Verification — issue-329

The testing plan executed (testing-plan.md, rows T1, T2, T3, T9, T11, T13). Commands run from the repository root at the head of claude/github-issue-329-fide3i. Every login, ref and path in the tests is a fixture (octocat, stranger, octo/repo, loop-lab, a pytest tmp path). Nothing here needed redaction.

Red → green

The new tests were written first. With the source files stashed back to cab21a6 (13.6.0) and only the tests in place, the twenty new service tests fail and ten pre-existing ones in the same selection still pass:

text

git stash push -- cli/the_loop
uv run --project cli python -m pytest -q cli/tests/test_poller.py cli/tests/test_poller_integration.py \
  cli/tests/test_webhook_routing_integration.py cli/tests/test_routing.py \
  -k "ended or reconciled or closer or stamp or untracked or tracked or reopen"
FAILED cli/tests/test_poller.py::test_provider_closure_carries_the_closer
FAILED cli/tests/test_poller.py::test_provider_closure_event_names_the_closer_as_sender
FAILED cli/tests/test_poller.py::test_a_closed_session_is_asked_once_and_not_again_once_stamped
FAILED cli/tests/test_poller.py::test_a_paused_sessions_closed_item_is_reconciled
FAILED cli/tests/test_poller.py::test_a_record_without_a_session_is_reconciled[control]
FAILED cli/tests/test_poller.py::test_a_record_without_a_session_is_reconciled[graph]
FAILED cli/tests/test_poller.py::test_a_record_without_a_session_is_reconciled[collaborators]
FAILED cli/tests/test_poller.py::test_a_listed_item_clears_its_ended_stamp
FAILED cli/tests/test_poller_integration.py::test_a_paused_sessions_closed_item_is_detected_and_stamped
FAILED cli/tests/test_poller_integration.py::test_a_polled_closure_by_an_authorized_closer_releases_the_item
FAILED cli/tests/test_poller_integration.py::test_a_polled_closure_by_an_unlisted_closer_is_deferred
FAILED cli/tests/test_webhook_routing_integration.py::test_a_closed_issue_with_no_session_is_stamped_ended
FAILED cli/tests/test_webhook_routing_integration.py::test_reopening_an_issue_clears_the_stamp
FAILED cli/tests/test_routing.py::test_an_issue_close_stamps_the_work_item_ended
FAILED cli/tests/test_routing.py::test_a_close_with_no_session_still_stamps_a_tracked_record
FAILED cli/tests/test_routing.py::test_a_close_for_an_untracked_ref_creates_no_record
FAILED cli/tests/test_routing.py::test_a_closed_session_record_counts_as_tracked
FAILED cli/tests/test_routing.py::test_a_pr_merge_stamps_the_prs_own_record_and_not_the_linked_issue
FAILED cli/tests/test_routing.py::test_a_reopen_clears_the_ended_stamp
FAILED cli/tests/test_routing.py::test_the_stamp_is_logged_as_work_item_ended
20 failed, 10 passed, 398 deselected in 2.09s
git stash pop

The task-1 tests (test_workitem.py, test_control.py, test_reset.py) were run red the same way before the store changed — 4 failed, 1 passed, 98 deselected — and the seven dashboard tests were red against the unchanged model.ts / grouping.ts / fixture.ts (Tests 7 failed | 212 passed (219)) because the edit script that carried those three files stopped before reaching them; the tests then went green once the edits were applied.

Task	Red (before the change)	Green
1 the fact	`ENDED` absent from `SECTIONS`; no `record_ended` / `ended` / `clear_ended`; reset leaves the section; `work_item.*` unknown to the catalog	`test_a_record_with_only_ended_is_kept_and_indexed`, `test_record_ended_and_clear_ended`, `test_a_malformed_ended_reads_as_not_ended`, `test_reset_clears_the_ended_section`; `test_eventlog.py::test_every_emitted_event_type_is_documented`
2 the writer	the close path clears `control` / `collaborators` and nothing else; a session-less close returns with a debug line; `reopened` is an ordinary event	the eight `test_routing.py` tests above; the two webhook scenarios
3 more closures	only active sessions reconciled; `Closure` has no `actor`; the polled event has no `sender`	the ten `test_poller.py` tests above (one rewritten); the three poll scenarios
4 the readers	`attention` reports the stale question and arming; the join has no `ended`, the gated record lands in `needs-you`	`test_an_ended_item_asks_for_no_attention`; `model.test.ts › an ended work item (issue-329)` ×4; `grouping.test.ts › sidebarGroup for an ended item` ×3
5 docs	—	`make lint` (markdownlint over every `*.md`), `test_docs_parity.py`

Existing assertions changed: one — test_an_already_closed_session_is_not_reconciled_again pinned the rule that was the gap and is now test_a_closed_session_is_asked_once_and_not_again_once_stamped. Four existing test files gained a tmp portable directory (test_routing.py's factory, test_workspace.py's two builders, test_tmux_runner_integration.py's layout): a close now writes a record, and RoutingConfig's default portable directory is the checkout's own — found by checking the checkout for a stray .the-loop/portable/ after each full run.

Rows T1, T2, T3, T9, T11

text

== T1
uv run --project cli python -m pytest -q cli/tests/test_workitem.py cli/tests/test_portable_index.py cli/tests/test_control.py cli/tests/test_routing.py cli/tests/test_poller.py cli/tests/test_reset.py cli/tests/test_core_attention.py
496 passed in 4.06s
== T2
cd ui && bun run test
 Test Files  16 passed (16)
      Tests  219 passed (219)
== T3
uv run --project cli python -m pytest -q cli/tests/test_webhook_routing_integration.py cli/tests/test_poller_integration.py
63 passed in 21.47s
== T9
uv run --project cli python -m pytest -q cli/tests/test_routing.py cli/tests/test_poller.py cli/tests/test_core_attention.py -k "ended or reopened or closer or untracked"
11 passed, 364 deselected in 0.20s
== T11
uv run --project cli python -m pytest -q cli/tests/test_eventlog.py cli/tests/test_docs_parity.py
19 passed in 0.67s

Row T13 — `make check` and the dashboard's three commands

text

make check
uv run ruff check cli hooks
All checks passed!
npx --yes markdownlint-cli2@0.18.1 "**/*.md"
Summary: 0 error(s)
uv run ruff format --check cli hooks
281 files already formatted
uv run pyright cli
0 errors, 0 warnings, 0 informations
uv run python scripts/validate_config.py
VALID   .the-loop/harness-config.yaml
VALID   skills/the-loop/templates/harness-config.yaml
VALID   cli/the_loop/harness-config.default.yaml
VALID   .the-loop/collaborators.yaml
VALID   skills/the-loop/templates/collaborators.yaml
VALID   .the-loop/cli-config.yaml
VALID   skills/the-loop/templates/cli-config.yaml
uv run --project cli python -m pytest -q cli
3165 passed, 1 skipped in 146.75s (0:02:26)

cd ui
bun run lint     → oxlint --type-aware, clean
bun run test     → Test Files 16 passed (16) · Tests 219 passed (219)
bun run build    → tsc --noEmit && vite build · ✓ built in 1.38s

Security

See security-review.md — abuse cases A1–A6, each closed by a named test; no human sign-off at tier 3.