Дмитрий
832fadbcc3
7 секций фиксов A-G закрывают 16 из 18 новых holes (2 → §9 acceptable residual): - A §3.1 protected paths +7 (registry, helpers, prompt-prefix, gate-config, package.json, composer.json) — закрывает C1-C5 - B §3 chain-state: user-content-only substring (H1), keyword list trimmed (S7), PostToolUse chain_step++ (S3), transcript SoT clarified (S2) - C §5.1 Bash: tokenizer +& (H3), session-scoped file-watcher (C6), git --output blacklist (C8), path-deny на все read commands (C9/S4), node -r blacklist (C10). §5.2 glob-aware (C11) + 1-level imports (E1) - D §4.5 default-CLOSE: unmatched answer → gate remains locked (H4) - E §3.2.0 smoke-test env propagation pre-impl (H2) + §3.2 path hardening derive from session-id (C7) + §3.4 BLOCKED-protocol enforced parent-side (S5) - F §4 Поведение 1 case-insensitive + morphology (E3) + stale-registry fall-through (S1). §4.7 length-ratio 4× (E2) - G §9 open questions: S6 (2-AskUser limit) UX-tradeoff acceptable, S8 (BRANCH-SWITCH controller-writable) → follow-up эпик Implementation budget: 13.5-20h → 18-27h (+5-7h за smoke-test, PostToolUse migration, Bash hardening, path-args overlay). Audit methodology: audit-context-building skill + ручной adversarial разбор по 13 attack-зонам. Brainstorming через superpowers:brainstorming для дизайна правок (scope=all, H4=default-CLOSE, H2=smoke-test через AskUserQuestion). cspell-words.txt sync: +4 валидных терминов (уйте/инкрементирован/матчащий/неверифицирована). Verify-sentinel: 1179/1179 vitest tools-only GREEN. 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.