Дмитрий
81f52fd1c6
v2.1 → v2.2. Полный critical audit (экономия 0%) нашёл 16 пробелов;
все закрыты. Новый §22 — changelog с таблицей всех находок.
Критическое (1):
- D1: §17 formal text (§6) включал Continuation в exempt-список direct,
но Continuation наследует non-conversation classification → §17 enforce
должен применяться. §6 п.4 переписан: exempt только Acknowledgment +
Cancellation; Continuation — нет (NON_BLOCKING_TASK_TYPES не содержит её).
Внутренние несогласованности (6):
- A1: schema alternative_better типизирован только null → заметка string|null.
- A2/C3: task_cost reviewer tokens vs usd — взаимоисключающи по пути review.
- A3: §11.3 step 3.2 «создать reviewer-agent.md» → файл уже создан (49aa4ba7).
- A4: prompt-enrichment risk добавлен в §14.
- A5/B3: missed-activations.mjs читает archived map → adapter task §11.2 task 9
(переключение на nodes.yaml) + §9.5 detail + §14 risk.
- A6: §18.7 «reviewer-agent будет добавлен» → уже добавлен.
Underspecified (5):
- B1/B5: inheritance state→episode chain описан (§4.1 проверка 2).
- B2: reviewer fallback path = brain-retro-opus-reviewer.mjs сохраняется из v2.0.
- B4: discipline-metrics.mjs keep/remove → task §11.1 task 17.
Typo / ellipsis (4):
- C1: «existed» → «создана» в DoD.
- C4: «ОНLY» (кириллица) → «ONLY» в reviewer-agent.md.
- C5: anchor list финализирован (28 существительных + 10 императивов, ellipsis убран).
Сравнение аудитов: v1.0 имел 35 находок, v2.1 — 16, обе волны закрыты.
Остаточные implementation-вопросы (model ID, clustering algo) — в §15,
решаются на старте фаз.
Реализация — после закрытия Биллинга v2 Спек C.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Observer infrastructure
Passive evidence-loop for the Лидерра «brain» per ADR-011.
Files
episodes-YYYY-MM.jsonl— append-only JSONL, one line per Stop-event. Schema v2 (schema_version: 2): the 5 mandatory fields +decision_provenance(who chose the node),environment(economy_level / model / post_compaction / session_turn / parallel_session),task_size,task_ref,prompt_signal, and anoutcomethat isunknownat write time (refined by/brain-retro). On an internal hook failure a minimalobserver_errormarker line is written instead of a silent skip. Written bytools/observer-stop-hook.mjsviatools/observer-transcript-parser.mjs.notes/YYYY-MM-DD-<slug>.md— optional MD notes for sessions with qualitative history.STATUS.md— auto-generated dashboard. Regenerated per-commit bytools/status-md-generator.mjs..read-counter.json— C3 observer-of-observer counter. Updated on Read of observer files.dashboard.html+dashboard.js+dashboard-core.js— Brain Dashboard: visualises the episode log over the automation-graph topology (4 views — Карта / Разбор / Лента / Агрегат). Runnpm run brain:dashboard, open the printed localhost URL.dashboard-core.jsis pure logic, unit-tested intools/brain-dashboard-core.test.mjs.
Lifecycle
- Write: every Stop-event appends one JSONL line, parsed from the session transcript (Stop-hook).
- Aggregate:
/brain-retroskill reads JSONL each sprint, proposes regulatory candidates. - Surface:
STATUS.mdshows controllers + monthly stats. - Self-prune: C3 warns if 54 weeks pass without any read of observer files.
Routing-tag discipline
When the user dictates a specific method/node (e.g. «запусти discovery-interview»), Claude must emit one line in its response:
<!-- routing: provenance=user_directed_method node=<chosen> counterfactual=<node Claude would have chosen autonomously> -->
The Stop-hook routing-gate (tools/observer-routing-detector.mjs + routingGateDecision) detects a dictated method; if the tag is missing it returns decision: block, so the turn cannot end without the tag. The gate fires at most once per turn (stop_hook_active guard). This makes decision_provenance reliable — factor analysis can separate a router error from a user-dictated one.
Privacy
PII filter (phone numbers, emails, tokens) is applied before every write — see tools/observer-pii-filter.mjs. gitleaks pre-push also scans observer files as part of full-history sweep.
Don't
- Don't edit
episodes-*.jsonlmanually — it's append-only. - Don't write outside
docs/observer/notes/for hand-curated notes. - Don't change
.read-counter.jsonmanually — it's maintained by hooks.
HK1 pre-check (Pravila ADR-010) — verified 2026-05-19
Before registering tools/observer-stop-hook.mjs on Stop event (Task B5), verified collision against 6-component economy/skill-discipline architecture:
- User-level
~/.claude/settings.jsonalready has Stop hook: agent-type Sonnet-4.6 economy compliance verifier (analyzes transcript for claim-without-evidence violations). - Project-level
.claude/settings.json— Stop slot empty.
Result: no overwrite. observer-stop-hook will be added as command-type entry in project-level Stop array. Project + user scopes are independent slots in Claude Code 2.x — both run on the same Stop event without conflict. The agent verifier (user scope) and the JSONL appender (project scope) have non-overlapping responsibilities.