Tasks: a closed work item is recorded as ended, and the board stops asking for a human on it

The last spec artifact. A DAG derived from the design and testing plan; each task names the testing-plan row that proves it. TDD: the test first, red, then green.

Task list

• [x] 1. The fact — ENDED in workitem.py (SECTIONS), record_ended / ended / clear_ended on ControlStore, ENDED in reset.py, the two event types in eventlog.py
  • Depends on: none
  • Requirements: R1.4, R1.5, R1.6
  • Test: T1 — the store, control and reset tests; T11 — the event catalog
• [x] 2. The writer — Dispatcher._record_closure on the closed branch (matched and session-less), the reopened clearing in handle
  • Depends on: 1
  • Requirements: R1.1–R1.3, R2.1, R2.3
  • Test: T1 — the routing tests; T3 — the two webhook scenarios; T9 — A1, A5
• [x] 3. More closures — the poller's candidate set, the listed-item clearing, GhItemState.closed_by → Closure.actor → sender
  • Depends on: 1
  • Requirements: R2.2, R3.1–R3.4, R4.1–R4.3
  • Test: T1 — the poller and provider tests; T3 — the three poll scenarios; T9 — A2, A3, A4
• [x] 4. The readers — core/attention.py; types.ts, model.ts, grouping.ts, the demo fixture
  • Depends on: 1
  • Requirements: R5.1–R5.5
  • Test: T1 — the attention test; T2 — the dashboard tests; T9 — A6
• [x] 5. Docs, capability docs, decision — docs/cli/state.md, docs/capabilities/webhook-triggers.md, docs/capabilities/control-plane.md, decision-113 + index row, ui/README.md
  • Depends on: 2, 3, 4
  • Requirements: R6.1, R6.2, the capability-docs gate
  • Test: T11 — docs parity; T13 — make check
• [x] 6. Verification — execute testing-plan.md, record evidence/verification.md and evidence/security-review.md
  • Depends on: 5
  • Requirements: all
  • Test: T1, T2, T3, T9, T11, T13, T14

Dependency graph (DAG)

flowchart LR
  T1["1 the fact"] --> T2["2 the writer"]
  T1 --> T3["3 more closures"]
  T1 --> T4["4 the readers"]
  T2 --> T5["5 docs + decision"]
  T3 --> T5
  T4 --> T5
  T5 --> T6["6 verification"]

Checkpoints

After task 1: the store, control, reset and catalog tests green. After tasks 2–4: T1, T2 and T3 green. After task 5: make check and the dashboard's three commands. Then the verification node, the self-review rounds and the security review gate (evidence/security-review.md), then the PR with the reviewer briefing.

Review comments

Appended by the-loop's record-feedback hook when a human gate approves with comments (issue-109). Append-only and attributed: an approval never silently discards a reviewer's suggestions, and the feedback travels with the document it concerns rather than living in a side-channel tracker.