b7a857d12f
fix(control-plane): derive the audit-log root from the registry in the binding and the shell defaults
GodopuandClaude Fable 5.12026-10-13 09:57:42 +00:00
e32c957464
merge: fix/u1e-workspace-root (u1e no longer writes the checkout's .mam/shim/herdr; v4.9.3)
GodopuandClaude Opus 5.52026-10-13 12:33:10 +09:00
121eab71d4
release(v4.9.3): u1e no longer writes the checkout's .mam/shim/herdr - archive the review reports and bump the lockstep version
GodopuandClaude Opus 5.52026-10-13 12:33:02 +09:00
3a30d7da89
fix(tests): u1e gives the wrapper its own WORKSPACE_ROOT - it wrote the checkout's .mam/shim/herdr
GodopuandClaude Opus 5.52026-10-13 11:57:46 +09:00
42e2a7ea9b
merge: docs/readme-refresh (README rewritten against the code, README.ko.md in step; v4.9.2)
GodopuandClaude Opus 5.52026-10-13 11:57:34 +09:00
e3bcdee834
release(v4.9.2): README rewritten against the code, README.ko.md in step - archive the review reports and bump the lockstep version
GodopuandClaude Opus 5.52026-10-13 11:57:33 +09:00
2ccf43911c
merge: fix/test-herdr-leak (tests no longer reach a real herdr server: a *-shim ancestor hid the mock, the sandbox kept the pane's socket; v4.9.1)
GodopuandClaude Opus 5.52026-10-13 10:33:04 +09:00
50c8da8269
release(v4.9.1): tests no longer reach a real herdr server - archive the review reports and bump the lockstep version
GodopuandClaude Opus 5.52026-10-13 10:33:04 +09:00
db7f4cab9c
test(sandbox): mam_sandbox also drops HERDR_SESSION (reviews 63919c5d / 191f0ec3, non-blocking)
GodopuandClaude Opus 5.52026-10-13 10:01:55 +09:00
525df2bb58
fix(tests): tests no longer reach a real herdr server - a *-shim ancestor hid the mock, the sandbox kept the pane's socket
GodopuandClaude Opus 5.52026-10-13 09:15:22 +09:00
8b0b4a07dc
merge: fix/issue-50-55 (Issue #50 / #55: onboarding and dispatch wait by the registry, one pane claim per server, tests stay out of the real agent homes; v4.9.0)
Godopu2026-10-12 12:38:55 +00:00
ffe8665b27
release(v4.9.0): Issue #50 / #55 - archive the review reports and bump the lockstep version
GodopuandClaude Opus 5.52026-10-12 21:38:50 +09:00
21f0c12b9e
fix(issue-50): the monitor labels a row with the server that answered
GodopuandClaude Opus 5.52026-10-12 10:37:50 +09:00
efd27def47
fix(issue-50): a claimed dry-run job holds its session (review 211037ab F1)
GodopuandClaude Opus 5.52026-10-12 10:12:05 +09:00
320aa9f7a0
test(issue-50): c10 makes the same-second tie instead of timing it (reviews 98b31bb2 / 352066ec / 244b8072 F-B)
GodopuandClaude Fable 5.12026-10-11 12:05:19 +00:00
6c99fedf40
fix(issue-50): an injected record holds its session whatever its FIFO place; the ambient pane claim locks the caller's server (reviews 5a09a13f F-A, e8e471ff R1)
GodopuandClaude Fable 5.12026-10-11 08:50:46 +00:00
d50610c1ab
fix(issue-50): a waiter whose own attempt is already running answers clear, not superseded
GodopuandClaude Fable 5.12026-10-11 06:28:01 +00:00
7844ae2c5f
fix(create): one pane claim at a time per herdr server; a create keeps the pane the monitor registered for it
GodopuandClaude Fable 5.12026-10-11 06:05:52 +00:00
30c25f193c
fix(issue-50): the waiter and its hold are pinned to one attempt and session; an unsignalled pending record lives by its ack budget (review 55219be1 R1, R2)
GodopuandClaude Fable 5.12026-10-11 02:58:16 +00:00
9cbe08e8de
fix(issue-50): the registry wait admits one job per session and holds it (review aa1bff92 F1, F2)
GodopuandClaude Fable 5.12026-10-11 01:29:57 +00:00
9e78ff3ba3
fix(issue-50): the loop path reads this attempt's token before it waits (review reviewer-grok-01)
GodopuandClaude Fable 5.12026-10-10 18:14:48 +00:00
3705bc72a8
fix(issue-50): the loop path waits by the registry on its pending record (review F1-F3)
GodopuandClaude Fable 5.12026-10-10 17:28:09 +00:00
630f04bfdc
test(issue-50): the two new knobs join the #26 table; the herdr box sweeps a late state write
GodopuandClaude Fable 5.12026-10-10 16:05:59 +00:00
e9efabf74b
feat(issue-50): the delegate-job wrapper waits by the registry before it injects (C)
GodopuandClaude Fable 5.12026-10-10 15:39:21 +00:00
c29c5b2142
feat(issue-50): create --onboard waits for the onboarding job to end (A)
GodopuandClaude Fable 5.12026-10-10 15:39:21 +00:00
40293841e7
test(issue-55): tests no longer write into the user's real agent homes
GodopuandClaude Fable 5.12026-10-10 14:29:36 +00:00
42e51735ba
merge: test/loop-marker-tmp (tests no longer take the checkout's loop lock; v4.8.9)
Godopu2026-10-09 01:59:58 +00:00
ca688dcef4
release(v4.8.9): tests no longer take the checkout's loop lock - archive the review reports and bump the lockstep version
GodopuandClaude Opus 5.52026-10-09 10:59:38 +09:00
a93c9bf292
test(loop-lock): the b13 runs of the checkout's run_loop.sh take a loop lock of their own
GodopuandClaude Opus 5.52026-10-09 09:01:34 +09:00
97c2946ee0
merge: test/issue17-workspace-root (Issue #17 follow-up: tests that source lib.sh no longer write into an inherited WORKSPACE_ROOT; v4.8.8)
Godopu2026-10-08 20:01:25 +00:00
e0181ea9be
release(v4.8.8): Issue #17 follow-up - archive the review reports and bump the lockstep version
GodopuandClaude Opus 5.52026-10-09 05:01:05 +09:00
ac48cb7861
test(issue-17): test_o1_rebuttal and test_o2_race_free_lock run run_loop.sh under their own WORKSPACE_ROOT (review claude, grok)
GodopuandClaude Opus 5.52026-10-09 02:20:24 +09:00
8b47cc76ee
test(issue-17): every test that sources lib.sh does it under its own WORKSPACE_ROOT (review grok, codex)
GodopuandClaude Opus 5.52026-10-09 00:21:38 +09:00
d198dd50aa
test(issue-17): item-4 helpers run lib.sh in the test's own WORKSPACE_ROOT
GodopuandClaude Opus 5.52026-10-08 23:10:35 +09:00
6b00b6e06f
merge: fix/issue17-followups (Issue #17 follow-ups: a failed decision ends the wrapper with 1; resume closes herdr 0.9.3's placeholder; v4.8.7)
Godopu2026-10-08 02:32:57 +00:00
cd82ab6e72
release(v4.8.7): Issue #17 follow-ups - archive the review reports and bump the lockstep version
GodopuandClaude Opus 5.52026-10-08 11:32:39 +09:00
65b78b6ec4
fix(issue-17): a failed yes/no decision is not a "no" (review codex R1)
GodopuandClaude Opus 5.52026-10-08 09:26:41 +09:00
541c3febe8
test(issue-17): the frozen wrapper literals follow _decided/_exit_for (review agy)
GodopuandClaude Opus 5.52026-10-08 09:26:41 +09:00
8241bd53b1
test(issue-17): the failing wait-rc test sources the wrapper in its own WORKSPACE_ROOT
GodopuandClaude Opus 5.52026-10-08 08:14:05 +09:00
d6cb01987c
docs(issue-17): wrapper.py imports only the stdlib, the package it runs in loads paho (N1)
GodopuandClaude Opus 5.52026-10-08 08:12:44 +09:00
a670ef53ee
fix(issue-17): a decision the wrapper cannot make ends it with 1 (N2)
GodopuandClaude Opus 5.52026-10-08 08:12:16 +09:00
f19f94099f
fix(issue-17): resume closes herdr 0.9.3's placeholder for a removed directory
GodopuandClaude Opus 5.52026-10-08 07:53:58 +09:00
25c99704f2
test(issue-17): herdr 0.9.3 keeps a removed directory as the restored pane's cwd
GodopuandClaude Opus 5.52026-10-08 07:52:40 +09:00
56d0bfda54
merge: refactor/issue-17-item4 (Issue #17 item 4: the delegate-job wrapper's decisions move to Python; v4.8.6)
Godopu2026-10-07 16:37:53 +00:00
8fde39be33
release(v4.8.6): Issue #17 item 4 - archive PR #51 reviews and bump the lockstep version
GodopuandClaude Opus 5.52026-10-08 01:37:48 +09:00
4c86107846
fix(issue-17): review R1 - a finalize wait is measured on the monotonic clock, like bash's sleep
GodopuandClaude Opus 5.52026-10-08 00:53:09 +09:00
518bedac5d
fix(issue-17): review R1 round 2 - an exited, unreaped wrapper is not a live owner
GodopuandClaude Opus 5.52026-10-06 08:09:13 +00:00
c5cebe6c32
fix(issue-17): review R1 - the finalizer ends with its wrapper
GodopuandClaude Opus 5.52026-10-06 07:02:26 +00:00
66d1490230
test(issue-17): item 4 - l15's stand-in control root carries lib_py/ints.py next to lib_int.sh
GodopuandClaude Opus 5.52026-10-06 05:37:02 +00:00
6947d3c785
test(issue-17): item 4 - the #42 inventory follows runner.py's shorter header
GodopuandClaude Opus 5.52026-10-06 05:36:48 +00:00
3b3ce307df
test(issue-17): item 4 - the #42 inventory points at main() where it now is
GodopuandClaude Opus 5.52026-10-06 05:36:23 +00:00
6bf9f92462
test(issue-17): item 4 - the frozen-mapping test imports the wrapper module directly (it has no scripts/ shim)
GodopuandClaude Opus 5.52026-10-06 05:08:35 +00:00
63d9c205ff
refactor(issue-17): item 4 - the delegate-job wrapper's decisions move to Python
GodopuandClaude Opus 5.52026-10-06 04:59:58 +00:00
7209750ac7
merge: fix/issue-45 (Issue #45 A+B: MAM herdr servers start with restore-on-restart off; resume closes its own restored empty shell; v4.8.5)
Godopu2026-10-05 09:56:39 +00:00
bf8a583afb
docs(v4.8.5): archive PR #48 reviews (three rounds) and record the merge-workspace review
GodopuandClaude Opus 5.52026-10-05 18:56:38 +09:00
ab93776d6d
fix(issue-45): review round 2 - an unrunnable config check is not a pass, odd pids never close a pane, nan does not drop settings
GodopuandClaude Opus 5.52026-10-05 17:50:15 +09:00
e8d9151226
test(issue-45): review - the real-server tests run on macOS, and the bootstrap config check follows the bootstrap's python3
GodopuandClaude Opus 5.52026-10-05 15:59:26 +09:00
36f0b8dfd6
test(issue-45): the herdr-config tests collect on Python < 3.11; the docs say what that Python loses
GodopuandClaude Opus 5.52026-10-05 15:56:27 +09:00
35935f9e8a
fix(issue-45): review round 2 R3 - the .mam fallback obeys the same no-overwrite rules
GodopuandClaude Opus 5.52026-10-04 11:49:08 +00:00
7c544c06ab
test(issue-45): S11 base follows the review-R1 heredoc change; VERSIONS notes
GodopuandClaude Opus 5.52026-10-04 11:18:09 +00:00
811b080966
fix(issue-45): review R1-R3 - herdr checks the config, any TOML spelling kept, source never overwritten
GodopuandClaude Opus 5.52026-10-04 11:17:59 +00:00
ce91cc11b9
release(v4.8.5): bump framework and eight skills to 4.8.5 in lockstep
GodopuandClaude Opus 5.52026-10-04 10:53:14 +00:00
44999ff2fb
release(v4.8.3): bump framework and eight skills to 4.8.3 in lockstep; archive PR #46 reviews (two rounds)
GodopuandClaude Opus 5.52026-10-04 08:36:15 +09:00
17eba0904e
docs(issue-42): review - attach names the herdr server, --dry-run is not "write nothing", a deleted block stays deleted
GodopuandClaude Opus 5.52026-10-04 08:06:32 +09:00
8d26106195
test(issue-42): review R2 - a passthrough code must be in the exit table
GodopuandClaude Opus 5.52026-10-03 14:10:59 +00:00
e6a46cd7c4
fix(issue-42): review R1 - stop's exit 1 also comes after the stop
GodopuandClaude Opus 5.52026-10-03 13:56:36 +00:00
7670b41afd
docs(issue-42): stages 1-2 - SKILL.md code-block inventory as test data; remove blocks the scripts already do
GodopuandClaude Opus 5.52026-10-03 13:42:08 +00:00
3563a6fc17
release(v4.8.2): bump framework and eight skills to 4.8.2 in lockstep; archive the follow-up loop's reports
GodopuandClaude Opus 5.52026-10-03 00:27:24 +09:00
75f364e4b4
fix(followups): F2 review R1 - a failed env load is not cached as loaded
GodopuandClaude Opus 5.52026-10-02 09:29:11 +09:00
95eabc693f
fix(followups): F2 - job_subscriber returns the env-file failure by name
GodopuandClaude Opus 5.52026-10-02 08:39:22 +09:00
1e92ddc563
test(followups): F6 - the wrapper vanishes inside the second wait; the <= 60 s is the herdr wait's
GodopuandClaude Opus 5.52026-10-02 08:23:31 +09:00
fb1b93c665
fix(followups): F4b - bash integer knobs that reach [ -ge/-le/-gt ] or sleep go through mam_int_knob
GodopuandClaude Opus 5.52026-10-02 08:23:31 +09:00
8b046390be
fix(followups): F4a/F4c - the runner's wait knobs follow the integer rule; MAM_WAIT_GRACE_S documented
GodopuandClaude Opus 5.52026-10-02 08:13:44 +09:00
71ea8095c5
fix(followups): F3 - .mam.env BOM, repeated keys and a missing MAM_ENV_FILE
GodopuandClaude Opus 5.52026-10-02 08:10:42 +09:00
5edf913d22
fix(followups): F2 - Python entry points report an unreadable env file in one line
GodopuandClaude Opus 5.52026-10-02 08:09:46 +09:00
75936f11c6
test(followups): F1 - A8 writes the marker lstart as loop_lock.sh reads it
GodopuandClaude Opus 5.52026-10-02 08:08:26 +09:00
0051fd6c3b
merge: fix/issue-26 (Issue #26 one .mam.env parser, decimal integer knobs, tests on a local MQTT broker; v4.8.1)
Godopu2026-10-01 15:23:26 +00:00