lilleman
|
1d411be20d
|
Refuse in rule-hook a Bash or Monitor wait that has no end
CI / tests (pull_request) Failing after 3m3s
|
2026-11-18 19:07:42 +01:00 |
|
lilleman
|
8210754780
|
Answer prose round 1: 0 🔴, 8 🟡 (0 misread); reword the decision, the rule texts and item 341
|
2026-11-18 15:54:24 +01:00 |
|
lilleman
|
a7b2fd0ae1
|
Work in progress: Bash output ceiling and refusing Reads of session transcripts
|
2026-11-18 08:28:38 +01:00 |
|
lilleman
|
934a398be0
|
Have larv-review round-title print a round commit's counts from its report, and rule-hook refuse a title that differs
|
2026-11-17 07:47:45 +01:00 |
|
lilleman
|
31da587415
|
Answer stability round 1: 0 🔴, 14 🟡 (0 misread); tests that kill the surviving mutants, with a versioned MemoryStore and appended_to, plan/named_paths.py, TakenIn's docstring and the run-heavy rule naming LAUNCHERS; three places left to item 300
CI / tests (pull_request) Successful in 3m10s
|
2026-11-16 16:53:36 +01:00 |
|
lilleman
|
ba21a1efcd
|
Answer principles round 1: 2 🔴, 9 🟡 (0 misread); explicit record versions in place of the Confluence store's hidden state, the routing ruling filed as item 393 for 0.12.0, rule-hook catching heavy runs through uv, poetry, sudo and doas, one unreleased heading, one owner of the no-item message, typed Jira issues, a turn that gives its failure as a value, outcome tests for the Atlassian client, an Emptied reason type, an injected clock and git in touch-predictions, and paired record names
|
2026-11-16 16:12:36 +01:00 |
|
lilleman
|
f58250d0ce
|
Answer architecture round 1: 4 🔴, 5 🟡 (0 misread); one-way plan packages with a plan core and an Atlassian client, one parse and print of an item's cells, a pure chunk-start decision, a settings table, one plan read and one standing per pick, one config load in chunk-pr, notifier backends in notifier, and the bin regroup filed as item 392
|
2026-11-16 14:47:46 +01:00 |
|
lilleman
|
bedd304479
|
Merge remote-tracking branch 'origin/main' into patch-0.10.1
# Conflicts:
# CHANGELOG.md
# README.md
# bin/loop_claude/loop/chunk_start.py
# bin/loop_claude/loop/close.py
# bin/loop_claude/loop/main.py
# bin/loop_claude/routing.py
# bin/loop_claude/test_loop_claude.py
# bin/project_config/main.py
# bin/project_config/test_project_config.py
# docs/decisions.md
|
2026-11-15 19:54:42 +01:00 |
|
lilleman
|
c5acadf8d3
|
Keep the shipped routing ladder beside the module that reads it, so a copy of bin/ carries it
|
2026-11-15 18:59:23 +01:00 |
|
lilleman
|
f98e8a84dd
|
Answer stability round 1: 1 🔴, 6 🟡 (0 misread); read every title through both templates, hold a title to its template only on an item's branch, and fail on a config fault
|
2026-11-14 01:47:41 +01:00 |
|
lilleman
|
ce87a519f0
|
Read the prose language and the title and branch templates from the project's config
|
2026-11-14 01:14:08 +01:00 |
|
lilleman
|
82f4bbd91b
|
Read a project's language, title templates and branch template from its config
|
2026-11-14 01:08:48 +01:00 |
|
lilleman
|
3213971f98
|
Answer stability round 1: 1 🔴, 3 🟡 (0 misread); say a refused search searches a folder holding the file and what to do, move the folder search into its own module, read -g only at the start of a word, lift a skip on a case-blind glob, and fix the prose findings of the product-owner and prose rounds in this answer
|
2026-11-13 22:34:18 +01:00 |
|
lilleman
|
153a85c78f
|
Refuse a Bash command that recursively searches a folder holding an env-file
|
2026-11-13 21:36:17 +01:00 |
|
lilleman
|
0cc242ac1b
|
Answer stability round 1: 0 🔴, 4 🟡 (0 misread); say a command names the file rather than reads it, name the way on, refuse redirect targets and a quoted $HOME, check each file once, and drop the README caveat the rule outlived
|
2026-11-13 21:25:15 +01:00 |
|
lilleman
|
60a1f5c385
|
Answer security round 1: 0 🔴, 1 🟡 (0 misread); refuse a Grep with a glob over a folder holding an ignored env-file, and file item 370 for recursive Bash searches
CI / tests (pull_request) Successful in 2m36s
|
2026-11-13 21:12:38 +01:00 |
|
lilleman
|
333ebe58b8
|
Refuse a Read, Grep or Bash call that reads ~/.claude/.env or a config's env-file
CI / tests (pull_request) Failing after 2m29s
|
2026-11-13 20:56:17 +01:00 |
|
lilleman
|
e1ca619e96
|
Name the four kinds of breach, the two level rules and the round title's spelling, and table the step-up rules
|
2026-11-13 02:45:35 +01:00 |
|
lilleman
|
5c7969e7c8
|
Answer stability round 1: 0 🔴, 1 🟡 (0 misread); say the checkout's commits, not git, hold no version of a file outside it, and find a symlinked config folder
|
2026-11-13 02:10:33 +01:00 |
|
lilleman
|
cfb5e50904
|
Commit a chunk's changes to the config's folder in its own repo
CI / tests (pull_request) Failing after 2m33s
|
2026-11-13 01:58:50 +01:00 |
|
lilleman
|
c5dc8b5755
|
Answer stability round 1: 1 🔴, 6 🟡 (0 misread); say what a plan outside the checkout leaves undone in scoring-run-owed, the release standing and rank_todo.py
|
2026-11-13 01:44:33 +01:00 |
|
lilleman
|
14e6756a53
|
Answer prose round 1: 0 🔴, 6 🟡 (0 misread); rewrite each <verb> nothing to the negation rule, tighten the new docstrings and comments
|
2026-11-12 20:33:32 +01:00 |
|
lilleman
|
7afc59a9d7
|
Merge remote-tracking branch 'origin/main' into item-293
# Conflicts:
# AGENTS.md
# agents/stability-reviewer.md
# docs/reading-cost.md
# todo.md
|
2026-11-11 19:24:50 +01:00 |
|
lilleman
|
a52ebb94bf
|
Run mutation tests on a chunk's changed lines before review with bin/mutate-changes, and hand stability-reviewer the survivors
CI / tests (pull_request) Successful in 2m14s
|
2026-11-11 18:58:31 +01:00 |
|
lilleman
|
47c10c4565
|
Answer prose round 1: 0 🔴, 5 🟡 (0 misread); reword the help paragraph, the changelog bullet and the decision, and restore the entry's old sentence
|
2026-11-11 14:19:48 +01:00 |
|
lilleman
|
3e289c95d9
|
Check the files a commit would commit against rule-hook's file rules, and sort .gitignore's allowlist block
CI / tests (pull_request) Failing after 2m2s
|
2026-11-11 14:07:34 +01:00 |
|
lilleman
|
797c5dd4fc
|
Answer product-owner round 1: 1 🔴, 2 🟡 (0 misread); check chunk-pr commands for an attribution; the default-branch refusal and the PR summary went in the stability and prose commits; tea's --draft and --ready are verified by its help and #448
|
2026-11-09 02:14:12 +01:00 |
|
lilleman
|
bd96cb7c03
|
Leave items a chunk drops by a 'Drop item' title out of score-outcomes, check this repo's branches with their own rank_todo.py, and test the replays' model, floor and checkout
|
2026-11-09 01:14:42 +01:00 |
|
lilleman
|
8d9a39de1c
|
Group bin's root modules into packages named by purpose, and make rank_todo.py a launcher of release_plan/ranking.py
|
2026-11-08 20:35:52 +01:00 |
|
lilleman
|
2321f4930d
|
Keep the rule for which commands reach run-heavy's docker beside that docker in capped_docker.py
|
2026-11-08 11:09:17 +01:00 |
|
lilleman
|
452a0c4271
|
Answer architecture round 1: 1 🔴, 12 🟡 (0 misread); route every verdict through next_route with one step-up writer, give a release one home in release.py and rename chunk_pick to release_standing, split loop/main.py, delete the unreachable lowered-pass code, render agent-followed Markdown from one list, refuse a round title naming no pass, and rename the clashing names
|
2026-11-04 22:57:30 +01:00 |
|
lilleman
|
2ee9df2e07
|
Answer stability round 1: 0 🔴, 4 🟡 (0 misread); refuse a security skill in a subagent not launched as the security pass, split skill_breaches out, tell a subagent by agent_id
|
2026-10-30 13:22:25 +01:00 |
|
lilleman
|
a212b4ef91
|
Refuse a security-review or claude-security Skill call from the worker's own context below the project's rung
|
2026-10-30 13:10:56 +01:00 |
|
lilleman
|
6bc1e9ac9c
|
Run the architecture and principles passes only in a release review before each cut
|
2026-10-29 21:03:25 +01:00 |
|
lilleman
|
0254e427f2
|
Give each review pass at most two rounds per chunk, with no third round and no ask for one
|
2026-10-28 17:37:09 +01:00 |
|
lilleman
|
dd90d109fd
|
Answer product-owner round 2: 0 🔴, 1 🟡 (0 misread); say that a Read goes through until the agent read the file with Read
|
2026-10-28 14:41:52 +01:00 |
|
lilleman
|
b1a90abff4
|
Answer stability round 1: 1 🔴, 8 🟡 (0 misread); read only a regular file, only as far as the shown lines, skip cut-short output, and test the surviving mutants
|
2026-10-28 14:14:53 +01:00 |
|
lilleman
|
2d3ec3a636
|
Block a read of file lines already in the agent's context, unless a compaction dropped them or the file changed
|
2026-10-28 12:43:46 +01:00 |
|
lilleman
|
6a2f29635e
|
Answer stability round 2: 1 🔴, 2 🟡 (0 misread); drop a --memory of 0 so the added limit stands, pin the 6m floor's edge, test env --unset, env --ignore-environment and doas, and say on Linux in the rule text
|
2026-10-24 11:42:53 +02:00 |
|
lilleman
|
055c2d2df0
|
Answer product-owner round 2: 0 🔴, 1 🟡 (0 misread); a size has at most one point, so -m 1.2.3 in a container's command is no size and crashes nothing
|
2026-10-24 11:39:12 +02:00 |
|
lilleman
|
c87e9f19d6
|
Answer stability round 1: 1 🔴, 5 🟡 (0 misread); the 🔴, the rule text and the PATH half of the exemption were answered in the principles round 1 commit; exempt no docker under sudo or doas, test unknown memory and the shim without a share, keep helped beside unlimited, drop a dead fallback
|
2026-10-24 11:37:08 +02:00 |
|
lilleman
|
2fa328a406
|
Answer principles round 1: 0 🔴, 4 🟡 (0 misread); exempt only a bare docker under run-heavy with no PATH of its own, drop the stale item citation, unit-test the slot share in MiB, place the limits by the subcommand's index; also a named Option, exit 126, clearer names and the launcher's import order
|
2026-10-24 11:31:10 +02:00 |
|
lilleman
|
52641079b7
|
Answer architecture round 1: 0 🔴, 3 🟡 (0 misread); keep the slot's share beside ROOM_SHARE in machine_load, move the rule's predicate into rule_hook, and have capped set both limits itself instead of leaning on docker taking the last option
|
2026-10-24 11:24:21 +02:00 |
|
lilleman
|
196abe18a9
|
Answer principles round 1: 0 🔴, 5 🟡 (0 misread); require a swap limit beside the memory limit, ask in CLAUDE.md for both where rule-hook can't see the container, remove items 225 and 228 with a changelog entry; the -m finding was answered in the product-owner round 1 commit
|
2026-10-24 00:42:39 +02:00 |
|
lilleman
|
2d219ff6c8
|
Answer product-owner round 1: 0 🔴, 1 🟡 (0 misread); read -m and --memory as a limit only with a size, so python -m pytest in the container is no limit, and read timeout as a launcher; also take the technical principles' nits: --help counts only beside docker's subcommand, sort STARTS_CONTAINER, print the example limit once, and name item 227 in the decision
|
2026-10-24 00:36:19 +02:00 |
|
lilleman
|
a292ec4cae
|
Answer architecture round 1: 0 🔴, 3 🟡 (0 misread); move docker's parsing to bin/ for item 227's shim, read launchers once in shell_command.launched, and give the memory limit its own decision, sized to what the work needs
|
2026-10-24 00:31:41 +02:00 |
|
lilleman
|
7b61fd5b4b
|
Sort rule_hook's imports by their first binding
|
2026-10-24 00:18:50 +02:00 |
|
lilleman
|
fbaa3075c4
|
Block a docker run or docker create that sets no memory limit, and read docker's TLS and buildx --builder options as taking a value
|
2026-10-24 00:15:44 +02:00 |
|
lilleman
|
6ea148bdf4
|
Answer principles round 1: 1 🔴, 0 🟡 (0 misread); sort the heavy_command import before review_folder
|
2026-10-23 14:45:45 +02:00 |
|
lilleman
|
578de9231f
|
Move the run-heavy rule's command parsing into its own module and pick docker's position without a nested conditional
|
2026-10-23 14:35:06 +02:00 |
|