Tasks: the thread is the work item's

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 state — channels/state.py: conversations map, bind(…, origin, permalink) writing both maps, thread_for reading it first with the legacy scan + backfill, conversation(), eviction dropping the conversation, ChannelState.locked(path) over <path>.lock via runlock.fcntl (unlocked with one debug line when absent)
  • Depends on: none
  • Requirements: R1.4, R3.1, R3.4, R3.5
  • Test: T1 — test_channels.py::test_bind_records_the_conversation_per_work_item, ::test_a_pre_issue_312_state_file_backfills_its_conversations, ::test_eviction_drops_the_conversation_with_the_thread, ::test_locked_sections_on_one_path_serialize; T8 — A5 ::test_without_flock_the_lock_degrades_to_today
• [x] 2. The root and the reply — channels/slack.py: render_root, _open_thread (root → permalink → bind → save → channel.thread_opened), post replying outside the lock, bind/advance/advance_kickoff/kickoff baseline under the lock; eventlog.py gains channel.thread_opened; inbound.process_kickoff binds with origin="kickoff"
  • Depends on: 1
  • Requirements: R1.1–R1.3, R1.5, R1.6, R2.1–R2.3, R3.2
  • Test: T1 — test_channels.py::test_the_first_post_opens_a_root_and_replies_into_it, ::test_the_second_post_is_one_reply, ::test_a_failed_reply_opens_no_second_thread, ::test_thread_opened_is_emitted_with_ids_only, test_eventlog.py catalog; T8 — A1 ::test_a_members_root_shaped_message_binds_nothing, A2 ::test_the_root_is_built_from_the_ref_alone, A3 ::test_a_failed_permalink_still_binds_the_thread, A4 ::test_a_corrupt_state_file_opens_a_fresh_thread
• [x] 3. The command — commands/channels_cmd.py: channels threads [--work-item] [--json]; status counting work items
  • Depends on: 1
  • Requirements: R3.3
  • Test: T1 — test_channels.py::test_channels_threads_lists_and_filters_conversations, ::test_channels_status_counts_work_items
• [x] 4. Scenarios and re-pointed tests — the five Gherkin scenarios; existing scenarios that indexed posted[0] as the event re-pointed at the reply; the standing-session scenarios re-pointed (announcement is the first reply)
  • Depends on: 2, 3
  • Requirements: R1.1–R1.6, R2.1, R3.1, R3.3, R3.4
  • Test: T2 — test_channels_integration.py, test_bus_integration.py, test_standing_channels_integration.py; T10 — the legacy-file case
• [x] 5. Docs, capability doc, decision — docs/cli/commands/channels.md, docs/cli/state.md, docs/config/cli/channels-options.md, docs/capabilities/channels.md, skills/the-loop/reference/collaboration.md, decision-105 + index row
  • Depends on: 2, 3
  • Requirements: R3.3 (documented), the capability-docs gate
  • Test: T12 — make check (markdownlint included)
• [x] 6. Verification — execute testing-plan.md, record evidence/verification.md and evidence/security-review.md
  • Depends on: 4, 5
  • Requirements: all
  • Test: T1, T2, T8, T10, T12, T13

Dependency graph (DAG)

flowchart LR
  T1["1 state + lock"] --> T2["2 root + reply"]
  T1 --> T3["3 channels threads"]
  T2 --> T4["4 scenarios"]
  T3 --> T4
  T2 --> T5["5 docs + decision"]
  T3 --> T5
  T4 --> T6["6 verification"]
  T5 --> T6

Checkpoints

After task 1 and after task 2: test_channels.py red → green recorded in evidence/verification.md. After task 4: the three integration files green. After task 5: make check. Then the verification node, then 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.