Commit Graph

  • 38f4274c85 Answer stability round 2: 0 🔴, 1 🟡 (0 misread); keep a chunk open, and read its gate again, only where the judge ruled it finished Lilleman 2026-11-19 18:32:13 +01:00
  • dae767b085 Test that the worker keeps its --finish sentence where a person merges, and how long an unread gate waits Lilleman 2026-11-19 18:22:42 +01:00
  • 7e141eaa1e Answer prose round 1: 0 🔴, 3 🟡 (0 misread); reword the approval docstring and tell the worker it awaited a person's merge Lilleman 2026-11-19 18:07:54 +01:00
  • 22654a5e05 Answer product-owner round 1: 0 🔴, 3 🟡 (1 misread); say which plans finish at once, keep a repo with its own merge method from merging, and file item 457 Lilleman 2026-11-19 18:07:54 +01:00
  • 24bd382d48 Answer stability round 1: 0 🔴, 2 🟡 (0 misread); keep a finished chunk open until a person's merge is read and its worktree removed, and assert facts in place of sentences Lilleman 2026-11-19 18:07:54 +01:00
  • 2e0ceef6ae Sort the chunk-pr test's imports Lilleman 2026-11-19 17:53:57 +01:00
  • f49d176103 Test that a parked chunk says once what it awaits, and stops where its new gate cannot be recorded Lilleman 2026-11-19 17:44:37 +01:00
  • e6f7d50943 Finish item 448 in the plan Lilleman 2026-11-19 17:41:51 +01:00
  • 50dd61c039 Merge origin/main into item-448, passing the label table to the chunk's next-step command Lilleman 2026-11-19 17:41:39 +01:00
  • 402b0690c8 Answer stability round 2: 0 🔴, 0 🟡 (0 misread); no finding Lilleman 2026-11-19 17:37:48 +01:00
  • 5013fbdf1f Answer product-owner round 2: 0 🔴, 0 🟡 (0 misread); no finding Lilleman 2026-11-19 17:22:38 +01:00
  • f8cabf78bf Test that the prompt and the refusal name the team's labels Lilleman 2026-11-19 17:21:20 +01:00
  • 846d9cd197 Finish item 455 in the plan Lilleman 2026-11-19 17:11:23 +01:00
  • 62b4825cea Answer stability round 2: 1 🔴, 0 🟡 (0 misread); the fake's blocker check types, and the imports in order Lilleman 2026-11-19 17:06:29 +01:00
  • b7d7a1f1ea Kill the surviving mutant in the e2e's follow-up check Lilleman 2026-11-19 16:58:43 +01:00
  • b383a89c5b Answer security round 1: 0 🔴, 0 🟡 (0 misread); /security-review found nothing in the merge setting or the parked chunk's gate Lilleman 2026-11-19 16:57:04 +01:00
  • b832c1e669 Answer stability round 1: 0 🔴, 4 🟡 (0 misread); refuse an empty blocker or parent, scope the fake's blocker check to the epic, check the link's direction in the e2e, assert the next-step command only Lilleman 2026-11-19 16:46:27 +01:00
  • 90352d74c4 Kill the surviving mutants of the blocker check and the worker's next-step sentence Lilleman 2026-11-19 16:40:45 +01:00
  • ff3961ca77 File item 456, so the child-shell stop test stops failing under load Lilleman 2026-11-19 16:40:35 +01:00
  • 84a0f1d1be Test who merges at each default, a failed ready mark, the worker's sentence and an unreadable config Lilleman 2026-11-19 16:29:31 +01:00
  • 26cadbb79e Answer prose round 1: 0 🔴, 3 🟡 (0 misread); give the label table its own README bullet and say who closes an item a person split Lilleman 2026-11-19 16:28:14 +01:00
  • adae921cd6 Answer product-owner round 1: 0 🔴, 3 🟡 (0 misread); count picks, not labels, and name the team's labels where a label is asked for Lilleman 2026-11-19 16:28:14 +01:00
  • bfe32bc1fa Answer stability round 1: 0 🔴, 1 🟡 (0 misread); leave an item a person split to be closed by its team, and refuse a setting below the label table Lilleman 2026-11-19 16:28:14 +01:00
  • 144986e671 File item 456 into 0.13.0, so loop-claude waits for a person's merge instead of restarting the chunk Lilleman 2026-11-19 16:22:41 +01:00
  • 68ed7ac95a Answer product-owner round 2: 0 🔴, 1 🟡 (0 misread); tell the worker to file only the next step, so the steps chain Lilleman 2026-11-19 16:16:53 +01:00
  • dae6f586ab Answer prose round 1: 0 🔴, 6 🟡 (0 misread); reworded in the commit of product-owner round 1 Lilleman 2026-11-19 16:06:43 +01:00
  • 5d9d880f44 Answer product-owner round 1: 0 🔴, 3 🟡 (0 misread); name the issue to block by in the worker's prompt, and link the issue as Jira creates it Lilleman 2026-11-19 16:06:42 +01:00
  • 90dd09e566 Name the label table and a person's sub-tasks in do-epic's help Lilleman 2026-11-19 15:55:09 +01:00
  • ecd8a99335 Test the team's label in a failed run's split, a second blocker outside the epic and the finish task's heading Lilleman 2026-11-19 15:53:53 +01:00
  • b2630b8148 Let do-epic --file take --blocked-by and --parent, so a change a repo ships as two pull requests gets an issue for each Lilleman 2026-11-19 15:52:49 +01:00
  • e2f5661760 Add the project setting merge = "human", under which chunk-pr merge leaves the pull request ready and a do-epic chunk parks on your merge Lilleman 2026-11-19 15:51:44 +01:00
  • b09d30b48e Answer security round 1: 0 🔴, 0 🟡 (0 misread); no finding Lilleman 2026-11-19 15:48:02 +01:00
  • b26256ce44 Sort the imports the label table added Lilleman 2026-11-19 15:37:03 +01:00
  • 95535b4fb3 Read a team's own repo labels as do-epic's picks, and treat an item a person filed sub-tasks under as split Lilleman 2026-11-19 15:33:51 +01:00
  • 017cf33611 File items 448 to 455 into 0.13.0, so do-epic fits a team's Jira labels, workflow, issue types and merge rule Lilleman 2026-11-19 15:12:20 +01:00
  • e37430a8a2 Move items 446, 445, 365, 368, 309, 419, 444, 366, 339, 341, 298, 395, 425, 447, 367 and 423 into 0.14.0, to clear 0.13.0 for do-epic's fixes Lilleman 2026-11-19 15:07:27 +01:00
  • e1aa89d244 Cut release 0.12.0 v0.12.0 lilleman 2026-11-19 14:15:16 +01:00
  • a7e1297805 Record the comprehension check: paired ruling better at 914b8dc2 against aedc5ed3, Mid A better, Inherited architect better lilleman 2026-11-19 14:08:16 +01:00
  • 04063e1bd8 Say in chunk_start's docstring that the route moved to chunk_driver lilleman 2026-11-19 14:08:11 +01:00
  • 914b8dc2d4 Answer stability round 1: 0 🔴, 4 🟡 (0 misread); a cycle's close takes the task from the cycle, a resolved thread is skipped unread, and tests run git in their own environment lilleman 2026-11-19 14:04:11 +01:00
  • aafe75e16f Answer prose round 1: 0 🔴, 11 🟡 (0 misread); loop-claude and do-epic are entry points everywhere, and each new sentence holds one idea lilleman 2026-11-19 13:44:20 +01:00
  • 2806a8edc9 Kill the mutants that survived on the lines the release review changed lilleman 2026-11-19 13:44:12 +01:00
  • 2edbde8071 Answer security round 1: 0 🔴, 0 🟡 (0 misread); nothing to fix lilleman 2026-11-19 12:57:03 +01:00
  • 762614353c Answer principles round 2: 0 🔴, 2 🟡 (0 misread); workflow-bench hands its effects in as arguments, and no Python file of the repo patches a global lilleman 2026-11-19 12:55:35 +01:00
  • cb3135bc8c Answer principles round 1: 1 🔴, 7 🟡 (0 misread); every effect is handed in as an argument, no test patches a global, and the traces principle scopes tools that run on their user's machine lilleman 2026-11-19 12:42:03 +01:00
  • 8ca612b7a4 Answer architecture round 4: 0 🔴, 0 🟡 (0 misread); loop-claude's suites sit beside their modules, and what tests share across packages lives in each package's doubles.py lilleman 2026-11-19 11:54:34 +01:00
  • 9bc2887ee8 Answer architecture round 3: 0 🔴, 1 🟡 (0 misread); every suite sits beside the module it tests, and test_loop_claude.py holds only loop-claude's own lilleman 2026-11-19 11:43:38 +01:00
  • df210aaeca Answer architecture round 2: 0 🔴, 3 🟡 (0 misread); the cycle's and the chunk run's suites sit beside their modules, and the claim rule and the model override have one home each lilleman 2026-11-19 11:29:32 +01:00
  • b8595db647 Answer architecture round 1: 0 🔴, 36 🟡 (0 misread); chunk_driver holds what both drivers share, and Jira and review-state reads have one home each lilleman 2026-11-19 11:14:44 +01:00
  • ab8bf249cc Record the reading cost before the release review of 0.12.0 lilleman 2026-11-19 10:34:11 +01:00
  • d488b8a324 Release review of 0.12.0 lilleman 2026-11-19 10:34:03 +01:00
  • 00c156d75e Finish item 422 in the plan lilleman 2026-11-19 10:28:15 +01:00
  • aedc5ed33e Record the comprehension check: two scoring runs at 81e8d8a1, Overall 6.50 and 5.88, and refresh floor item 300 lilleman 2026-11-19 10:28:08 +01:00
  • 81e8d8a1b8 Answer stability round 2: 0 🔴, 0 🟡 (0 misread); every answer holds lilleman 2026-11-19 10:14:25 +01:00
  • d06eea3149 Kill the mutants of the judged shells, the named process's lookup and a tool named as a third word lilleman 2026-11-19 10:11:51 +01:00
  • 2471db9108 Answer prose round 2: 0 🔴, 1 🟡 (0 misread); say a heavy slot no longer exempts a shell rather than marking it, and untangle the help's pronouns lilleman 2026-11-19 10:02:12 +01:00
  • a49a790a36 Answer product-owner round 2: 0 🔴, 1 🟡 (0 misread); judge a shell again once it used CPU and went idle, retry a failed judgment, and name a process without a command line lilleman 2026-11-19 10:02:12 +01:00
  • bfcff168c2 Answer prose round 1: 0 🔴, 5 🟡 (0 misread); rewrite the stuck check's help, docstring and prompt, put the verdict first, and drop the nothing-negations lilleman 2026-11-19 09:56:43 +01:00
  • f3af8fda13 Answer product-owner round 1: 0 🔴, 3 🟡 (0 misread); check a shell that holds a heavy slot, name the process's command, and judge each shell once lilleman 2026-11-19 09:56:43 +01:00
  • ed241f3ab5 Answer stability round 1: 1 🔴, 3 🟡 (0 misread); kill each test shell's whole process group, check the CPU sum without timing, and file the headless calls' missing time limit as item 447 lilleman 2026-11-19 09:56:43 +01:00
  • dbfe87d36f Sort the stuck check's imports by their first binding lilleman 2026-11-19 09:26:57 +01:00
  • 53ef45dbd8 Sort the stuck check's imports by their first binding lilleman 2026-11-19 09:26:30 +01:00
  • d251d50472 Kill the mutants of the evidence separator and the CPU sum lilleman 2026-11-19 09:16:14 +01:00
  • a12710ebc4 Kill the mutants of the stuck check, its rung and the /proc readings it hands the investigation lilleman 2026-11-19 09:06:26 +01:00
  • 6b8fbc6ff5 Answer security round 1: 0 🔴, 0 🟡 (0 misread); nothing to fix lilleman 2026-11-19 08:56:07 +01:00
  • 08038d93eb Judge a worker that stalls before its result with a headless claude, and report the verdict without a kill lilleman 2026-11-19 08:55:14 +01:00
  • a559bbfcbf Finish item 420 in the plan lilleman 2026-11-19 08:10:50 +01:00
  • 1e19f7759d Answer stability round 2: 1 🔴, 1 🟡 (0 misread); say a subagent's shells are stopped like the worker's own, and share one polling helper in the tests lilleman 2026-11-19 08:07:00 +01:00
  • 46dc401736 Merge remote-tracking branch 'origin/main' into item-420 lilleman 2026-11-19 07:46:23 +01:00
  • 5d350c6210 Merge remote-tracking branch 'origin/main' into item-310 lilleman 2026-11-19 05:43:00 +01:00
  • 983e7328cf Kill the mutants of stdin, undecodable lines and repeated record writes lilleman 2026-11-19 05:25:51 +01:00
  • 3649a83c0c Remove item 310 again, which merging main into item-442 brought back, and file item 445 to refuse such a plan lilleman 2026-11-19 05:24:50 +01:00
  • 33b5c8babf Finish item 417 in the plan lilleman 2026-11-19 05:23:00 +01:00
  • 330e3f1d09 Merge remote-tracking branch 'origin/main' into item-417 lilleman 2026-11-19 05:22:54 +01:00
  • 47b863fd9f Merge remote-tracking branch 'origin/main' into item-417 lilleman 2026-11-19 05:20:20 +01:00
  • 52fb218e97 Answer stability round 2: 0 🔴, 2 🟡 (0 misread); stopped the timeout test's sleep at once, dropped a duplicate test, reported a found issue without a key lilleman 2026-11-19 05:20:01 +01:00
  • 566df67639 Finish item 442 in the plan lilleman 2026-11-19 05:14:37 +01:00
  • 610edb1ada Merge remote-tracking branch 'origin/main' into item-442 lilleman 2026-11-19 05:12:04 +01:00
  • cff76d2172 Kill the mutants of the child-shell reader and the streamed claude, and end a command at its last terminator lilleman 2026-11-19 05:02:06 +01:00
  • d1136ebb8d Merge remote-tracking branch 'origin/main' into item-442 lilleman 2026-11-19 04:58:11 +01:00
  • 5f4bbdcbe0 Test a refused ranking and the hint for a bare repo name lilleman 2026-11-19 04:57:47 +01:00
  • e019a6b78c Finish item 310 in the plan lilleman 2026-11-19 04:53:47 +01:00
  • 587363c3ed Answer stability round 1: 2 🔴, 5 🟡 (0 misread); fixed the import order and the unused return, tested the scenario steps offline, passed env and timeout in, reused the paged search lilleman 2026-11-19 04:52:10 +01:00
  • d9875ee460 File item 443: measure the compaction's saving lilleman 2026-11-19 04:47:57 +01:00
  • e55b8826fb Merge remote-tracking branch 'origin/main' into item-310 lilleman 2026-11-19 04:47:53 +01:00
  • 194e3e18ad Answer prose round 1: 0 🔴, 7 🟡 (0 misread); fixes are in the stability answer lilleman 2026-11-19 04:46:55 +01:00
  • 112536ba4e Answer product-owner round 1: 0 🔴, 3 🟡 (0 misread); fixes are in the stability answer lilleman 2026-11-19 04:46:55 +01:00
  • 0ce3cc6ac6 Answer stability round 1: 0 🔴, 5 🟡 (1 misread); start chunks after a partial planning and keep what planning wrote lilleman 2026-11-19 04:46:55 +01:00
  • 1fe59960fb File item 398: measure the compaction's saving lilleman 2026-11-19 04:45:19 +01:00
  • f0f9d91396 Answer product-owner round 1: 0 🔴, 1 🟡 (0 misread); announce the compaction on a line of its own, fixed in the stability answer lilleman 2026-11-19 04:44:37 +01:00
  • a2ea673822 Answer prose round 1: 0 🔴, 2 🟡 (0 misread); reword the timer's docstrings, comments and help, rename the no-compaction test lilleman 2026-11-19 04:44:37 +01:00
  • 708347382b Check the stopped timer once stop returns, not after a wait lilleman 2026-11-19 04:44:37 +01:00
  • 78dd557057 Answer stability round 1: 0 🔴, 2 🟡 (1 misread); compact on the worker's rung, announce the compaction and skip it once the cache expired lilleman 2026-11-19 04:42:14 +01:00
  • 999fca6f17 Pin planning's progress line and the repo a goals failure names lilleman 2026-11-19 04:23:53 +01:00
  • 6ee55681a3 Check that a refused search is reported as the refusal lilleman 2026-11-19 04:15:05 +01:00
  • 30595cfdf4 Test the compaction's boundaries, an unreadable transcript, a usage limit and the timer's daemon thread lilleman 2026-11-19 04:11:13 +01:00
  • 94b34e509f Test what planning prints, asks and refuses lilleman 2026-11-19 04:04:32 +01:00
  • 7a798a9e37 Kill the last surviving mutants of the sandbox and settings of the e2e command lilleman 2026-11-19 04:04:13 +01:00
  • 59c02ef8d1 Answer security round 1: 0 🔴, 0 🟡 (0 misread); no finding lilleman 2026-11-19 04:02:58 +01:00
  • 444f5824d9 Answer security round 1: 0 🔴, 0 🟡 (0 misread); no finding lilleman 2026-11-19 04:00:32 +01:00