `score-outcomes` waits for before it shows a score's medians, today 10.** | 1 | 1 | 2 | 1.2 by 5 | 7.3 |
| 80 | 0.2.0 | | defect | **State in README → Install that every machine taking this checkout must be a sandbox, as `settings.json`'s auto mode assumes.** | 1 | 1 | 1 | 2 by 4 | 17.5 |
| 21 | 0.2.0 | | defect | **Give `settings.json`'s model one writer, and sort its keys.** | 3 | 3 | 4 | 2 by 5 | 13.7 |
| 29 | 0.2.0 | | | **Make the contributor a standing member of every project's audience, unless its `## Audience` says maintainability does not matter.** | 2 | 2 | 3 | 1.1 by 5, 1.4 by 5 | 13.7 |
| 119 | 0.2.0 | | | **Retitle the comprehension-floor decision so it says a chunk lifting a project towards the floor works first on its lowest-scoring dimension, and that the floor blocks nothing.** | 1 | 1 | 1 | 1.1 by 4, 1.4 by 4 | 13.1 |
| 127 | 0.2.0 | | | **Define seat, place, paired ruling, scoring run and the judge's records in `CLAUDE.md` → "Vocabulary".** | 1 | 1 | 1 | 1.1 by 4, 1.4 by 4 | 13.1 |
| 129 | 0.2.0 | | | **Explain `bin/rule_hook/rules.py`'s `OVERRIDE` pattern beside it: why the reason is lazy, what the lookahead splits and why `-->` is stripped.** | 1 | 1 | 1 | 1.1 by 4, 1.4 by 4 | 13.1 |
| 114 | 0.2.0 | | decision | **Run git through `git_repo.py` in `rule_hook`, `metric_bench`, `reading_cost`, `stability_replay` and `loop_claude`, keeping rule-hook's guards as an option, so one module owns how a tool runs git.** | 3 | 4 | 4 | 1.1 by 9, 1.4 by 9 | 12.3 |
| 120 | 0.2.0 | | | **Rewrite larv-review §3's "Round cap" and "Re-entry" as one table that says, for each kind of fix, which passes rerun and what a round reviews.** | 2 | 4 | 5 | 1.1 by 6, 1.2 by 6, 1.4 by 6 | 11.9 |
| 49 | 0.2.0 | | principle | **Check that `AGENTS.md` lists exactly the titles of `docs/decisions.md`.** | 1 | 2 | 2 | 1.1 by 6 | 10.9 |
| 82 | 0.2.0 | | principle | **Sort `AGENTS.md`'s decision titles and `docs/decisions.md`'s entries by title.** | 1 | 2 | 2 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 117 | 0.2.0 | | | **Give `todo_table.items` one way to say a plan has no items table, so no caller tells it from an empty plan by truthiness.** | 2 | 2 | 3 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 126 | 0.2.0 | | | **State when work has landed towards a release once, in `bin/larv_review/plan.py`, and have larv-review's help point to it.** | 1 | 2 | 2 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 131 | 0.2.0 | | | **Describe the transcript archive's layout in `bin/transcript_archive/`, which owns it, and have loop-claude's help point there.** | 1 | 2 | 1 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 133 | 0.2.0 | | | **Show, beside `bin/rule_hook/override_log.py`'s `this_chunks`, one example per case it decides: an amended or rebased commit, a renamed branch, and a commit on the default branch.** | 1 | 2 | 2 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 134 | 0.2.0 | | | **Name in `bin/shell_command.py`'s `segments` the shell forms it does not parse, so a caller knows what it misses.** | 1 | 2 | 2 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 140 | 0.2.0 | | | **Split `docs/decisions.md` → `routing: one ladder` into one entry per decision, so a reader finds the rules for the ladder, the worker's rung, the routing figures and delegation by their titles.** | 1 | 2 | 1 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 154 | 0.2.0 | | defect | **Have `bin/shell_command.py`'s `segments` keep a command's words together across a redirect, so `git > out commit` reads as one git commit.** | 2 | 2 | 3 | 1.1 by 6 | 10.9 |
| 157 | 0.2.0 | | | **Split `docs/decisions.md` → `todo.md: one table of items …`'s 2026-10-02 paragraph into one paragraph per choice.** | 1 | 2 | 1 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 106 | 0.2.0 | | | **Make larv-review's "Do not over-read a clean result" cite a rule that says Renovate or Dependabot owns dependency CVEs, or drop the citation.** | 1 | 1 | 1 | 1.1 by 5 | 10.9 |
| 139 | 0.3.0 | | | **Split `bin/routing_figures/measure.py` by concern, and keep the figures' measured cycle and each raised call as fields loop-claude reads back, not as the prose it renders.** | 2 | 3 | 3 | 1.1 by 5, 1.4 by 5 | 10.3 |
| 158 | 0.3.0 | | decision | **Measure the fixed cost of a chunk from loop-claude's archive, in tokens and hours against its items' S, and set F in `Score = 100 * V / (2**S + F)` to it.** | 1 | 3 | 4 | 1.1 by 5, 1.4 by 5 | 10.3 |
| 149 | 0.3.0 | | decision | **Fill each release by priority up to a capacity it states, in place of the bar, and move what does not fit to a later release.** | 3 | 3 | 3 | 1.1 by 7 | 9.6 |
| 116 | 0.3.0 | | | **Report L by the share of chunks with a 🔴, once 10 chunks that each shipped one item record finding counts and most record no 🔴.** | 1 | 2 | 2 | 1.1 by 5 | 9.1 |
| 110 | 0.3.0 | | | **Count, for each finished chunk, the later changes that fixed code it wrote, and show them beside its guessed E.** | 2 | 4 | 6 | 1.1 by 6, 1.4 by 6 | 8.2 |
| 112 | 0.3.0 | | | **Count an item as shipped only where its chunk built it, so an item a chunk dropped from the plan gets no outcome.** | 2 | 3 | 5 | 1.1 by 6 | 8.2 |
| 130 | 0.3.0 | | | **Move the vote counting out of `bin/larv_review/ruling.py`'s `votes`, so reading a votes file decides nothing.** | 2 | 3 | 3 | 1.1 by 4, 1.4 by 4 | 8.2 |
| 137 | 0.3.0 | | | **Archive a chunk's route as a record loop-claude reads back field by field, in place of the `Worker:` prose `ROUTED` parses, and name `stepped` and the judge's `bundled` by what they do.** | 2 | 3 | 3 | 1.1 by 4, 1.4 by 4 | 8.2 |
| 125 | 0.3.0 | | decision | **Give "record", "outcome" and "places" one meaning each across `bin/`, and rename `bin/larv_review/git.py` by what it holds.** | 2 | 5 | 3 | 1.1 by 9, 1.4 by 9 | 7.4 |
| 81 | 0.3.0 | | principle | **Have `workflow-bench/bench.py` return each run's failure as a value, so one failed clone does not hide the later runs' results.** | 2 | 2 | 3 | 1.1 by 4 | 7.3 |
| 102 | 0.3.0 | | | **Give loop-claude's retry delay an environment override, so the black-box test stops replacing `time.sleep` through `sitecustomize`.** | 2 | 2 | 2 | 1.1 by 4 | 7.3 |
| 66 | 0.3.0 | | principle | **Hold systems-architect's method thresholds only while the comprehension panel finds the flagged code hard.** | 3 | 3 | 4 | 1.1 by 5 | 6.8 |
| 67 | 0.3.0 | | principle | **Point stability-reviewer at the technical principles in place of its own restatement of them.** | 2 | 3 | 2 | 1.1 by 5 | 6.8 |
| 101 | 0.3.0 | | principle | **Have `larv_review/record.py`'s `read_seats` take the time, the commit check and the history as inputs, so it parses and decides without effects.** | 3 | 3 | 3 | 1.1 by 5 | 6.8 |
| 111 | 0.3.0 | | | **Measure the chunks of a project that merges by squash or fast-forward, which `score-outcomes` skips today.** | 3 | 3 | 5 | 1.1 by 5 | 6.8 |
| 113 | 0.3.0 | | | **Compute an item's S from the code it names, where it names any, and show it beside the finished items `score-outcomes` lists.** | 3 | 4 | 6 | 1.1 by 5, 1.4 by 5 | 6.8 |
| 121 | 0.3.0 | | | **Check that `bin/round_titles.py`'s `PASSES` lists exactly the passes of larv-review §3's table.** | 1 | 3 | 2 | 1.1 by 5 | 6.8 |
| 122 | 0.3.0 | | | **Assemble the judge's prompt in `bin/loop_claude/cycle/judge.py`, importing only wording the worker also sees, and call the worker "the worker" in it.** | 3 | 4 | 4 | 1.1 by 5, 1.4 by 5 | 6.8 |
| 4 | 0.4.0 | | | **Name each chunk's archive folder by the chunk's start time, so `<key>/` lists chunks in the order they ran.** | 3 | 2 | 3 | 1.1 by 3 | 5.5 |
| 12 | 0.4.0 | | | **Run the test suite in CI on every PR, or record in `docs/decisions.md` why the local run is enough.** | 2 | 4 | 3 | 1.1 by 6 | 5.5 |
| 50 | 0.4.0 | | principle | **Grant the bench's runs only the tools a review needs, in place of skipping every permission.** | 2 | 2 | 3 | 1.1 by 3 | 5.5 |
| 65 | 0.4.0 | | defect | **Archive each judge reply the loop rejects.** | 3 | 4 | 3 | 3 by 3 | 5.5 |
| 105 | 0.4.0 | | decision | **Compare review rounds per pass and reading cost for the five chunks before and after the `worker:` decision moved the principles out of the worker's context, and file an item if either rose.** | 1 | 2 | 3 | 1.4 by 6 | 5.5 |
| 135 | 0.4.0 | | | **Fingerprint only the modules loop-claude imports, so editing another tool's code no longer reloads a running loop.** | 3 | 3 | 4 | 1.1 by 4 | 5.5 |
| 156 | 0.4.0 | | decision | **Drop `rank_todo.py`'s ranking of an old-shape table once no project under `~/code` holds one.** | 2 | 3 | 2 | 1.1 by 4 | 5.5 |
| 68 | 0.4.0 | | principle | **Test `workflow-bench/compare.py`'s blinding and `bench.py`'s run plan.** | 2 | 4 | 4 | 1.1 by 5 | 4.6 |
| 84 | 0.4.0 | | | **Check, at each release's scoring run, that seats in a paired ruling agree more often than two scoring runs on near-identical code do, as the paired-ruling decision assumes.** | 2 | 2 | 5 | 1.4 by 5 | 4.6 |
| 153 | 0.4.0 | | decision | **Run the judge on the rung the highest R + C in its chunk picks, an item the worker takes in included.** | 2 | 3 | 4 | 1.2 by 5 | 4.6 |
| 59 | 0.4.0 | | | **Bound a run of unfinished verdicts on one chunk, so a worker and judge that disagree cannot cycle unattended without end.** | 3 | 3 | 5 | 1.4 by 6 | 4.1 |
| 123 | 0.4.0 | | | **Have `bin/comprehension_history.py` parse each history entry once into seats, votes and places, so `latest_places`, `worse_places` and `worse_votes` neither depend on table order nor split a label to find a seat.** | 3 | 5 | 5 | 1.1 by 5, 1.4 by 5 | 4.1 |
| 132 | 0.4.0 | | | **Lift each dimension still under the comprehension floor after release 0.2.0's scoring run, at the places that run names.** | 3 | 5 | 5 | 1.1 by 5, 1.4 by 5 | 4.1 |
| 155 | 0.4.0 | | | **Lower the reading cost of the files a chunk reads most before its first commit.** | 2 | 3 | 4 | 1.4 by 6 | 4.1 |
| 124 | 0.4.0 | | decision | **Group the shared modules at `bin/`'s root into packages named by purpose, so the root holds only launchers and stand-alone tools.** | 3 | 6 | 3 | 1.1 by 9, 1.4 by 9 | 4.1 |
| 1 | 0.5.0 | | question | **Say in README → Install how the transcript archive is backed up, once bender-infra backs up ai-sandbox's data volume.** | 3 | 5 | 2 | | 0.0 |
| 146 | 0.5.0 | | decision | **Score R as L + E + I − 2 from anchored levels, at least 7 at I5 and 5 at I4, with L, E and I in their own columns and `rank_todo.py` computing R, drop C, and route a task's model and effort on its R alone.** | 6 | 7 | 7 | 1.1 by 8, 1.2 by 8, 1.4 by 8 | 2.8 |
| 152 | 0.5.0 | 146 | decision | **Anchor each score level in `rules/todo-items.md` with words, and each project's with two or three items it shipped, and have a scorer compare a new item against them.** | 2 | 4 | 3 | 1.1 by 7 | 6.4 |
| 5 | 0.5.0 | | decision | **Key a project on its checkout's full path, and refuse to start while another session uses that key.** | 5 | 5 | 7 | 1.1 by 6 | 3.3 |
| 77 | 0.5.0 | | | **Teach `bin/metric-bench` Go and TypeScript, rerun it once two projects' comprehension histories name places, and bring into review only a metric that tracks the panel.** | 3 | 6 | 6 | 1.1 by 7, 1.4 by 7 | 3.2 |
| 6 | 0.5.0 | | decision | **Move what `loop-claude` writes out of `~/.claude.json`, and ask the human about each write that must stay.** | 5 | 5 | 6 | 1.1 by 5 | 2.7 |
| 51 | 0.5.0 | | principle | **Hand loop-claude's terminal output to an injected effect, so no storage or retry function prints.** | 3 | 5 | 5 | 1.1 by 5 | 2.7 |
| 151 | 0.5.0 | | decision | **Keep 10 to 20 settled decisions per project as its goals' test set, and rerun them after a goal changes.** | 3 | 5 | 5 | 1.2 by 7 | 2.6 |
| 73 | 0.5.0 | | | **Have each reviewer write its full report to a file and return the worker only its findings' one-line summaries.** | 3 | 4 | 4 | 1.4 by 4 | 1.8 |
| 52 | 0.5.0 | | principle | **Record why this checkout lives in Claude Code's own directory, or move it out and link it in.** | 6 | 6 | 4 | 1.1 by 4 | 1.2 |
| 67 | 0.2.0 | | principle | **Point stability-reviewer at the technical principles in place of its own restatement of them.** | 2 | 3 | 2 | 1.1 by 5 | 6.8 |
| 146 | 0.2.0 | | decision | **Score R as L + E + I − 2 from anchored levels, at least 7 at I5 and 5 at I4, with L, E and I in their own columns and `rank_todo.py` computing R, drop C, and route a task's model and effort on its R alone.** | 6 | 7 | 7 | 1.1 by 8, 1.2 by 8, 1.4 by 8 | 2.8 |
| 152 | 0.2.0 | 146 | decision | **Anchor each score level in `rules/todo-items.md` with words, and each project's with two or three items it shipped, and have a scorer compare a new item against them.** | 2 | 4 | 3 | 1.1 by 7 | 6.4 |
| 50 | 0.2.0 | | principle | **Grant the bench's runs only the tools a review needs, in place of skipping every permission.** | 2 | 2 | 3 | 1.1 by 3 | 5.5 |
| 65 | 0.2.0 | | defect | **Archive each judge reply the loop rejects.** | 3 | 4 | 3 | 3 by 3 | 5.5 |
| 84 | 0.2.0 | | | **Check, at each release's scoring run, that seats in a paired ruling agree more often than two scoring runs on near-identical code do, as the paired-ruling decision assumes.** | 2 | 2 | 5 | 1.4 by 5 | 4.6 |
| 153 | 0.2.0 | | decision | **Run the judge on the rung the highest R + C in its chunk picks, an item the worker takes in included.** | 2 | 3 | 4 | 1.2 by 5 | 4.6 |
| 73 | 0.2.0 | | | **Have each reviewer write its full report to a file and return the worker only its findings' one-line summaries.** | 3 | 4 | 4 | 1.4 by 4 | 1.8 |
| 80 | 0.3.0 | | defect | **State in README → Install that every machine taking this checkout must be a sandbox, as `settings.json`'s auto mode assumes.** | 1 | 1 | 1 | 2 by 4 | 17.5 |
| 21 | 0.3.0 | | defect | **Give `settings.json`'s model one writer, and sort its keys.** | 3 | 3 | 4 | 2 by 5 | 13.7 |
| 29 | 0.3.0 | | | **Make the contributor a standing member of every project's audience, unless its `## Audience` says maintainability does not matter.** | 2 | 2 | 3 | 1.1 by 5, 1.4 by 5 | 13.7 |
| 119 | 0.3.0 | | | **Retitle the comprehension-floor decision so it says a chunk lifting a project towards the floor works first on its lowest-scoring dimension, and that the floor blocks nothing.** | 1 | 1 | 1 | 1.1 by 4, 1.4 by 4 | 13.1 |
| 127 | 0.3.0 | | | **Define seat, place, paired ruling, scoring run and the judge's records in `CLAUDE.md` → "Vocabulary".** | 1 | 1 | 1 | 1.1 by 4, 1.4 by 4 | 13.1 |
| 129 | 0.3.0 | | | **Explain `bin/rule_hook/rules.py`'s `OVERRIDE` pattern beside it: why the reason is lazy, what the lookahead splits and why `-->` is stripped.** | 1 | 1 | 1 | 1.1 by 4, 1.4 by 4 | 13.1 |
| 114 | 0.3.0 | | decision | **Run git through `git_repo.py` in `rule_hook`, `metric_bench`, `reading_cost`, `stability_replay` and `loop_claude`, keeping rule-hook's guards as an option, so one module owns how a tool runs git.** | 3 | 4 | 4 | 1.1 by 9, 1.4 by 9 | 12.3 |
| 120 | 0.3.0 | | | **Rewrite larv-review §3's "Round cap" and "Re-entry" as one table that says, for each kind of fix, which passes rerun and what a round reviews.** | 2 | 4 | 5 | 1.1 by 6, 1.2 by 6, 1.4 by 6 | 11.9 |
| 49 | 0.3.0 | | principle | **Check that `AGENTS.md` lists exactly the titles of `docs/decisions.md`.** | 1 | 2 | 2 | 1.1 by 6 | 10.9 |
| 82 | 0.3.0 | | principle | **Sort `AGENTS.md`'s decision titles and `docs/decisions.md`'s entries by title.** | 1 | 2 | 2 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 117 | 0.3.0 | | | **Give `todo_table.items` one way to say a plan has no items table, so no caller tells it from an empty plan by truthiness.** | 2 | 2 | 3 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 126 | 0.3.0 | | | **State when work has landed towards a release once, in `bin/larv_review/plan.py`, and have larv-review's help point to it.** | 1 | 2 | 2 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 131 | 0.3.0 | | | **Describe the transcript archive's layout in `bin/transcript_archive/`, which owns it, and have loop-claude's help point there.** | 1 | 2 | 1 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 133 | 0.3.0 | | | **Show, beside `bin/rule_hook/override_log.py`'s `this_chunks`, one example per case it decides: an amended or rebased commit, a renamed branch, and a commit on the default branch.** | 1 | 2 | 2 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 134 | 0.3.0 | | | **Name in `bin/shell_command.py`'s `segments` the shell forms it does not parse, so a caller knows what it misses.** | 1 | 2 | 2 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 154 | 0.3.0 | | defect | **Have `bin/shell_command.py`'s `segments` keep a command's words together across a redirect, so `git > out commit` reads as one git commit.** | 2 | 2 | 3 | 1.1 by 6 | 10.9 |
| 157 | 0.3.0 | | | **Split `docs/decisions.md` → `todo.md: one table of items …`'s 2026-10-02 paragraph into one paragraph per choice.** | 1 | 2 | 1 | 1.1 by 4, 1.4 by 4 | 10.9 |
| 106 | 0.3.0 | | | **Make larv-review's "Do not over-read a clean result" cite a rule that says Renovate or Dependabot owns dependency CVEs, or drop the citation.** | 1 | 1 | 1 | 1.1 by 5 | 10.9 |
| 139 | 0.4.0 | | | **Split `bin/routing_figures/measure.py` by concern, and keep the figures' measured cycle and each raised call as fields loop-claude reads back, not as the prose it renders.** | 2 | 3 | 3 | 1.1 by 5, 1.4 by 5 | 10.3 |
| 158 | 0.4.0 | | decision | **Measure the fixed cost of a chunk from loop-claude's archive, in tokens and hours against its items' S, and set F in `Score = 100 * V / (2**S + F)` to it.** | 1 | 3 | 4 | 1.1 by 5, 1.4 by 5 | 10.3 |
| 149 | 0.4.0 | | decision | **Fill each release by priority up to a capacity it states, in place of the bar, and move what does not fit to a later release.** | 3 | 3 | 3 | 1.1 by 7 | 9.6 |
| 116 | 0.4.0 | | | **Report L by the share of chunks with a 🔴, once 10 chunks that each shipped one item record finding counts and most record no 🔴.** | 1 | 2 | 2 | 1.1 by 5 | 9.1 |
| 110 | 0.4.0 | | | **Count, for each finished chunk, the later changes that fixed code it wrote, and show them beside its guessed E.** | 2 | 4 | 6 | 1.1 by 6, 1.4 by 6 | 8.2 |
| 112 | 0.4.0 | | | **Count an item as shipped only where its chunk built it, so an item a chunk dropped from the plan gets no outcome.** | 2 | 3 | 5 | 1.1 by 6 | 8.2 |
| 130 | 0.4.0 | | | **Move the vote counting out of `bin/larv_review/ruling.py`'s `votes`, so reading a votes file decides nothing.** | 2 | 3 | 3 | 1.1 by 4, 1.4 by 4 | 8.2 |
| 137 | 0.4.0 | | | **Archive a chunk's route as a record loop-claude reads back field by field, in place of the `Worker:` prose `ROUTED` parses, and name `stepped` and the judge's `bundled` by what they do.** | 2 | 3 | 3 | 1.1 by 4, 1.4 by 4 | 8.2 |
| 125 | 0.4.0 | | decision | **Give "record", "outcome" and "places" one meaning each across `bin/`, and rename `bin/larv_review/git.py` by what it holds.** | 2 | 5 | 3 | 1.1 by 9, 1.4 by 9 | 7.4 |
| 81 | 0.4.0 | | principle | **Have `workflow-bench/bench.py` return each run's failure as a value, so one failed clone does not hide the later runs' results.** | 2 | 2 | 3 | 1.1 by 4 | 7.3 |
| 102 | 0.4.0 | | | **Give loop-claude's retry delay an environment override, so the black-box test stops replacing `time.sleep` through `sitecustomize`.** | 2 | 2 | 2 | 1.1 by 4 | 7.3 |
| 66 | 0.4.0 | | principle | **Hold systems-architect's method thresholds only while the comprehension panel finds the flagged code hard.** | 3 | 3 | 4 | 1.1 by 5 | 6.8 |
| 101 | 0.4.0 | | principle | **Have `larv_review/record.py`'s `read_seats` take the time, the commit check and the history as inputs, so it parses and decides without effects.** | 3 | 3 | 3 | 1.1 by 5 | 6.8 |
| 111 | 0.4.0 | | | **Measure the chunks of a project that merges by squash or fast-forward, which `score-outcomes` skips today.** | 3 | 3 | 5 | 1.1 by 5 | 6.8 |
| 113 | 0.4.0 | | | **Compute an item's S from the code it names, where it names any, and show it beside the finished items `score-outcomes` lists.** | 3 | 4 | 6 | 1.1 by 5, 1.4 by 5 | 6.8 |
| 121 | 0.4.0 | | | **Check that `bin/round_titles.py`'s `PASSES` lists exactly the passes of larv-review §3's table.** | 1 | 3 | 2 | 1.1 by 5 | 6.8 |
| 122 | 0.4.0 | | | **Assemble the judge's prompt in `bin/loop_claude/cycle/judge.py`, importing only wording the worker also sees, and call the worker "the worker" in it.** | 3 | 4 | 4 | 1.1 by 5, 1.4 by 5 | 6.8 |
| 4 | 0.5.0 | | | **Name each chunk's archive folder by the chunk's start time, so `<key>/` lists chunks in the order they ran.** | 3 | 2 | 3 | 1.1 by 3 | 5.5 |
| 12 | 0.5.0 | | | **Run the test suite in CI on every PR, or record in `docs/decisions.md` why the local run is enough.** | 2 | 4 | 3 | 1.1 by 6 | 5.5 |
| 105 | 0.5.0 | | decision | **Compare review rounds per pass and reading cost for the five chunks before and after the `worker:` decision moved the principles out of the worker's context, and file an item if either rose.** | 1 | 2 | 3 | 1.4 by 6 | 5.5 |
| 135 | 0.5.0 | | | **Fingerprint only the modules loop-claude imports, so editing another tool's code no longer reloads a running loop.** | 3 | 3 | 4 | 1.1 by 4 | 5.5 |
| 156 | 0.5.0 | | decision | **Drop `rank_todo.py`'s ranking of an old-shape table once no project under `~/code` holds one.** | 2 | 3 | 2 | 1.1 by 4 | 5.5 |
| 68 | 0.5.0 | | principle | **Test `workflow-bench/compare.py`'s blinding and `bench.py`'s run plan.** | 2 | 4 | 4 | 1.1 by 5 | 4.6 |
| 59 | 0.5.0 | | | **Bound a run of unfinished verdicts on one chunk, so a worker and judge that disagree cannot cycle unattended without end.** | 3 | 3 | 5 | 1.4 by 6 | 4.1 |
| 123 | 0.5.0 | | | **Have `bin/comprehension_history.py` parse each history entry once into seats, votes and places, so `latest_places`, `worse_places` and `worse_votes` neither depend on table order nor split a label to find a seat.** | 3 | 5 | 5 | 1.1 by 5, 1.4 by 5 | 4.1 |
| 132 | 0.5.0 | | | **Lift each dimension still under the comprehension floor after release 0.2.0's scoring run, at the places that run names.** | 3 | 5 | 5 | 1.1 by 5, 1.4 by 5 | 4.1 |
| 155 | 0.5.0 | | | **Lower the reading cost of the files a chunk reads most before its first commit.** | 2 | 3 | 4 | 1.4 by 6 | 4.1 |
| 124 | 0.5.0 | | decision | **Group the shared modules at `bin/`'s root into packages named by purpose, so the root holds only launchers and stand-alone tools.** | 3 | 6 | 3 | 1.1 by 9, 1.4 by 9 | 4.1 |
| 1 | 0.6.0 | | question | **Say in README → Install how the transcript archive is backed up, once bender-infra backs up ai-sandbox's data volume.** | 3 | 5 | 2 | | 0.0 |
| 5 | 0.6.0 | | decision | **Key a project on its checkout's full path, and refuse to start while another session uses that key.** | 5 | 5 | 7 | 1.1 by 6 | 3.3 |
| 77 | 0.6.0 | | | **Teach `bin/metric-bench` Go and TypeScript, rerun it once two projects' comprehension histories name places, and bring into review only a metric that tracks the panel.** | 3 | 6 | 6 | 1.1 by 7, 1.4 by 7 | 3.2 |
| 6 | 0.6.0 | | decision | **Move what `loop-claude` writes out of `~/.claude.json`, and ask the human about each write that must stay.** | 5 | 5 | 6 | 1.1 by 5 | 2.7 |
| 51 | 0.6.0 | | principle | **Hand loop-claude's terminal output to an injected effect, so no storage or retry function prints.** | 3 | 5 | 5 | 1.1 by 5 | 2.7 |
| 151 | 0.6.0 | | decision | **Keep 10 to 20 settled decisions per project as its goals' test set, and rerun them after a goal changes.** | 3 | 5 | 5 | 1.2 by 7 | 2.6 |
| 52 | 0.6.0 | | principle | **Record why this checkout lives in Claude Code's own directory, or move it out and link it in.** | 6 | 6 | 4 | 1.1 by 4 | 1.2 |
## Details
@@ -82,6 +82,63 @@
`ENOUGH` in `bin/score_outcomes/outcomes.py` matches the 10 figures the reading-cost trend waits for. No published standard sets either, and `CLAUDE.md` → "Web research" asks the human for a threshold a consumer relies on.
### 140. Split `docs/decisions.md` → `routing: one ladder` into one entry per decision, so a reader finds the rules for the ladder, the worker's rung, the routing figures and delegation by their titles.
The entry holds 7 paragraphs under one title of about 70 words chaining three decisions: the ladder and bands, the worker below the project's rung with its reviewers, the routing figures and calls, delegation, and when the worker changes rung. A reader after one rule reads the whole entry. The prose pass of the chunk that split out the judge's rung asked for the rest. Once split, the delegation entry's title names the floor that `todo.md: only a session or subagent on the project's rung or above scores or reshapes the plan…` puts under a subagent that files or scores items.
### 67. Point stability-reviewer at the technical principles in place of its own restatement of them.
DRY, in `principles/technical-principles.md`: `agents/stability-reviewer.md` restates principles in its own words, and the copy has drifted: it asks for "properly randomised UUIDs" where the principle requires UUID v7.
### 146. Score R as L + E + I − 2 from anchored levels, at least 7 at I5 and 5 at I4, with L, E and I in their own columns and `rank_todo.py` computing R, drop C, and route a task's model and effort on its R alone.
Each level is about ×10 of a real quantity; `docs/decisions.md` → `todo.md: …` says why. The anchors, for `rules/todo-items.md`:
| Level | L — chance the change holds a defect | E — chance the defect escapes | I — damage, worst of availability, data, secrets, reach and reversibility |
|---|---|---|---|
| 1 | S 1–2, and nothing new | a test or the gate fails on it | one person's tooling or something cosmetic; a revert fixes it in minutes |
| 2 | S 3–5, in one component, a pattern the repo already has | only review or a manual check sees it | one feature or service degraded; a revert within the hour |
| 3 | one of: S 6 or more, several components, a new tool or pattern, live config or state | nothing checks it before users hit it | a service down, or every consumer affected; hours to restore |
| 4 | two or more of level 3's, or a rewrite of code with known defects | — | everything down, a backup restore, or a secret or PII exposed |
| 5 | — | — | irreversible: data lost for good, or a one-way door |
`rank_todo.py` ranks a table with C and without L, E and I until its project reshapes it, and says so, as it does for the old shape since PR #283, so no project's commits stop the day this lands. Reshaping is then that project's only work until done, as `CLAUDE.md` → "Working loop" says.
C goes in the same chunk, so no reshaped table meets bands built for R + C: R alone runs 1 to 10, and the R + C bands would step most tasks well down the ladder. The bands (`docs/decisions.md` → `routing: one ladder…`) become bands over R, spread across the R values the rescored plans hold. The bands are the human's call, so the chunk proposes them and waits, and puts one option beside them: the worker routed on L and its reviewers on E + I, since impact makes a task riskier, not harder. `CLAUDE.md`'s delegation rule, `subagent-rung` and its `--plan`, loop-claude's chunk routing and prompts, the judge's rung, escalation trigger 1 and `score-outcomes` drop C. The routing decision's "S never moves a rung" no longer holds, since size sets L, so the chunk revises it. Once it merges, append a `Reset` to `routing-calls-raised.md`, as `AGENTS.md` says of a band change, so the routing figures count only chunks routed on R.
### 152. Anchor each score level in `rules/todo-items.md` with words, and each project's with two or three items it shipped, and have a scorer compare a new item against them.
LLM ratings drift to the middle of an unanchored scale, and worked examples cut estimation error. A project's anchors are shipped items, cited by ID and commit, chosen from what `score-outcomes` measured for them.
### 50. Grant the bench's runs only the tools a review needs, in place of skipping every permission.
"Always grant explicit access to data", in `principles/technical-principles.md`: `workflow-bench/bench.py` starts claude with `--dangerously-skip-permissions`, and no rule confines it to a sandbox, so a run on the host reaches the host's credentials and other checkouts.
### 65. Archive each judge reply the loop rejects.
A defect: goal 3, keeping unattended runs' transcripts backed up, does not hold while a rejected judge reply reaches only stderr. Raised by the principles audit of PR #215: the judge runs with `--no-session-persistence`, so the archive cannot answer the 3am question of why the loop rejected a verdict.
### 84. Check, at each release's scoring run, that seats in a paired ruling agree more often than two scoring runs on near-identical code do, as the paired-ruling decision assumes.
Raised by the principles audit of PR #228: the `review: the comprehension check records …` entry in `docs/decisions.md` is valid only while that holds, and `docs/comprehension-history.md` keeps every vote, yet nothing reads the agreement rate. Scoring runs on near-identical code are rare now, so compare with those already in the history, such as PR #224's and PR #225's, or make a second run for the check.
### 153. Run the judge on the rung the highest R + C in its chunk picks, an item the worker takes in included.
The judge reports the R + C of an item the worker takes in, and `routing.stepped` lifts the worker to the rung it picks, but `judge_rung` reads only the routed item's R + C, which the archived route line holds. So when the worker takes in a harder item, the judge stays on a lower rung than the work, where a missed interrupt lets the worker settle a decision that is lilleman's (goal 1.2). The route must carry the highest R + C taken in, archived so a rerun keeps it.
### 73. Have each reviewer write its full report to a file and return the worker only its findings' one-line summaries.
2026-10-08, lilleman: rescored from A10/G8 to A4/G4 on the measurement below, so other items that lower token use go first.
Measured over the 68 transcripts in loop-claude's archive on 2026-10-08, matching each Agent call to its tool result and to any `<task-notification>` naming its tool-use id. No report reached the worker twice: 813 arrived once as the hand-back, and 363 once as a notification after Claude Code ran the agent in the background. Reports took a median 19k tokens per chunk, 8% of the worker's peak context (24% at most). Lines carrying a severity mark are 9–24% of a reviewer's report, and the worker would still open the file for each 🔴 or 🟡 it fixes. A third of the report volume is the comprehension seats', whose votes and places the worker copies into the history. File reads fill more of the worker's context: `sed -n` output is 30% of its Bash results and `grep` output 11%, which `reading-cost` targets.
The reports would live in the session's scratchpad, such as `<scratchpad>/larv-review/prose-round-1.md`, so no project's checkout sees them. A later round gets the path of that pass's earlier report in place of the findings pasted in. How a report reaches the file:
- **Write access.** Grant each reviewer Write. No reviewer under `agents/` holds it today, an agent's tool list cannot confine Write to one folder, and `~/.claude/CLAUDE.md` → "Working with the codebase" denies edits to a reviewer that only reads. Check whether a PreToolUse hook can tell a subagent's call from the worker's and deny a Write outside the scratchpad.
- **A wrapper.** A general-purpose subagent launches the reviewer, saves the report and returns only the summary lines. Claude Code's documentation says a subagent cannot start another; drop this route unless a test shows one can.
This risks goal 1.1: a worker fixing from a one-line summary can miss context it needs.
### 80. State in README → Install that every machine taking this checkout must be a sandbox, as `settings.json`'s auto mode assumes.
A defect: goal 2, keeping the setup shareable between computers, does not hold while a new machine is never told it must be a sandbox. Raised by the principles audit of PR #224. The `settings.json` decision is valid while every machine taking this checkout is a sandbox, but Install asks for a sandbox only for `loop-claude`.
@@ -146,10 +203,6 @@ Junior A, Mid A and the maintainability senior of PR #257's scoring run named `b
Junior A and Mid A of PR #257's scoring run named `bin/shell_command.py:31` `segments`: a hand-rolled tokenizer with dense regexes and a second lexing path for unbalanced quotes, which the rule hook and `reading-cost` both depend on. Its comments state each quirk but not its limits. Make `segments` easier to read.
### 140. Split `docs/decisions.md` → `routing: one ladder` into one entry per decision, so a reader finds the rules for the ladder, the worker's rung, the routing figures and delegation by their titles.
The entry holds 7 paragraphs under one title of about 70 words chaining three decisions: the ladder and bands, the worker below the project's rung with its reviewers, the routing figures and calls, delegation, and when the worker changes rung. A reader after one rule reads the whole entry. The prose pass of the chunk that split out the judge's rung asked for the rest. Once split, the delegation entry's title names the floor that `todo.md: only a session or subagent on the project's rung or above scores or reshapes the plan…` puts under a subagent that files or scores items.
### 154. Have `bin/shell_command.py`'s `segments` keep a command's words together across a redirect, so `git > out commit` reads as one git commit.
`segments('echo hi > f arg')` returns `arg` as a command of its own, and `segments('git > /tmp/x commit -m hi')` returns `git` and `commit -m hi` apart, though the shell runs one `git commit`. rule-hook's commit rules read the commands `segments` finds, so such a commit escapes `commit-one-line`, `no-ai-attribution`, `round-title-counts` and `todo-rank-check`. The comprehension seats' scoring run in PR #280 found it.
@@ -212,10 +265,6 @@ Found by the principles audit of PR #242. `bin/loop_claude/test_black_box.py:113
KISS, in `principles/technical-principles.md`, which keeps a limit only while readers find the flagged code hard. `agents/systems-architect.md` makes cyclomatic 7 and 24 lines gate rules from the first commit, a 🔴 unless a reason is stated.
### 67. Point stability-reviewer at the technical principles in place of its own restatement of them.
DRY, in `principles/technical-principles.md`: `agents/stability-reviewer.md` restates principles in its own words, and the copy has drifted: it asks for "properly randomised UUIDs" where the principle requires UUID v7.
### 101. Have `larv_review/record.py`'s `read_seats` take the time, the commit check and the history as inputs, so it parses and decides without effects.
Found by the principles audit of PR #242, against "Functional core, imperative shell" (`principles/technical-principles.md`). `bin/larv_review/record.py:67` reads `time.time()`, then checks git (`:70`, `commit_exists`) and the history file (`:75`) inside the function that parses the seats file and decides. A test cannot fix the heading's time, and every test of `read_seats` touches the real checkout.
@@ -246,14 +295,6 @@ The repo has no CI, so larv-review's gate is the local run, and larv-review's Ga
When CI runs a rule check, such as one of `bin/rule-hook`'s, let `<!-- rule-override <rule>: <reason> -->` in the PR body override it.
### 50. Grant the bench's runs only the tools a review needs, in place of skipping every permission.
"Always grant explicit access to data", in `principles/technical-principles.md`: `workflow-bench/bench.py` starts claude with `--dangerously-skip-permissions`, and no rule confines it to a sandbox, so a run on the host reaches the host's credentials and other checkouts.
### 65. Archive each judge reply the loop rejects.
A defect: goal 3, keeping unattended runs' transcripts backed up, does not hold while a rejected judge reply reaches only stderr. Raised by the principles audit of PR #215: the judge runs with `--no-session-persistence`, so the archive cannot answer the 3am question of why the loop rejected a verdict.
### 105. Compare review rounds per pass and reading cost for the five chunks before and after the `worker:` decision moved the principles out of the worker's context, and file an item if either rose.
The `worker:` decision is valid while review rounds per pass do not rise. Count the rounds from each PR's `Answer <pass> round <n>:` commits, and take reading cost from `docs/reading-cost.md`. Run it once five chunks have merged after the one that recorded the decision.
@@ -270,14 +311,6 @@ While `rank_todo.py` ranks two shapes, projects follow two sets of scoring rules
"Most, if not all, code should have automated tests", in `principles/technical-principles.md`: `compare.blind()` keeps the comparison blind, and a version name it misses invalidates the bench silently.
### 84. Check, at each release's scoring run, that seats in a paired ruling agree more often than two scoring runs on near-identical code do, as the paired-ruling decision assumes.
Raised by the principles audit of PR #228: the `review: the comprehension check records …` entry in `docs/decisions.md` is valid only while that holds, and `docs/comprehension-history.md` keeps every vote, yet nothing reads the agreement rate. Scoring runs on near-identical code are rare now, so compare with those already in the history, such as PR #224's and PR #225's, or make a second run for the check.
### 153. Run the judge on the rung the highest R + C in its chunk picks, an item the worker takes in included.
The judge reports the R + C of an item the worker takes in, and `routing.stepped` lifts the worker to the rung it picks, but `judge_rung` reads only the routed item's R + C, which the archived route line holds. So when the worker takes in a harder item, the judge stays on a lower rung than the work, where a missed interrupt lets the worker settle a decision that is lilleman's (goal 1.2). The route must carry the highest R + C taken in, archived so a rerun keeps it.
### 59. Bound a run of unfinished verdicts on one chunk, so a worker and judge that disagree cannot cycle unattended without end.
Raised by the product-owner review of PR #211. The resumed worker gets the judge's quote of the work remaining, but a worker that still thinks the chunk done can answer each cycle the same way, and each cycle costs a worker and a judge run (goal 1.4).
@@ -308,26 +341,6 @@ It lives under `~/.claude`, which `.gitignore` keeps from the remote, so a lost
The README line covers the bench's transcripts in `workflow-bench/logs/` too (goal 3).
### 146. Score R as L + E + I − 2 from anchored levels, at least 7 at I5 and 5 at I4, with L, E and I in their own columns and `rank_todo.py` computing R, drop C, and route a task's model and effort on its R alone.
Each level is about ×10 of a real quantity; `docs/decisions.md` → `todo.md: …` says why. The anchors, for `rules/todo-items.md`:
| Level | L — chance the change holds a defect | E — chance the defect escapes | I — damage, worst of availability, data, secrets, reach and reversibility |
|---|---|---|---|
| 1 | S 1–2, and nothing new | a test or the gate fails on it | one person's tooling or something cosmetic; a revert fixes it in minutes |
| 2 | S 3–5, in one component, a pattern the repo already has | only review or a manual check sees it | one feature or service degraded; a revert within the hour |
| 3 | one of: S 6 or more, several components, a new tool or pattern, live config or state | nothing checks it before users hit it | a service down, or every consumer affected; hours to restore |
| 4 | two or more of level 3's, or a rewrite of code with known defects | — | everything down, a backup restore, or a secret or PII exposed |
| 5 | — | — | irreversible: data lost for good, or a one-way door |
`rank_todo.py` ranks a table with C and without L, E and I until its project reshapes it, and says so, as it does for the old shape since PR #283, so no project's commits stop the day this lands. Reshaping is then that project's only work until done, as `CLAUDE.md` → "Working loop" says.
C goes in the same chunk, so no reshaped table meets bands built for R + C: R alone runs 1 to 10, and the R + C bands would step most tasks well down the ladder. The bands (`docs/decisions.md` → `routing: one ladder…`) become bands over R, spread across the R values the rescored plans hold. The bands are the human's call, so the chunk proposes them and waits, and puts one option beside them: the worker routed on L and its reviewers on E + I, since impact makes a task riskier, not harder. `CLAUDE.md`'s delegation rule, `subagent-rung` and its `--plan`, loop-claude's chunk routing and prompts, the judge's rung, escalation trigger 1 and `score-outcomes` drop C. The routing decision's "S never moves a rung" no longer holds, since size sets L, so the chunk revises it. Once it merges, append a `Reset` to `routing-calls-raised.md`, as `AGENTS.md` says of a band change, so the routing figures count only chunks routed on R.
### 152. Anchor each score level in `rules/todo-items.md` with words, and each project's with two or three items it shipped, and have a scorer compare a new item against them.
LLM ratings drift to the middle of an unanchored scale, and worked examples cut estimation error. A project's anchors are shipped items, cited by ID and commit, chosen from what `score-outcomes` measured for them.
### 5. Key a project on its checkout's full path, and refuse to start while another session uses that key.
The key names the project's session, archive folder and models entry. `project_key` drops the leading `/`, so `~/code/a` and `/a` share all three, and one loop can resume or delete the other's session.
@@ -376,19 +389,6 @@ Claude Code owns that file and writes it. `trust` reads it, changes it and write
Each case is a question the project settled and the answer it chose. A fresh-context agent given only the README's premises answers each case; a case it answers differently says the goals no longer decide it. Written guidance can fail to change behaviour, which is why the goals get a test.
### 73. Have each reviewer write its full report to a file and return the worker only its findings' one-line summaries.
2026-10-08, lilleman: rescored from A10/G8 to A4/G4 on the measurement below, so other items that lower token use go first.
Measured over the 68 transcripts in loop-claude's archive on 2026-10-08, matching each Agent call to its tool result and to any `<task-notification>` naming its tool-use id. No report reached the worker twice: 813 arrived once as the hand-back, and 363 once as a notification after Claude Code ran the agent in the background. Reports took a median 19k tokens per chunk, 8% of the worker's peak context (24% at most). Lines carrying a severity mark are 9–24% of a reviewer's report, and the worker would still open the file for each 🔴 or 🟡 it fixes. A third of the report volume is the comprehension seats', whose votes and places the worker copies into the history. File reads fill more of the worker's context: `sed -n` output is 30% of its Bash results and `grep` output 11%, which `reading-cost` targets.
The reports would live in the session's scratchpad, such as `<scratchpad>/larv-review/prose-round-1.md`, so no project's checkout sees them. A later round gets the path of that pass's earlier report in place of the findings pasted in. How a report reaches the file:
- **Write access.** Grant each reviewer Write. No reviewer under `agents/` holds it today, an agent's tool list cannot confine Write to one folder, and `~/.claude/CLAUDE.md` → "Working with the codebase" denies edits to a reviewer that only reads. Check whether a PreToolUse hook can tell a subagent's call from the worker's and deny a Write outside the scratchpad.
- **A wrapper.** A general-purpose subagent launches the reviewer, saves the report and returns only the summary lines. Claude Code's documentation says a subagent cannot start another; drop this route unless a test shows one can.
This risks goal 1.1: a worker fixing from a one-line summary can miss context it needs.
### 52. Record why this checkout lives in Claude Code's own directory, or move it out and link it in.
"One owner per value" and "Prefer the vendor-neutral option", in `principles/technical-principles.md`: git, Claude Code and loop-claude all write `~/.claude`, kept apart only by the `.gitignore` allowlist, and the repo's records live in a vendor directory.