ДОМЕНЫ - ВОПРОСЫ - ОРРКЕСТРАЦИЯ - БАЗА -Зафиксировать в orchestration loop правило domain enablement для новых неразмеченных доменов

.codex/skills/domain-case-loop/SKILL.md

@@ -29,6 +29,8 @@ Read `references/repo_runtime_map.md` before the first real cycle.
 Use these repo-native capture paths:
 - automated capture: `python scripts/domain_case_loop.py run-case ...`
 - import existing technical export: `python scripts/domain_case_loop.py import-export ...`
+- `run-case` defaults to the repo's live local profile: `local / qwen2.5-14b-instruct-1m / http://127.0.0.1:1234/v1`
+- override with `--llm-provider`, `--llm-model`, `--llm-base-url`, `--llm-api-key` when needed
 
 ## Workflow
 
@@ -69,6 +71,11 @@ Spawn `domain_analyst` and provide:
 
 Require a full verdict using `references/verdict_template.md`.
 
+The verdict must explicitly say whether the case is:
+- an existing in-contour regression;
+- a missing route/intent/capability inside project scope;
+- a true out-of-scope request.
+
 ### Step 4 - Domain patch
 
 Spawn `domain_coder` with:
@@ -80,6 +87,7 @@ Require:
 - a minimal patch
 - zero architecture drift
 - rerun after changes
+- if the domain is in project scope but outside the current contour, convert the verdict into capability enablement work instead of closing the case as unsupported
 
 ### Step 5 - Rerun
 
@@ -104,6 +112,8 @@ Write `final_status.md` with one of:
 - blocked
 - needs_exact_capability
 
+`needs_exact_capability` is the default status when the business/domain request is valid for the project, but the current contour is missing the route, intent, capability, or domain bootstrap needed to answer it.
+
 Accepted requires:
 - quality score >= 80
 - no unresolved P0 defects
@@ -114,6 +124,7 @@ Accepted requires:
 - Do not count heuristic candidates as confirmed business answers.
 - If exact data should exist in 1C/MCP, prefer exact route work over prompt cosmetics.
 - If exact data does not exist yet in the reachable contour, return a technical insufficiency with a crisp blocker.
+- If the user case belongs to a project-relevant domain but is outside the current contour, do not treat that as a terminal rejection. Treat it as domain enablement work and record the missing route/intent/capability explicitly.
 - Never fabricate 1C data.
 - Keep domain fixes minimal and localized.
 - Preserve successful baseline scenarios.
@@ -127,6 +138,7 @@ For this repository:
 - analyst output must be detailed and business-readable;
 - answers should be suitable for product hardening, not just debugging notes;
 - machine-readable turn artifacts are first-class inputs for analysis.
+- New user domains may be unmarked in the current repo. Missing markup is expected and should be handled as enablement, not as a reason to stop the loop.
 
 ## Recommended artifact set
 

.codex/skills/domain-case-loop/references/case_brief_template.md

@@ -15,6 +15,12 @@
 ## Expected result mode
 - confirmed_balance / confirmed_tax_liability / partial / technical_insufficiency / other
 
+## Contour status
+- in_contour / outside_current_contour / unknown
+
+## Enablement hypothesis
+- missing_route / missing_intent / missing_capability / missing_data_access / unknown
+
 ## Constraints
 - no architecture changes
 - 1C/MCP first

.codex/skills/domain-case-loop/references/domain_constraints.md

@@ -6,4 +6,6 @@
 - Не считать heuristic ответ продуктовым успехом.
 - Математика вне 1С допустима только как детерминированный постпроцесс над уже подтвержденными фактами.
 - Analyst read-only, Coder implementation-focused.
+- Если домен или маршрут еще не прокинут, но кейс лежит внутри целевого project scope, это считается работой на enablement, а не поводом закрыть кейс как unsupported.
+- Для новых неразмеченных доменов outer loop должен явно фиксировать, чего не хватает: intent, route, capability, data access или bootstrap слоя.
 - Accepted требует score >= 80, zero unresolved P0 и отсутствия silent fallback masking.

AGENTS.md

@@ -18,4 +18,6 @@ Rules:
 - Use `.codex/skills/domain-case-loop` for repeatable domain hardening loops on one concrete case.
 - Preserve current architecture: domain loop may automate capture, review, rerun, and artifact storage, but must not rewrite runtime foundations.
 - Prefer machine-readable case artifacts in `artifacts/domain_runs/<case_id>/`, especially `baseline_turn.json` / `rerun_turn.json`, over ad hoc prose-only summaries.
+- If a case falls outside the current routed contour because the route/intent/capability is not wired yet, treat it as domain enablement work for this project, not as automatic out-of-scope rejection.
+- For new unmarked domains, `needs_exact_capability` means "bootstrap or extend the contour" rather than "close the case as unsupported".
 - A case can be marked `accepted` only when analyst verdict is at least `80/100`, no unresolved `P0` remains, and the rerun does not mask heuristic output as confirmed.

docs/orchestration/domain_case_loop_repo_adapter.md

@@ -46,6 +46,9 @@
 
 Использует живой backend:
 
+- по умолчанию helper запускает кейс через `local / qwen2.5-14b-instruct-1m / http://127.0.0.1:1234/v1`;
+- для переопределения доступны `--llm-provider`, `--llm-model`, `--llm-base-url`, `--llm-api-key`.
+
 ```powershell
 python scripts/domain_case_loop.py run-case `
   --domain open_contracts `
@@ -103,3 +106,23 @@ python scripts/domain_case_loop.py import-export `
 - `partial`
 - `blocked`
 - `needs_exact_capability`
+
+`needs_exact_capability` здесь означает не "закрыть кейс как чужой", а "доменный запрос валиден для проекта, но текущий контур еще не умеет его отрабатывать точно".
+
+## Политика для новых и неразмеченных доменов
+
+Outer loop должен считать нормальным, что пользователь будет приносить кейсы из доменов, которые еще не размечены в текущем runtime.
+
+Если baseline показывает, что:
+
+1. вопрос лежит внутри целевого project scope;
+2. 1С/MCP/данные проекта концептуально относятся к этому домену;
+3. ответ не получен из-за отсутствующего intent/route/capability/bootstrap;
+
+то кейс нельзя автоматически считать `out_of_scope`.
+
+Такой кейс должен переходить в режим domain enablement:
+
+- явно зафиксировать, чего не хватает: `intent`, `route`, `capability`, `data access`, `contour bootstrap`;
+- сформировать минимальную задачу на прокидывание контура;
+- после этого повторно прогонять baseline/rerun уже как продуктовый кейс.

llm_normalizer/data/eval_cases/assistant_autogen_runtime_job-Bv_o7Vircc.json

@@ -0,0 +1,22 @@
+{
+  "suite_id": "assistant_autogen_runtime_job-Bv_o7Vircc",
+  "suite_version": "0.1.0",
+  "schema_version": "assistant_autogen_runtime_v0_1",
+  "scenario_count": 1,
+  "case_ids": [
+    "AUTO-001"
+  ],
+  "cases": [
+    {
+      "case_id": "AUTO-001",
+      "scenario_tag": "autogen_runtime",
+      "question_type": "direct",
+      "broadness_level": "medium",
+      "turns": [
+        {
+          "user_message": "Какие товары сейчас лежат на складе"
+        }
+      ]
+    }
+  ]
+}

llm_normalizer/data/eval_cases/assistant_autogen_runtime_job-vHloW0L_3W.json

@@ -0,0 +1,22 @@
+{
+  "suite_id": "assistant_autogen_runtime_job-vHloW0L_3W",
+  "suite_version": "0.1.0",
+  "schema_version": "assistant_autogen_runtime_v0_1",
+  "scenario_count": 1,
+  "case_ids": [
+    "AUTO-001"
+  ],
+  "cases": [
+    {
+      "case_id": "AUTO-001",
+      "scenario_tag": "autogen_runtime",
+      "question_type": "direct",
+      "broadness_level": "medium",
+      "turns": [
+        {
+          "user_message": "Какие товары лежат на складе сейчас или на любую выбранную дату"
+        }
+      ]
+    }
+  ]
+}

scripts/domain_case_loop.py

@@ -18,6 +18,13 @@ DEFAULT_ARTIFACTS_ROOT = REPO_ROOT / "artifacts" / "domain_runs"
 DEFAULT_SESSIONS_DIR = REPO_ROOT / "llm_normalizer" / "data" / "assistant_sessions"
 DEFAULT_REPORTS_DIR = REPO_ROOT / "llm_normalizer" / "reports"
 DEFAULT_BACKEND_URL = "http://127.0.0.1:8787"
+DEFAULT_PROMPT_VERSION = "address_query_runtime_v1"
+DEFAULT_LLM_PROVIDER = "local"
+DEFAULT_LLM_MODEL = "qwen2.5-14b-instruct-1m"
+DEFAULT_LLM_BASE_URL = "http://127.0.0.1:1234/v1"
+DEFAULT_LLM_API_KEY = ""
+DEFAULT_TEMPERATURE = 0.0
+DEFAULT_MAX_OUTPUT_TOKENS = 900
 TECH_SECTION_HEADER = "### technical_debug_payload_json"
 
 
@@ -444,6 +451,15 @@ def handle_run_case(args: argparse.Namespace) -> int:
     )
 
     payload: dict[str, Any] = {
+        "normalizeConfig": {
+            "llmProvider": args.llm_provider,
+            "apiKey": args.llm_api_key,
+            "model": args.llm_model,
+            "baseUrl": args.llm_base_url,
+            "temperature": args.temperature,
+            "maxOutputTokens": args.max_output_tokens,
+            "promptVersion": args.prompt_version,
+        },
         "eval_target": "assistant_stage1",
         "questions": [args.question],
         "useMock": bool(args.use_mock),
@@ -564,6 +580,13 @@ def build_parser() -> argparse.ArgumentParser:
     run_case.add_argument("--output-root", default=str(DEFAULT_ARTIFACTS_ROOT))
     run_case.add_argument("--sessions-dir", default=str(DEFAULT_SESSIONS_DIR))
     run_case.add_argument("--reports-dir", default=str(DEFAULT_REPORTS_DIR))
+    run_case.add_argument("--prompt-version", default=DEFAULT_PROMPT_VERSION)
+    run_case.add_argument("--llm-provider", default=DEFAULT_LLM_PROVIDER, choices=["openai", "local"])
+    run_case.add_argument("--llm-model", default=DEFAULT_LLM_MODEL)
+    run_case.add_argument("--llm-base-url", default=DEFAULT_LLM_BASE_URL)
+    run_case.add_argument("--llm-api-key", default=DEFAULT_LLM_API_KEY)
+    run_case.add_argument("--temperature", type=float, default=DEFAULT_TEMPERATURE)
+    run_case.add_argument("--max-output-tokens", type=int, default=DEFAULT_MAX_OUTPUT_TOKENS)
     run_case.add_argument("--timeout-seconds", type=int, default=300)
     run_case.add_argument("--poll-interval-seconds", type=float, default=1.5)
     run_case.add_argument("--expected-capability")