The .claude/skills check accepts both local layouts and fails on a dangling
link. Review: round 1 (@e1ed270) claude agy grok PASS, codex NOT PASS (a
dangling top-level link was skipped); fixed in 150b493; round 2 4/4 PASS.
Reports and the round-2 identity sidecars go to .agents/reports/<session>/.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
Issue #42 stages 1-2: the SKILL.md code-block inventory as test data and
the removal of blocks the entry scripts already do. Merge-workspace review:
round 1 (@8d26106) claude agy grok PASS, codex NOT PASS (R1-R3); fixed in
17eba09 (docs, inventory data and the guard test only); round 2 4/4 PASS.
Full suite at 8d26106 1408 passed / 0 failed. Reports and the round-2
identity sidecars go to .agents/reports/<session>/.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
PR #41/#43 review follow-ups F1-F6, implemented by creator-claude-01 in a
loop and reviewed by claude, agy, grok and codex: round 1 (@95eabc6)
codex NOT PASS (a failed env load cached as loaded), fixed in 75f364e;
round 2 4/4 PASS. Full suite at 75f364e 1380 passed / 0 failed. Reviewer
reports (round 2 with identity sidecars) and the creator's reports go to
.agents/reports/<session>/.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
delegate-job SKILL.md said a watcher outlives a killed wrapper by "one
wait, <= 60 s". That bound is the herdr wait's; a wrapper that goes while a
notice is being published is noticed when that publish ends. The text now
says both (PR #41 codex round 3).
"a wrapper gone during the wait" switched alive() by call count, so the
watcher saw the wrapper gone at the next loop head and never entered the
second wait (ticks == [1] with that band). The wrapper now vanishes inside
the second herdr wait (_during_notice's in_wait hook), and the test checks
that the second wait happened and that alive() was true until then. The
same test fails when the post-wait _wrapper_gone() check is removed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HCGpJXhSZasJLU38ksDghe
Issue #26: one .mam.env parser for the wrapper and the Python tools,
decimal integer knobs, tests on a local MQTT broker. Merge-workspace
review: round 1 (@b58edac) claude agy PASS, codex NOT PASS (R1-R4), grok
no verdict (its late report, archived as report-c9ed5b14.md, found R1 too);
round 2 (@0eec795) 4/4 PASS. Reports and round-2 identity sidecars go to
.agents/reports/<session>/.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
Merge-workspace review of PR #43 (codex R1-R4, claude R1/R2).
- R1: envfile.parse raises EnvFileError when the file exists but cannot be
read or decoded as a whole; `--shell` prints the cause and exits 1, so the
wrapper stops (rc 1, as before #26) instead of running on defaults, and
mqtt_common's entry points stop on the same error. A bad single line is
still a warning. The wrapper's message no longer claims a read failure
when the cause is elsewhere ("cannot load ... see the line above").
- R2: lib.sh mam_int spells out the ASCII blanks and digits instead of
[[:space:]]/[0-9], which follow the locale: U+2003 30 U+2003 was 30 in the
wrapper and invalid in the runner under en_US.UTF-8. The test oracle
_expected uses the same ASCII rule.
- R3: an integer flag given explicitly with an empty value is refused like
any other invalid value; only an absent flag keeps its default.
- R4: a line holding a NUL byte is skipped with a warning by both loaders;
before, the wrapper read "leftright" and the Python loader died.
- .mam.env.example states the format: no expansion, `export` and bash's own
variables skipped, a trailing `# ...` is part of the value, the environment
wins, an unreadable file is an error.
Tests: unreadable (0000) and invalid-UTF-8 files stop the wrapper and the
Python loader; a NUL line is skipped by both; Unicode blanks/digits are
invalid for mam_int, _positive_int, the wrapper and the runner under C and
en_US.UTF-8; explicit empty values for the five integer flags are refused.
All of them fail on b58edac.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
R1 (P1): a quoted export still ran code when the key was a bash special
variable (RANDOM=a[$(cmd)] is evaluated arithmetically on assignment).
envfile.RESERVED (bash's own variables, plus BASH_/COMP_/READLINE_
prefixes) is skipped with a line-numbered warning by both loaders; a test
keeps it covering every variable a clean bash defines. The wrapper also
hands SCRIPT_DIR/REPO_ROOT/TARGET_ENV to the parser as environment, so the
file cannot replace the variables it has already set.
R2 (P2): SKS_QUIESCENT_TRIES reached `for ((i < tries))` through a function
argument; it is normalised at the call site (minimum 0). B5 now requires
every integer-looking env default in the shell code to be normalised or
listed in NOT_ARITHMETIC with where it goes, and exercises _pane_quiescent.
R3 (P2): _positive_int strips leading zeros and rejects more than nine
significant digits before int(), like mam_int; 5000-digit inputs no longer
raise in the runner.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
bash read MAM_JOB_TIMEOUT_S=0700 as octal 448 (and failed on 0900) while
the loop runner read 700. One rule now, with a per-knob minimum:
lib.sh mam_int / mam_int_knob and runner._positive_int (blanks allowed,
digits only, decimal, min <= n <= 999999999).
- minimum 1: MAM_JOB_TIMEOUT_S, MAM_HEARTBEAT_SEC, MAM_HEARTBEAT_MISSES,
MAM_ACK_TIMEOUT_S (invalid env: warning + default) and the wrapper's
--timeout/--idle-timeout/--heartbeat/--heartbeat-misses/--ack-timeout
(invalid: rc 1); the runner's job_budget, --job-timeout, contract_args.
- minimum 0 (0 = check once): MAM_INJECT_ACK_GRACE_S,
delegate_wait_started's ack_s, SKS_DIALOG_TIMEOUT, SKS_BUSY_TIMEOUT,
SKS_DELIVERY_GRACE_S.
- tests/test_issue26_ints.py: INT_KNOBS is the one table of minimums; a
static guard keeps env defaults out of arithmetic and checks each
normalised knob against it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- lib_py/delegation/envfile.py (stdlib only) holds the one set of rules:
comments and blank lines skipped, key and value stripped, one matching
quote pair removed, no shell interpretation, keys must match
^[A-Za-z_][A-Za-z0-9_]*$ (bad or `=`-less lines warn with the line
number, never the value).
- mqtt_common._parse_env_file delegates to envfile.apply.
- The delegate-job wrapper runs envfile.py --shell by path and evals the
shlex-quoted exports; values never reach a command line. The environment
now wins over the file in both loaders.
- .mam.env.example: commented assignments no longer use $HOME, which the
parser keeps literally.
The shell parsed the env file in one place only (wrapper :38), not three.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Issue #33 (job budget, wall_clock_timeout vs deadline_timeout) and
Issue #30 (notify-only stall watcher). Reviews of PR #41 in the merge
workspace: round 1 (@1673818) claude N1 / codex R1 R2 NOT PASS, agy grok
PASS; round 2 (@af0a2a2) grok N1 / codex D1 D2 NOT PASS, claude agy PASS;
round 3 (@9200562) 4/4 PASS. Reports and the round-3 identity sidecars go
to .agents/reports/<session>/. The design consensus on items A-D is in the
VERSIONS.md entry.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
grok N1: the wrapper spent up to 5 s on a TERM budget before it finalized the
job, so when the watcher swallowed the TERM the silent job's close came ~5 s
late and a `completed` written in that gap won (test_issue14 idle-grace).
codex D1: `_notify` blocks on a publish for up to 120 s, and the prompt went
out right after it; a completed, a re-dispatch or an unreadable record in that
window still got the nudge. codex D2: the wrapper pid was checked only at the
loop head, so a watcher whose wrapper vanished during a wait could still
notify and nudge once.
- multi-agent-mux-delegate-job `stop_stall_watch`: KILL + wait, no budget.
The watcher holds nothing to flush (its TERM handler is a bare os._exit),
and KILL cannot be swallowed.
- lib_py/stall_watch.py: `_wrapper_gone()` after every wait (before any
notice) and at the loop head; `_may_prompt()` right before the keys - the
wrapper is alive and the record, read then, is open and this dispatch;
`_dispatch_open(strict=True)` treats an unreadable record as "no".
- SKILL.md: the lifetime contract states the one-wait lag and that nothing is
sent after the wrapper is seen gone.
- tests/test_issue30_stall_watch_stop.py: the stop returns at once against a
TERM-ignoring stand-in (took 6 s on the round-2 code); completed,
re-dispatch and a torn record during the notice publish withhold the nudge;
a wrapper gone during the wait gets no notice and no nudge. All five fail
on the round-2 code.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
Round-1 review of PR #41 (claude N1, codex R1/R2).
N1 - the wrapper stopped its stall watcher with `kill` followed by an
unbounded `wait`. A TERM that lands in the watcher's fork-exec window is
swallowed (bash defers terminating signals in the forked child until the
exec), the watcher then installs its own handler and lives until the job
closes, and the wrapper hung until someone killed the watcher by hand
(reproduced twice in test_issue14's ack test; the wrapper trace stops at
`kill` + `wait`).
R1 - a wrapper that is itself terminated left the watcher orphaned.
R2 - a `completed` or a re-dispatch that landed during a blocking herdr
wait (up to 60 s) was still followed by the idle notice and the nudge.
- multi-agent-mux-delegate-job: `stop_stall_watch` sends TERM, polls up
to 5 s, then KILLs and reports "ignored TERM; killed"; `start_stall_watch`
passes `--watch-pid $$`.
- lib_py/stall_watch.py: `--watch-pid` (rc 0 once that pid is gone, checked
between herdr calls); `_dispatch_open()` re-reads the record after every
wait and right before the nudge - terminal status, another dispatch token
or attempt, or a missing record ends the run without a notice; a refused
`stalled:idle` publish withholds the nudge (`_notify` returns False).
- delegate-job SKILL.md: the watcher's lifetime contract; exit-code rows 2
and 3 name the idle clock and the wall-clock label; "before v4.7.5" ->
v4.8.0 (M5, NB2).
- tests/test_issue30_stall_watch.py (M8): the module-wide Linux mark is
narrowed to S6 (/proc); r4's stub stamps time.time() so its cross-process
comparison holds on macOS too (47 passed, 1 skipped here).
- tests/test_issue30_stall_watch_stop.py (new, macOS and Linux): the stop
routine against a TERM-ignoring stand-in returns in 4-12 s; the wrapper
returns (rc 3) with a TERM-ignoring watcher shim; `--watch-pid` ends the
watcher; a TERM to the real wrapper leaves no watcher; a job completed or
re-dispatched during the wait is not nudged; a refused notice withholds
the nudge. Every one of them hangs or fails on the round-1 code.
Not in this commit (follow-ups): M7 test hook in product code (no adapter
is unreliable, a seam is needed), M9/B3 public-broker latency (on this host
even `started` sometimes never reaches the subscriber), the leading-zero
budget parse (NB1) and M1/M2/M4/M6 go on the #17 item 4 list.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
- job_subscriber.py returns only named codes (RC_COMPLETED 0, RC_ERROR 1,
RC_NO_EVENT 2, RC_BROKER_UNAVAILABLE 3, RC_WALL_CLOCK 4). The wrapper
mirrors them as SUB_RC_* and compares subscriber codes only by name; a test
keeps the mirror equal to the Python constants and forbids numeric
comparisons of $sub_rc/$sub_exit and numeric `return` in the subscriber.
- `delegate-job wait` passed the subscriber's code straight through, so a
spent budget would have reached its callers as 4; it is 2 again (B7 fails
without the mapping).
- tier1 G-8 gave both clocks 0.5 s and raced; the wall clock now runs out
first by construction (idle 30 s) and the test asserts RC_WALL_CLOCK.
Other subscriber rc asserts use the names.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- R1 (P1): without rules the input region runs from the last prompt line to
the end of the pane; only a last line matching the adapter's footer pattern
(codex: ` <model> · ~/<cwd> · …`) is dropped, so a draft on a second line
keeps the nudge withheld.
- R2: while the synchronous submit runs, a relay thread finds the job it
registers (same session, created since the submit) and prints its
stalled:/stall_watch: notices; before, they appeared only after the job.
- R3: the broker-unavailable disk wait that runs into the deadline is the wall
clock: wall_clock_timeout (caller rc stays 3).
- R4: a state_change_seq change seen right before the judgement restarts the
idle window (decision 9: S seconds of continuous idle).
Regression tests are the reviewer's reproductions, each with its passing
twin; the wrapper-level #30 tests now use this environment's broker instead
of the #18 harness's public default.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
lib_py/stall_watch.py runs beside the wrapper's subscriber for a local herdr
job (started after delivery, stopped when the subscriber ends; MAM_STALL_WATCH=0
turns it off). It follows `herdr agent wait` transitions and publishes only
`progress` events through one publisher; it has no path that closes a job.
- blocked: `stalled:blocked <detection line>` once per episode, never answered.
- idle for MAM_STALL_IDLE_S (600) with no background descendants of the agent
pid (pane process-info + /proc; adapter resident_children and the heartbeat
timer excluded) and an empty input box: `stalled:idle` and one nudge.
- input box judged from adapter facts measured on herdr 0.9.1 (codex '›',
grok frame '│', opencode '┃' block + stop pattern); the whole-pane fallback
is gone. No facts: withheld + `stall_watch:input_check_unavailable` once;
never empty for 3 x S: `stall_watch:input_never_empty` once.
- herdr failures are not "no event": warning + `stall_watch:herdr_unavailable:
<code>` (or agent_not_found), backoff, rc 2 after MAM_STALL_HERDR_FAIL_S
(120); the wrapper prints the watcher's last lines.
- remote and human jobs are skipped with one log line.
- agent_status_reliable is an adapter field; false turns idle judgement off.
- the loop runner relays stalled:/stall_watch: notices once each.
- scoped_env drops a pane-inherited workspace scope for name lookups (#34 rule).
The module sits at the top of lib_py: delegation and loop may not import the
agent adapters (their sibling-import rules). Idle pane captures of the five
CLIs are fixtures; the mock herdr answers process-info in the real shape.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- wrapper: TIMEOUT falls back to MAM_JOB_TIMEOUT_S (--timeout > env > 3600);
an invalid value warns and keeps 3600.
- loop: --job-timeout <sec> (run_loop.sh passes it on; refused in the legacy
shell loop). The runner always hands the wrapper --timeout, so the budget
the wrapper registers is the one its wait caps come from.
- runner: wait_for_job and the MAM_CHILD_WAIT_S default are budget + 300 s
(MAM_WAIT_GRACE_S), today's 3900 with no option; an explicit
MAM_CHILD_WAIT_S still wins.
- subscriber: the wall-clock budget returns 4 (RC_WALL_CLOCK), silence keeps
2 (RC_NO_EVENT). The wrapper maps both to caller rc 2 and labels the close
from the rc: wall_clock_timeout vs deadline_timeout; the disk-heartbeat
grace that ends on the deadline is the wall clock too.
- registry close accepts wall_clock_timeout (engine: -> error); the remote
runner's deadline and command-timeout details lead with wall_clock_timeout;
the sub-job table files it CLOSED-UNCONFIRMED (first token, colon stripped).
Test expectations changed only where the contract changed: the loop's
contract_args now starts with --timeout, the subscriber's exit-code docstring
names rc 2/4, and the registry --help capture lists the new close kind.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Issue #34: the loop waits for the creator job's sub-jobs, reviewers get
the sub-job table as evidence, a run with sub-jobs voids a round whose
source changed during review, and a pane caller reaches an agent of
another workspace on its own server. VERSIONS.md carries the entry; the
merge-side reports (round 1 @ed293dc and round 2 @e80186f with identity
sidecars, base ed293dc -> head e80186f, evidence review-tree) are archived
under .agents/reports/<session>/.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
- mam_herdr: for name lookups (agent / has-session / kill-session) a
HERDR_WORKSPACE_ID that is the caller pane's own workspace (it matches
the HERDR_PANE_ID prefix) is dropped. Live agent names are unique per
server (herdr --skill), so this cannot address a same-named agent of
another workspace; an id the caller chose, or one set outside a pane,
still scopes (H-18/H-21/H-23). The generated shim is unchanged (S11).
- wrapper: a preflight that cannot see the target closes the job
cancelled with detail target_not_found instead of session_busy; the
loop's sub-job table counts it as NOT-DISPATCHED.
- tests: T10-T12 with the measured pane env (same-server socket, caller
w6, target w2); T8 follows the sub-job-armed guard.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- registry: parent_job field, register --parent-job fails closed on an
unknown parent, children_of/descendants_of, list --parent
- wrapper: --parent-job; $MAM_JOB_ID default (ignored with a warning when
this registry lacks it); brief line and one sentence at the end of the
injected instructions
- create_session: the onboarding instructions carry the same sentence
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Issue #36: delegated prompts travel by file, a delegation that fails ends
the loop with rc 1. VERSIONS.md carries the entry; the merge-side reports
(round 1 @59be1c8 and round 2 @708262a with identity sidecars, base
59be1c8 -> head 708262a, evidence review-tree) are archived under
.agents/reports/<session>/.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
- runner._run_cmd: the write of the prompt file was outside the cleanup
and error handling that the exec had (merge review: claude H1, codex R1).
A failed write now removes the partial file and returns the same 126
failed submit an unstartable wrapper does, so the loop ends with rc 1
and no traceback. Test S4c.
- test S9 required the literal "File too large" in stderr; bash 3.2 on
macOS prints nothing for EFBIG, so the whole suite failed on this host
(agy, grok, codex R2). The contract asserted everywhere is rc != 0, no
leftover file and the portable diagnostic line; the libc wording is
checked on Linux only.
- the wrapper accepted an empty --prompt-file and registered an empty
prompt (claude H3); it is refused now. Test S1e.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
One argv string is capped at MAX_ARG_STRLEN (131,072 bytes on Linux), so a
prompt passed as --prompt could not reach the registry once it carried a
large diff. submit now takes --prompt-file and passes the same path to
`registry.py register --prompt-file`; the loop type's per-iteration update
goes by file too. --prompt keeps working unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Issues #23 and #24: cross-server herdr scope, agent-agnostic start-up
dialogs, onboarding delivery judged by the agent's own started event,
verified rollback. VERSIONS.md carries the entry; the merge-side reports
(round 1: claude/agy/grok; round 2 on 05b0ea4: all four, with identity
sidecars, base f4ba205 -> head 05b0ea4, evidence review-tree) are archived
under .agents/reports/<session>/. The round-1 codex report was lost to the
host sleeping mid-review; its two probes are recorded in the changelog.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
Review round 3 (planner-reviewer-codex-01, reviewer-opencode-01): R1 blocking,
R2 and the "ONE builder" wording non-blocking. All three reproduced; none
rebutted.
R1. _wait_started still returned 0 as soon as the record left `pending`,
ahead of the shared predicate, so `error`, `cancelled` and another dispatch's
`running` all acknowledged an injection that create's delegate_wait_started
rejects - the two contracts the ack-finding brief said must not diverge. Both
reviewers reproduced it with the real producer and no record manipulation;
my S7n had hidden it by resetting the record to `pending`.
The bypass and the console grep are gone. The wrapper's evidence is now
delegate_started_recorded alone, the same call create makes. For a remote
agent - no shared disk, so no line in events.log - the predicate reads the
subscriber's audit record of the `started` it received. That record is
attributable by construction: the subscriber verifies the HMAC and drops any
event whose data.dispatch_token differs from the job's expected token before
it writes anything. The token itself cannot go into the record (the audit
redactor masks token-named keys), so the subscriber now stamps a boolean
dispatch_verified, and with a token demanded only a verified record counts;
an untokened legacy record fails closed. The registry status is never
evidence; a terminal job is preserved by the terminal logic, not turned into
an acknowledgement.
Tests: W1/W2/W3 are the reviewers' three negatives and the positive, real
producer, record left exactly as it produced it; W4 runs create's wait and
the wrapper's wait side by side on all four and requires the same answer;
W5 pins the remote route (a verified record counts, an unverified or
non-started one does not); W6 drives the real _Watcher and shows it stamps
the flag on a matching started and records nothing for a mismatch; S7n no
longer resets the record. test_issue14's "registry ack over unconfirmed
inject" produced its ack with a bare status flip - the very bypass - and now
produces it the way an agent does, publish_event with the dispatch token;
its scenario and every other assertion are unchanged. test_issue18's remote
e2e accepted the status messages and now expects the shared one.
R2. _code_only blanked a docstring's whole line range, so code sharing that
line - `"""doc"""; call()` - vanished from the scan. It now removes only the
docstring node from the AST and unparses the rest (comments drop out with
it, code strings stay); test_g9b pins the four same-line shapes.
The builder comment claimed one builder for create and the wrapper; the
wrapper keeps _pub_cmd and S7l pins the two equal argument-for-argument. The
comment now says that.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Trial 2 of this branch: cross-server delivery was restored (f7cee00), the
agent completed its onboarding and published started and completed - and
create_session still waited 180 s, called the onboarding unconfirmed, and
tried to write `error` over the completed job (refused by the registry).
delegate_wait_started required the started line to carry data.dispatch_token
(review R1: a stale attempt's started must not vouch for this one). But
publish_event only writes that field when its command line carries
--expect-dispatch-token, and create_session's hand-rolled onboarding command
never did - the wrapper's _pub_cmd does, which is why every wrapper-dispatched
agent's started has the token and no create_session onboarding ever could.
Five reviewers missed it because the fixture wrote the token into the started
line by hand: a fake producer that differs from the real one.
Fix, along the wrapper's existing mechanism rather than a new one:
* delegate_publish_cmd (lib.sh) builds the publish command for an agent, with
--expect-dispatch-token and --attempt-id, and create_session uses it. The
agent's own started now names the dispatch, and publish_event's CAS check
refuses a stale dispatch's event at the source. S7l pins the builder
argument-for-argument against the wrapper's _pub_cmd.
* delegate_started_recorded (lib.sh) is the one definition of "this
dispatch's started is on disk" - JSON-parsed, token-matched when a token is
given - used by delegate_wait_started and by the wrapper's _wait_started in
place of its regex. The wrapper keeps its status and subscriber routes
(remote agents); the shared part is the evidence contract.
* An already-terminal job is left alone when the wait expires (req. 4): no
error over a completed/error/cancelled record.
The fixture now registers the record through registry.py and publishes
started through the real publish_event.py; the only synthetic line left (S7h)
exists to pin key-order independence and says so. S7i pins the trial-2 defect
as the negative it is (old command shape -> started without token -> not
acknowledged); S7j runs the builder's command through the real producer and
is acknowledged on the first probe; S7n shows the wrapper accepting the same
real line and rejecting a stale token; S7m pins no-write-over-terminal. S7f
moves from "publish error" to "leave the terminal job alone", per req. 4.
Option B (verify data.hmac_sig with the job's auth_token, plus record-token
still equal) was weighed and not taken: verify_hmac returns True with no
auth_token - PoC mode, which the trial and every sandbox run in - so it proves
nothing there; in production hmac_sig authenticates the signer but does not
identify the dispatch, so R1 would rest on a current-state check that a stale
line plus an unrotated token passes; and it would diverge from the wrapper
unless the wrapper changed too. Token exposure is not new: every wrapper-
dispatched instruction already carries the dispatch token in the pane, and it
is a per-dispatch CAS nonce, not the HMAC secret.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Issue #25: a promoted plan is reused only with --plan-file; the implicit
newest-plan load is gone in both runners, and an empty planner report no
longer passes silently. VERSIONS.md carries the entry; the four merge-side
review reports and their identity sidecars (base 6fc9194 -> head 1e82651,
evidence review-tree) are archived under .agents/reports/<session>/.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
Issue #21 phase-1 prerequisites (PR #22) add loop runner options and the
control-plane module: MINOR. VERSIONS.md gains the v4.6.0 entry, the
archived design reviews lose their markdown trailing whitespace so
`git diff --check` is clean, and the three merge-side PASS reports
(claude, agy, codex @ 4a81d15) are archived under .agents/reports/.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
The injected prompt, the start/success/error commands and the brief's Output
Report Path used cwd-relative `.mam/jobs/...`, and the wrapper ignored
MAM_JOBS_DIR. A creator whose cwd is a git worktree could not run them as
written, and its audit log fell back to <cwd>/.mam/delegate_job_logs.
- REGISTRY_DIR defaults to ${MAM_JOBS_DIR:-.mam/jobs} and is made absolute
right after `cd "$WORKDIR"` (a relative value still resolves against WORKDIR)
- the publish command is shell-quoted (printf %q) and prefixed with
DELEGATE_JOB_LOGS_DIR=<abs>, resolved by the same get_logs_dir() the
wrapper's subscriber uses; the heartbeat timer inherits it
- direct and loop jobs share one _instructions helper; brief and report paths
are $REGISTRY_DIR/$JOB_ID/... (absolute)
- create --onboard: absolute publish_event.py, --registry-dir and log root
Tests T1-1..T1-5 (a space in the control root, commands run from another cwd
with no workspace env).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017R3Eg7MEpSWnm1oB9ZPP6G
The subscriber is started before the injection, and the injection itself
(idle handshake, quiescence, delivery grace) can take tens of seconds, so
a first-event budget of T + K x N measured from the subscription could
expire before `started` was even possible. The wrapper now passes
--ack-timeout <wall clock>: it judges delivery with _wait_started (T from
the injection) and reaps the subscriber on failure; K x N applies from the
first event on. A standalone subscriber keeps T + K x N. Documented in
MESSAGING.md 4.3.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
In the v4.5.0 review round grok read the brief and published `started`
70 s after the injection; with T = 60 the wrapper had already closed the
job as agent_blocked and refused its later PASS. T bounds delivery, not
work, so a generous default costs nothing when the agent answers fast.
Code, docs and the CLI snapshot updated together.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
Review round 1 (claude NOT PASS) corrections:
- B2: the timer now watches a process - `--watch-pid`, or by default the
nearest agent-CLI ancestor of the publishing command - and stops within
one period once it is gone, so a crashed agent stops heartbeating and is
closed after K x N; a remote runner watches itself. Documented as process
liveness (an alive-but-stuck agent keeps heartbeating).
- R1: `_wait_started` accepts an events.log `started` only when it carries
the current dispatch token (loop jobs share one events.log).
- R2: the remote runner ignores a re-sent dispatch of an attempt that
already reached its terminal.
- R3: registry.md EOF blank line (git diff --check clean).
- R4: the ack error states the effective budget when the wall clock cut it.
- P3: stop_timer verifies the pidfile pid is this job's timer before
signalling; a retained state message keeps its own seq watermark so it
can never make the monotonic guard drop its own event.
- MULTI_AGENT_RULES.md timeout section updated for the contract.
- tests: agent-process death stops the timer, recycled-pid safety,
attempt-scoped started, re-sent dispatch; acceptance 3 now kills the
agent process instead of the timer. Help captures refreshed
(--watch-pid). The wrapper can be sourced with
MAM_DELEGATE_JOB_NO_MAIN=1 for helper-level tests.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
Contract (no new event types): heartbeat every N s (automatic timer),
unresponsive after K x N s without an event, `started` within T s of the
injection, retained per-job state, and a remote transport.
- registry: heartbeat_sec / heartbeat_misses / ack_timeout_sec / transport
fields; idle_timeout_sec derives K x N unless explicit (defaults 30/4/60
keep the historical 120 s idle window).
- heartbeat.py + publish_event --heartbeat N: detached fixed-cadence timer
publishing signed `progress` heartbeats (threaded wire publish, local
record first); stopped by the terminal event via pidfile, by re-dispatch
and by lifetime. Terminal events refuse --heartbeat.
- publish_event: retained `<prefix>/state` (status, last_event, seq, HMAC)
sent on the same connection after every event (--no-state to skip);
publisher.publish_raw takes extra messages.
- job_subscriber: subscribes the state topic, recovers terminal status from
retained state, resets idle on any accepted message, records seq_gap and
reconnected in the audit log, idle budget before the first event is
T + K x N (--ack-timeout), --remote mirrors started/terminal events into
the registry through the transition boundary.
- lib.sh send_keys_safe: an accepted-but-unconfirmed submission is confirmed
by agent_status working or the job's own `started` (_sks_job_started);
pane change and busy token are no longer delivery signals.
- wrapper: --heartbeat/--heartbeat-misses/--ack-timeout (MAM_* env),
injected start command arms the timer, brief states the contract,
_wait_started (registry / events.log / subscriber log, bounded by T and
the wall clock) replaces the 10 s inject grace, R1 grace uses only the
on-disk event age (herdr/pane logged once as a diagnostic),
remote:<name> sessions skip preflight/idle/pane and dispatch over the
broker (remote_dispatch).
- remote.py (+ scripts shim): signed retained dispatch (MAM_DISPATCH_SECRET)
and a runner that mirrors the job into its own registry with the
orchestrator's identity, publishes started --heartbeat, runs --exec and
publishes the terminal event.
- loop runner: passes MAM_HEARTBEAT_SEC / MAM_HEARTBEAT_MISSES /
MAM_ACK_TIMEOUT_S through; YAML rows with transport: remote delegate as
remote:<name>.
- docs: MESSAGING.md 2.5 contract + remote transport, delegate-job SKILL.md,
registry.md, loop SKILL.md.
- tests: tests/test_issue18_liveness_contract.py (16); issue 14 pane-signal
tests flipped to started-based; issue 6 working-agent tests acknowledge
with started; CLI snapshot/help captures refreshed.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
Review of 4176d75 (planner-reviewer-claude-01, N1): _cmd_scrub_secrets chose
records by the presence of broker.password, not by its value. The old
to_registry_block wrote "password": null even for an unauthenticated broker,
so on this repo's registry (901 of 903 records carry the null key, none a
real password) --dry-run reported "would scrub 901 record(s)" and a real run
would rewrite every one of them while holding the registry lock.
- registry.py: select on broker.get("password"); an empty key is no secret,
and _atomic_write_record drops it on the record's next write anyway
- tests: the migration test gains a legacy "password": null record that must
be neither listed nor rewritten (fails on 4176d75 with "would scrub 2")
- VERSIONS.md v4.4.4 item 2 and registry.md now describe the stored-value
rule
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HCGpJXhSZasJLU38ksDghe
PATCH release for the fix/issue-12-8-7 branch (Issue #12 broker password
kept out of job records, #8 fresh install stamps kept on update, #7
same-cwd workspaces packed). No code logic changes.
- lib.sh MAM_VERSION, lib_py/loop/eager.py FRAMEWORK_VERSION and the eight
SKILL.md frontmatters (create, stop, resume, status, monitor, delegate-job,
loop, orc-onboard) move from 4.4.3 to 4.4.4
- VERSIONS.md: current release v4.4.4 (2026-09-13 KST), the eight-row skill
matrix, and a v4.4.4 changelog section in the v4.4.3 format covering the
password-free registry block restored from MQTT_PASSWORD at connect time,
the scrub on every record write, `registry.py scrub-secrets [--dry-run]`,
why auth_token/dispatch_token stay, the update.sh stamp and install_state
merge rules, the sibling-workspace walk in new-session, and the 8 new tests
- VERSIONS.md v4.4.3 ledger correction: its new-test count predated fae70c1,
which added two purge tests; now 26 (19 + 7) with both listed
- README, README.ko, BOOTSTRAP and BOOTSTRAP.ko badges and the MESSAGING.md
"Applies to" header follow the current version; historical 4.4.3 mentions
(v4.4.3 changelog, reviewer reports) stay as history
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HCGpJXhSZasJLU38ksDghe
Issue #12 - MQTT_PASSWORD was copied into every job record, which any
sub-agent can read. BrokerConfig.to_registry_block no longer persists it;
broker_config_from_job layers the password back from the environment
(.mam.env) at connect time, so publish_event still connects from the record.
_atomic_write_record scrubs broker.password from any record it writes, and
`registry.py scrub-secrets [--dry-run]` migrates an existing registry.
auth_token/dispatch_token stay: the agent's own publish_event needs them.
Issue #8 - update.sh restored the pre-update asset_hashes.txt and version.txt
over the installer's fresh ones, so version.txt named the previous commit and
remove.sh backed up every pristine updated file as "locally modified". The
success path now keeps the installer's stamps and only merges install_state
keys the re-install did not write (gitignore_created); the failure path
still restores everything verbatim.
Issue #7 - the shim's new-session took the first same-cwd pane, always found
the full first workspace, judged it overflow and opened another workspace per
agent (4 + 1 + 1). It now walks every same-cwd workspace in pane order and
takes the first with an idle pane or room in its 2xK grid; a workspace is
created only when all are full (4 + 2).
Tests: test_issue7_workspace_packing.py (3; the mock herdr gains an opt-in
per-workspace 2x2 geometry), test_issue8_update_state.py (1, real
install+update cycle), test_issue12_registry_secrets.py (4); registry CLI
snapshot and help capture refreshed for the intentional scrub-secrets
subcommand; registry.md documents the broker block.
Refs #12#8#7
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
PATCH release for the fix/issue-10-11-13-15 branch (Issue #10 agy purge
sidecars, #11 orc_onboard message, #13 orchestrator conversation purge,
#15 workspace-anchored job registry). No code logic changes.
- lib.sh MAM_VERSION, lib_py/loop/eager.py FRAMEWORK_VERSION and the eight
SKILL.md frontmatters (create, stop, resume, status, monitor, delegate-job,
loop, orc-onboard) move from 4.4.2 to 4.4.3
- VERSIONS.md: current release v4.4.3 (2026-09-13 KST), the eight-row skill
matrix, and a v4.4.3 changelog section in the v4.4.2 format covering the
agy purge sidecars/annotation/presence lock and residual_artifacts,
`lib_py.agents purge` and `orc_onboard.sh --remove --purge-conversation`
with their exit codes, the past-tense message, the MAM_JOBS_DIR ->
WORKSPACE_ROOT -> cwd registry/log resolution and the 20 new tests
- README, README.ko, BOOTSTRAP and BOOTSTRAP.ko badges and the MESSAGING.md
"Applies to" header follow the current version; historical 4.4.2 mentions
(v4.4.2 changelog, reviewer reports) stay as history
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HCGpJXhSZasJLU38ksDghe
Issue #10 - agy purge_artifacts left resumable pieces behind: it now also
removes the SQLite sidecars (-wal/-shm/-journal), the title annotation and
the presence lock (the lock only once no live process references the
conversation), and residual_artifacts() reports anything under
~/.gemini/antigravity-cli still carrying the uuid.
Issue #13 - the orchestrator has no herdr_sessions row, so its conversation
could not be purged by any skill. `lib_py.agents purge <agent|auto> <uuid>`
resolves the adapter from what is on disk (orchestrator_uuids stores bare
uuids), refuses while a live process names the uuid (conversation_in_use:
process list minus the caller's own lineage, plus lsof), and prints what it
purged. `orc_onboard.sh --remove <uuid> --purge-conversation --yes [--agent]`
runs it before the registry update so a refusal changes nothing (exit 4
without --yes, 5 while in use).
Issue #11 - "Successfully removeed": conjugate add/remove explicitly.
Issue #15 - the job registry and audit log followed the caller's cwd, so the
pytest suite (create_session.sh with its default onboarding job, run from the
repo root) had left 593 test records in the real .mam/jobs. Defaults now
resolve MAM_JOBS_DIR / DELEGATE_JOB_LOGS_DIR, then $WORKSPACE_ROOT/.mam/...,
then the cwd (mqtt_common.default_registry_dir / get_logs_dir; registry.py,
publish_event.py, service.py, create_session.sh, stop_session.sh); the
sandbox fixture pins both. The tier suites that used to leak now leave the
real registry untouched (881 records before and after).
Tests: test_issue10_13_purge.py (13), test_issue15_registry_isolation.py (7);
T-O13 pins the corrected message; the R1 helper test and B-9 logs-dir test
now assert the workspace-anchored contract; publish_event snapshot updated.
Refs #10#11#13#15
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
Round-2 review of 952657e (reviewer-grok-01): R1's disk-heartbeat
evidence comes from events.log, but publish_event appended to it only
after with_retry(_publish_once) returned - after the broker ACK or all of
its retries. With the public broker at ~5 s per publish, an agent's
heartbeat reached the disk after the grace window and the live job was
closed as deadline_timeout (test_wrapper_idle_timeout_grace_accepts_disk_heartbeat
failed 2/2 on 952657e here, as it did for the reviewer).
- publish_event: new record_local_event appends the prepared payload to
events.log right after prepare_event (validation and seq allocation
first) and before the wire publish; record_outcome takes
local_event_recorded=True so the line is not written twice, while the
audit "published" entry and the status sync stay after the publish
- DelegationService.publish_event: the same local-first order
- wrapper comment and VERSIONS.md v4.4.2 item 6 now say the heartbeat is
on disk before the publish; new-test count 21
- tests: the heartbeat is already in events.log when the publish starts
(and is recorded once), and a publish failing on every retry leaves one
local record with published=False in the audit; both fail on 952657e.
The wrapper disk-heartbeat test keeps the real publish_event path and
now passes 3/3 at the same broker latency
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HCGpJXhSZasJLU38ksDghe
Round-2 review of d5a191c (planner-reviewer-claude-01, planner-reviewer-codex-01).
- delegate wrapper: the R1 idle grace and the broker_unavailable re-check
counted only their sleeps against TIMEOUT, so python/herdr probe time plus
fresh disk heartbeats let a job run far past --timeout (55 s for a 20 s
budget in the new test). Both loops now end at START_TIME + TIMEOUT read
off the clock on every pass; heartbeats extend only the idle window, and
the herdr probe is cut off at the deadline (_within_deadline)
- document the window policy d5a191c actually implements: an agent with no
evidence is closed ~2 x IDLE_TIMEOUT after its last wire event (Issue #6
promised IDLE_TIMEOUT + 5 s) and heartbeats may be ~2 x IDLE_TIMEOUT apart;
correct the claim that publish_event records a heartbeat on disk before
the wire publish - it records right after the publish attempt
- shim ls: every failed `agent list` other than a structured
server_not_running now exits nonzero with a "state unknown" diagnostic
instead of reaching the list parser, where an error document on stdout
read as an empty list and let drift A terminate live rows
- tests: stdout variants for down and unrelated errors, reconcile keeping
herdr_confirmed False on an unrelated failure, and a wall-clock upper bound
under continuous heartbeats and slow probes; the three new checks fail on
d5a191c
- VERSIONS.md v4.4.2: items 3 and 6 and the test count (19)
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HCGpJXhSZasJLU38ksDghe
The R1 grace after a subscriber idle timeout only continued while herdr
reported `working` or the pane showed a busy token, and gave up the moment
that evidence vanished. Two live jobs in the v4.4.2 review loop were closed
as deadline_timeout that way: a grok review whose heartbeat gap was 147 s,
and the creator running the full suite in a background shell (idle to herdr,
no busy token) while publishing progress every 45 s. Both later published
`completed`, which the terminal record refused.
- `_recent_disk_heartbeat`: publish_event appends every event to the job's
events.log before the wire publish, so a heartbeat the subscriber never saw
is still on disk; treat one at most IDLE_TIMEOUT old as liveness
- R1 loop: any evidence (herdr working, pane busy token, disk heartbeat)
opens a new IDLE_TIMEOUT window; a job with no evidence for a whole window
is closed as before, still inside the wall clock
- test: a human-agent job whose subscriber idles out stays open across five
disk heartbeats and picks up the out-of-band completion (rc 0)
- VERSIONS.md v4.4.2 item 6 and test count updated
Refs #14
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
The ls translation searched stderr and the whole stdout for the substring
server_not_running, so a successful `agent list` that merely carried the text
(an agent named server_not_running-codex, or metadata) made the shim report
"no server running". reconcile.sh reads that as a confirmed-empty server, and
drift A would then mark every live row on it terminated (codex review of
0918917).
- keep the RPC exit code and translate only a failed call whose error
envelope (the whole stream or any line of it) has
error.code == "server_not_running"; every other failure keeps the unknown
path (nonzero exit, herdr_confirmed False) as before
- tests: the real herdr 0.9 envelope, a successful list carrying the
substring in a name and in metadata, other failures staying unknown, and
reconcile keeping both live rows with no drift A; the new cases fail on
0918917
- delegate wrapper: log a non-pending status seen during the inject ack
grace as "registry advanced" instead of attributing it to the agent
- VERSIONS.md v4.4.2: describe the structured check and count 16 new tests
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HCGpJXhSZasJLU38ksDghe
PATCH release for the fix/issue-9-14 branch (Issue #9 herdr server env
isolation, Issue #14 accepted-prompt delivery confirmation). No code logic
changes.
- lib.sh MAM_VERSION, lib_py/loop/eager.py FRAMEWORK_VERSION and the eight
SKILL.md frontmatters (create, stop, resume, status, monitor, delegate-job,
loop, orc-onboard) move from 4.4.1 to 4.4.2
- VERSIONS.md: current release v4.4.2 (2026-09-12 KST), the eight-row skill
matrix, and a v4.4.2 changelog section in the v4.4.1 format covering the
server env marker denylist, MAM_NO_AUTOSTART, the shim ls "no server
running" translation, SKS_DELIVERY_GRACE_S, MAM_INJECT_ACK_GRACE_S, the R1
pane busy-token grace, the T1-7 trigger change and the 13 new tests
- README, README.ko, BOOTSTRAP and BOOTSTRAP.ko badges and the MESSAGING.md
"Applies to" header follow the current version; historical 4.4.1 mentions
(v4.4.1 changelog, IMPROVEMENTS.md ledger, docs/contracts/issue-6, reviewer
reports) stay as history
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01HCGpJXhSZasJLU38ksDghe
The R1 grace after a subscriber idle timeout only asked herdr whether the
agent was still `working`. herdr never reports that for codex, so a codex
review longer than the idle timeout was closed as deadline_timeout mid-turn
and its eventual `completed` refused - the same status blindness behind the
agent_blocked misjudgement. `_agent_still_working` now also accepts the pane's
own busy token, the evidence send_keys_safe already uses to confirm delivery.
Adds a wrapper test where a codex-like agent (idle per herdr, busy per pane)
idles out the subscriber and completes out of band afterwards.
Refs #14
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
- Drop per-session agent runtime markers (CLAUDECODE, CLAUDE_CODE_CHILD_SESSION,
CLAUDE_CODE_SESSION_ID, Claude Code messaging/bridge/pid markers, and the
agy/hermes/grok/opencode session ids) from the environment lib.sh's shim hands
to a bootstrapped `herdr --session <name> server`; user configuration such as
CLAUDE_PROJECT_DIR still inherits
- Add MAM_NO_AUTOSTART=1 so query-only callers never bootstrap a server as a side
effect; status.sh and reconcile.sh export it, and the shim's `ls` reports a down
server as "no server running" so reconcile keeps herdr_confirmed instead of
freezing drift detection for every server
- send_keys_safe: treat `timeout` / `agent_prompt_stalled` from `agent prompt
--wait` as an accepted submission and confirm delivery for SKS_DELIVERY_GRACE_S
(default 15 s) via agent_status working, a busy token the pane did not show
before, or the pane moving on from its pre-submit snapshot; no signal at all
still fails closed (rc 4). Codex under herdr 0.9.0 never reports `working`, so
every delivered prompt was closed as agent_blocked and its later
started/completed events refused
- delegate wrapper: before closing an unconfirmed inject as agent_blocked, honour
the registry's own acknowledgement for MAM_INJECT_ACK_GRACE_S (default 10 s)
- Switch T1-7's failure trigger to transport_error (timeout now means accepted);
add test_issue9_server_env.py (5 cases) and test_issue14_delivery_ack.py (7)
- Verified live: a freshly bootstrapped multi-agent-mux server carries no markers
and the claude panes save transcripts again; codex job d04ebcb5 had its
started/completed accepted with no agent_blocked
Closes#9Closes#14
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VCgu9Fa9iT7TbJUSzdq4Ff
GodopuandClaude Opus 5.5