Answer prose round 1: 0 🔴, 12 🟡 (0 misread); name the tmux and claude sessions apart, say parallel-chunks is a most, and put the fakes and keywords in order

This commit is contained in:
2026-11-16 12:46:08 +01:00
parent 986697ab52
commit 255d5f7bb1
7 changed files with 15 additions and 16 deletions
+3 -3
View File
@@ -71,7 +71,7 @@ git config filter.settings.clean bin/settings_filter.py
Then pick a model with `/model`.
Link `loop-claude` onto `PATH`: `ln -s ~/.claude/bin/loop-claude ~/.local/bin/loop-claude`. Every claude it starts, the interactive ones included, skips every permission prompt. It runs each chunk in its own window of a tmux session per project, so it needs tmux 3.2 or later.
Link `loop-claude` onto `PATH`: `ln -s ~/.claude/bin/loop-claude ~/.local/bin/loop-claude`. Every claude it starts, the interactive ones included, skips every permission prompt. loop-claude runs each chunk in its own window of a tmux session per project, so it needs tmux 3.2 or later.
To get loop-claude's questions and stops as DMs from a bot, write the bot's token, your username or member ID and, for Mattermost, the server into `~/.claude-projects/<key>.env`, in the keys `~/.claude/bin/notify --help` names, run `chmod 600` on it, and add `env-file = "~/.claude-projects/<key>.env"` and `notifier = "mattermost"` (or `"slack"`) to the project's config, as Install → "Set up a project" describes.
@@ -105,7 +105,7 @@ A project may also set `notifier` to `"mattermost"` or `"slack"`, the service th
A project's config may set four more things, each optional, as `project-config --help` describes. `language` names the language its owners write in, where that isn't English: the worker writes the project's prose in it, and the prose reviewer judges the prose in it. `commit-title`, `pull-request-title` and `branch` are templates for a project that titles and names these in another way than `CLAUDE.md` → "Git, commits, and PRs" says. `chunk-preflight` names the branch and the worktree by the `branch` template. On a branch that fits it, rule-hook refuses a commit, and `chunk-pr` a pull request, whose title doesn't fit its template; on any other branch, such as a release review's, a title is the summary alone. A tool that reads a commit title, such as a round's `Answer <pass> round <n>:`, reads the template's `{summary}`: it tries `commit-title`, then `pull-request-title`, since a squash merge titles the commit as the pull request, then takes the title whole. While a project's config holds a fault, rule-hook refuses every commit there, and the tools that read its titles fail naming the fault.
A legacy project with no container setup sets `tooling-on-host = true`, so the loop runs its language runtime, package manager, tests and task runner directly on the host. One whose code the comprehension panel scores low may lower `readability-floor` from 7.0: under it, the ranking weighs code health 4 times as much, and a scoring run asks for a floor item. To run more than one chunk at once, set `parallel-chunks` to how many; left out, it is 1.
A legacy project with no container setup sets `tooling-on-host = true`, so the loop runs its language runtime, package manager, tests and task runner directly on the host. One whose code the comprehension panel scores low may lower `readability-floor` from 7.0: under it, the ranking weighs code health 4 times as much, and a scoring run asks for a floor item. To let more than one chunk run at once, set `parallel-chunks` to the most that may; left out, it is 1. Beside the plan's next item, only items whose `Touches:` predictions overlap none of the running chunks' items start.
A project whose pull requests need a human's approval sets `wait-for-approval = true` in its config. `chunk-pr merge` then waits, once the checks have passed, until GitHub reports the pull request approved as its branch protection requires, and merges it. It refuses on Gitea, where a reviewer requested changes and where the protection requires no review. While it waits, other chunks merge. It stops where a reviewer's comment waits for an answer, on an approved pull request too; answer it and run it again. It gives up after 90 minutes without an approval; run it again once the pull request is approved. Left out, the setting is false.
@@ -174,7 +174,7 @@ Install puts only `loop-claude` on `PATH`; call the others by their path, such a
- `deploy-release [--cut] [<checkout>]` moves `~/.claude`, or the checkout named, to the newest release, as Update says; with `--cut`, run in the development clone, it first cuts the next release, as Release says.
- `judge-replay <model>` replays loop-claude's judge on archived verdicts of the projects run on the model, and prints what each rung a review climbs to misses, and the lowest rung below them that, like each candidate above it, returns them; `judge-replay --help` states the replay.
- `larv-review <command>` holds larv-review's mechanical rules: the chunk's review folder, the paired ruling's base and vote count, each pass's round count, the scoring run's trigger and setup; the comprehension history's format, its latest places and its count of how often seats agree; and each reviewer's prompt header. `larv-review --help` states them. `settings.json` → `hooks` runs `larv-review reviewer-context` as each subagent starts. A reviewer under `agents/` whose frontmatter has `omitClaudeMd: true` starts without `CLAUDE.md`, and the command hands it that file's "Shared goal", "Prose" and "Vocabulary" sections.
- `loop-claude <task>` runs a project's next chunk cycle after cycle; `loop-claude --help` states its contract.
- `loop-claude <task>` runs a project's next chunks, each in its own tmux window, cycle after cycle; `loop-claude --help` states its contract.
- `metric-bench [history]` benches cheap static metrics of a project's Python against the places its comprehension history names as hardest; `metric-bench --help` states the metrics and the scores.
- `mutate-changes [--base <ref>] [--prefix <shell text>] [--jobs <n>]` mutation-tests the lines a branch changes and prints each mutant its tests don't kill; `mutate-changes --help` says which tool it runs for each language.
- `notify [<message>]` sends a message, or else stdin, as a DM from the bot of the project's `notifier`, Mattermost or Slack, to the human; the project's `env-file` holds the bot's token, the recipient and, for Mattermost, the server, and `notify --help` names its keys. `loop-claude --help` says when loop-claude sends one.
+7 -8
View File
@@ -10,20 +10,19 @@ and quotes what it prints in the worker's prompt. It leaves the checkout as it i
todo.md as the default branch holds it, the plan loop-claude routes on. Where the project's config
puts todo.md or CHANGELOG.md outside the checkout, it reads that file as it stands. Such a plan
no longer names a release whose last item has shipped, so chunk-preflight doesn't see that release,
and prints that the release reviews and the cuts are the human's to call. With --item, the item is
the plan item of that ID, in place of the next release's first item that does not wait: loop-claude
names the item of each chunk it starts beside another. It prints:
and prints that the release reviews and the cuts are the human's to call. It prints:
- the last version cut: the newest v* tag on the default branch, or on a branch with no tag the
newest version CHANGELOG.md heads `## X.Y.Z - YYYY-MM-DD`; an undated heading doesn't count.
Then how many entries the default branch's CHANGELOG.md holds under ## Unreleased, or else how
many commits followed the cut: the tag, or the commit that added that heading;
- the next release and its first item that does not wait, the item loop-claude routes on, with
its L and E + I. An item waits while it is marked question, or while an item still in todo.md
- the next release and its first item that does not wait, the item loop-claude routes on, with its L
and E + I; with --item <ID>, the plan item of that ID instead, which loop-claude names in a window
named for an item. An item waits while it is marked question, or while an item still in todo.md
blocks it. Where the last cut is a version x.y.z, the next release is the earliest after x.y.z
that todo.md has named at that cut or since, less a release its ## Renamed section marks
renamed. Where nothing is cut, it is the earliest release todo.md has ever named, less those
renamed, so a release whose items are all done still reads as next.
that todo.md has named at that cut or since, less a release its ## Renamed section marks renamed.
Where nothing is cut, it is the earliest release todo.md has ever named, less those renamed, so a
release whose items are all done still reads as next.
Where the next release holds no item, it says so, naming it where todo.md named it since the cut:
its release review due where changes are unreleased (the ## Unreleased entries, or else the
commits since the cut) and no commit titled `Release review of <release>` followed the tag,
+1 -1
View File
@@ -126,7 +126,7 @@ class Main(unittest.TestCase):
result = reply(Rung(args[args.index("--model") + 1], args[args.index("--effort") + 1]), "said Verdict: interrupt" in prompt.rsplit("latest last:", 1)[-1])
return subprocess.CompletedProcess(args, 0, json.dumps(result), "")
effects = Effects(claude=claude, claude_on_terminal=unused, tmux=unused, tmux_on_terminal=unused, discard_typed=unused, dm=unused, execv=unused, git=unused, input=unused, local_time=unused, memory=unused, now=unused, on_terminal=unused, python=unused, sleep=unused, stream_claude=unused)
effects = Effects(claude=claude, claude_on_terminal=unused, discard_typed=unused, dm=unused, execv=unused, git=unused, input=unused, local_time=unused, memory=unused, now=unused, on_terminal=unused, python=unused, sleep=unused, stream_claude=unused, tmux=unused, tmux_on_terminal=unused)
out, err = io.StringIO(), io.StringIO()
with contextlib.redirect_stdout(out), contextlib.redirect_stderr(err):
code = main.main(argv, {}, self.home, effects)
+1 -1
View File
@@ -25,7 +25,7 @@ class LoopRun:
route: Route | None = None
# How many unfinished verdicts the chunk has had running; a finished chunk and a rerun each set it to 0. It stays apart from route and unfinished_evidence, though all three change together at each verdict: merging or renaming a field stops a running loop at its next reload.
unfinished_run: int = 0
# The chunk this loop runs in its tmux window, as chunk_windows names it: `item-372`, or `no-item`. None for the loop-claude the human started, which starts the windows; a loop started before chunks ran in windows reloads with None too, and so moves its chunk into one.
# The chunk this loop runs in its tmux window, as chunk_windows names it: `item-372`, or `no-item`. None for the loop-claude the human started, which starts the windows; a loop started before chunks ran in windows reloads with None too, and so starts the project's windows; its open claude session, named for its slot, is left in the session picker (docs/decisions.md → `loop-claude: each chunk runs in its own tmux window, …`).
chunk: str | None = None
+1 -1
View File
@@ -91,7 +91,7 @@ aims for. While the panel's overall score is under it, code health weighs 4 time
ranking, and a scoring run with a mean under it asks for a floor item. A legacy project, whose code
scores low, may lower it. Left out, it is 7.0.
`parallel-chunks` is an optional setting, a whole number of 1 or more: how many chunks of the project
`parallel-chunks` is an optional setting, a whole number of 1 or more: the most chunks of the project
loop-claude runs at once, each in its own tmux window. Left out, it is 1.
`project-config tooling-on-host`, `project-config readability-floor` and `project-config parallel-chunks`
+1 -1
View File
@@ -35,7 +35,7 @@ TOOLING_ON_HOST = "tooling-on-host"
READABILITY_FLOOR = "readability-floor"
# The floor where the config leaves it out, which every dimension and the overall should reach; docs/decisions.md → "comprehension floor" says it doesn't block anything.
DEFAULT_FLOOR = decimal.Decimal("7.0")
# How many chunks of the project loop-claude runs at once, each in its own tmux window; 1 where left out. Three readers seated as the README's audience picked the spelling 2-1 over concurrent-chunks.
# The most chunks of the project loop-claude runs at once, each in its own tmux window; 1 where left out. Three readers seated as the README's audience picked the spelling 2-1 over concurrent-chunks.
PARALLEL_CHUNKS = "parallel-chunks"
# The optional text settings. `language` is the language the project's prose is written in, by its English name. The others are templates, as `templates` says.
LANGUAGE = "language"
+1 -1
View File
@@ -267,7 +267,7 @@ class Main(unittest.TestCase):
result = result if "api_error_status" in result else {"type": "result", "structured_output": {"found": [n + 1 for n in result["found"]]}}
return subprocess.CompletedProcess(args, 0, json.dumps(result), "")
effects = Effects(claude=claude, claude_on_terminal=unused, tmux=unused, tmux_on_terminal=unused, discard_typed=unused, dm=unused, execv=unused, git=system_effects().git, input=unused, local_time=unused, memory=unused, now=unused, on_terminal=unused, python=unused, sleep=unused, stream_claude=unused)
effects = Effects(claude=claude, claude_on_terminal=unused, discard_typed=unused, dm=unused, execv=unused, git=system_effects().git, input=unused, local_time=unused, memory=unused, now=unused, on_terminal=unused, python=unused, sleep=unused, stream_claude=unused, tmux=unused, tmux_on_terminal=unused)
out, err = io.StringIO(), io.StringIO()
with contextlib.redirect_stdout(out), contextlib.redirect_stderr(err):
code = main.main(argv, {}, self.home, effects)