Дмитрий
49aa4ba725
v2.0 → v2.1 — 3 группы изменений (16 пунктов суммарно): Группа 1 — решения принятые после v2.0, не внесённые: - 1.1 Памятка classifier (4 паттерна: brainstorming / discovery-interview / writing-plans / systematic-debugging). +flag prompt-enrichment-mode. - 1.2 Reviewer как полноценный Claude Code subagent (tools=[Read,Grep,Glob, Skill], model=opus). Новый файл .claude/agents/reviewer-agent.md. +стоимость $240-1200/мес vs $40-80 direct API. Crash fallback на direct API. Context bloat cap 10 соседних эпизодов. - 1.3 Inheritance + 3 группы коротких prompt'ов (continuation/acknowledgment/ cancellation) + 30-минутный таймаут. +flag inheritance-mode. Новые поля в schema v4.1: inherited_from_task_id, inheritance_age_minutes, previous_direction_rejected, previous_task_id_rejected. Группа 2 — edge cases: - 2.1 Reviewer model явно opus в agent file. - 2.2 Reviewer subagent crash → fallback direct API call. - 2.3 Reviewer context bloat: max 10 episodes в agent system prompt. - 2.4 Manual override приоритет №1 в prefilter (раньше inheritance). - 2.5 Cancellation clears state + previous_task_id_rejected marker. Группа 3 — мелкие упущения: - 3.1 brain-retro SKILL.md description: раз в 1-2 недели (не sprint). - 3.2 recommended_chain_id nullable для custom chains. - 3.3 Embedding только для non-prefilter эпизодов. - 3.4 PII filter wraps sanity-check comments. - 3.5 requested_node fuzzy matching fallback. - 3.6 Anchor word list inline initial. - 3.7 Self-retrospect counter init в фазе 3 step 3.3. - 3.8 Sanity-check answer file schema_version=1. Cost rewrite: 720-1380 USD (v2.0) -> 1940-8200 USD (v2.1) на 6 месяцев из-за reviewer subagent. Granular rollback через reviewer-mode=direct-api возвращает к v2.0 ценам. §21 новый — changelog v2.0 → v2.1 со всеми 16 пунктами и где правка. 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.