Дмитрий
659f2b0757
Brain-retro #5 за период 2026-05-24T13:18Z .. 2026-05-26T05:09Z (202 эпизода).
Первый ненулевой reviewer-pass в истории brain-governance (раньше 0/414).
Key findings:
• 184 episodes reviewed via Opus 4.7 ProxyAPI, 18 errors (~$9 cost)
• outcome_reviewed: success 24.5% / soft_success 64.1% / rework 11.4%
• node_quality: correct 30% / disputable 59% / wrong_node 9% / over+under 1.6%
• 93.5% no_self_assessment — confirms self-assessment bug fixed in 752d80af
• Top ignored nodes (wrong_node): #19 Superpowers (5), #18 Pest (3),
#33 claude-md-management (2), #25 Semgrep (2)
• Discipline regressed in long session: regulated 19% → 4.5%
Artifacts:
• tools/brain-retro-batch-reviewer.mjs (new) — direct API batch driver
for retros >50 episodes (canonical Task() spawn impractical at scale).
• docs/observer/notes/2026-05-26-brain-retro.md (new) — full retro note
with 4 candidates A/B/C/D for owner review.
• docs/observer/sanity-checks/2026-05-26.json (new) — sanity Q&A.
• docs/observer/episodes-2026-05.jsonl — 184 episodes mutated with
review.* / outcome_reviewed / outcome_reviewed_source fields.
• docs/observer/STATUS.md — refreshed.
• docs/observer/.pii-counters.json / .read-counter.json / .self-retrospect-counter.json
— bumped by procedure.
Spec: brain-retro skill .claude/skills/brain-retro/SKILL.md.
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.