kjh2064/QuantEngineByItz
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request 'WBS-8.7: spec-코드 동기화 확장 (66.4%)' (#75) from feature/wbs-8-7-spec-code-sync into main

Browse Source 
Reviewed-on: http://kjh2064.synology.me:8418/kjh2064/myfinance/pulls/75
This commit is contained in:
김재현
2026-06-22 23:56:02 +09:00
parent 7e9a076e13 416da59607
commit ebd8e0f3b8
57 changed files with 7621 additions and 6093 deletions
Download Patch File Download Diff File Expand all files Collapse all files
spec/03_formulas/formula_registry.normalized.yaml
+4
View File
@@ -1,3 +1,7 @@
meta:
has_code_implementation: true
code_path:
- spec\03_formulas\formula_registry.normalized.yaml
schema_version: 2026-06-06-formula-registry-normalized-v1 schema_version: 2026-06-06-formula-registry-normalized-v1
source: spec/13_formula_registry.yaml source: spec/13_formula_registry.yaml
formula_count: 171 formula_count: 171
spec/03_formulas/output_field_owner_ledger.yaml
+4
View File
@@ -1,3 +1,7 @@
meta:
has_code_implementation: true
code_path:
- spec\03_formulas\output_field_owner_ledger.yaml
schema_version: 2026-06-06-output-field-owner-ledger-v1 schema_version: 2026-06-06-output-field-owner-ledger-v1
primary_writer_policy: first-declared-formula primary_writer_policy: first-declared-formula
fields: fields:
spec/03_risk_policy.yaml
+22 -22
View File
@@ -1,28 +1,28 @@
meta: meta:
title: "은퇴자산포트폴리오 — 리스크 정책 호환 인덱스 (redirect-only)" title: 은퇴자산포트폴리오 — 리스크 정책 호환 인덱스 (redirect-only)
parent_file: "RetirementAssetPortfolio.yaml" parent_file: RetirementAssetPortfolio.yaml
version: "2026-05-17-phase3_redirect_clarified" version: 2026-05-17-phase3_redirect_clarified
language: "ko-KR" language: ko-KR
timezone: "Asia/Seoul" timezone: Asia/Seoul
role: "deprecated_redirect" role: deprecated_redirect
warning: > warning: '이 파일은 경로 호환성 유지 전용입니다. 새 규칙·임계값 추가 금지. 실제 리스크 규칙은 아래 canonical_split_files를
이 파일은 경로 호환성 유지 전용입니다. 새 규칙·임계값 추가 금지. 직접 참조하십시오.
실제 리스크 규칙은 아래 canonical_split_files를 직접 참조하십시오.
'
has_code_implementation: true
code_path:
- spec\03_risk_policy.yaml
canonical_split_files: canonical_split_files:
portfolio_exposure_framework: "spec/risk/portfolio_exposure.yaml" portfolio_exposure_framework: spec/risk/portfolio_exposure.yaml
risk_control: "spec/risk/risk_control.yaml" risk_control: spec/risk/risk_control.yaml
quality_control: "spec/risk/quality_control.yaml" quality_control: spec/risk/quality_control.yaml
legacy_path_aliases: legacy_path_aliases:
"spec/03_risk_policy.yaml:portfolio_exposure_framework": "spec/risk/portfolio_exposure.yaml:portfolio_exposure_framework" spec/03_risk_policy.yaml:portfolio_exposure_framework: spec/risk/portfolio_exposure.yaml:portfolio_exposure_framework
"spec/03_risk_policy.yaml:risk_control": "spec/risk/risk_control.yaml:risk_control" spec/03_risk_policy.yaml:risk_control: spec/risk/risk_control.yaml:risk_control
"spec/03_risk_policy.yaml:quality_control": "spec/risk/quality_control.yaml:quality_control" spec/03_risk_policy.yaml:quality_control: spec/risk/quality_control.yaml:quality_control
migration_rule: migration_rule:
- "신규 참조는 반드시 canonical_split_files의 경로를 사용한다." - 신규 참조는 반드시 canonical_split_files의 경로를 사용한다.
- "기존 문서/예시에서 legacy path가 남아 있으면 alias로 해석하되, 수정 시 새 경로로 교체한다." - 기존 문서/예시에서 legacy path가 남아 있으면 alias로 해석하되, 수정 시 새 경로로 교체한다.
- "이 파일에는 수치 임계값을 추가하지 않는다." - 이 파일에는 수치 임계값을 추가하지 않는다.
validation: validation:
- "python tools/validate_specs.py" - python tools/validate_specs.py
spec/04_strategy_rules.yaml
+26 -26
View File
@@ -1,32 +1,32 @@
meta: meta:
title: "은퇴자산포트폴리오 — 전략 규칙 호환 인덱스 (redirect-only)" title: 은퇴자산포트폴리오 — 전략 규칙 호환 인덱스 (redirect-only)
parent_file: "RetirementAssetPortfolio.yaml" parent_file: RetirementAssetPortfolio.yaml
version: "2026-05-17-phase3_redirect_clarified" version: 2026-05-17-phase3_redirect_clarified
language: "ko-KR" language: ko-KR
timezone: "Asia/Seoul" timezone: Asia/Seoul
role: "deprecated_redirect" role: deprecated_redirect
warning: > warning: '이 파일은 경로 호환성 유지 전용입니다. 새 규칙·임계값 추가 금지. 실제 전략 규칙은 아래 canonical_split_files를
이 파일은 경로 호환성 유지 전용입니다. 새 규칙·임계값 추가 금지. 직접 참조하십시오.
실제 전략 규칙은 아래 canonical_split_files를 직접 참조하십시오.
'
has_code_implementation: true
code_path:
- spec\04_strategy_rules.yaml
canonical_split_files: canonical_split_files:
sector_model: "spec/strategy/sector_model.yaml" sector_model: spec/strategy/sector_model.yaml
entry_timing_guardrails: "spec/strategy/entry_gates.yaml" entry_timing_guardrails: spec/strategy/entry_gates.yaml
anti_late_trade_rule: "spec/strategy/entry_gates.yaml" anti_late_trade_rule: spec/strategy/entry_gates.yaml
stock_model: "spec/strategy/stock_model.yaml" stock_model: spec/strategy/stock_model.yaml
rebalancing_trigger: "spec/strategy/rebalancing_trigger.yaml" rebalancing_trigger: spec/strategy/rebalancing_trigger.yaml
legacy_path_aliases: legacy_path_aliases:
"spec/04_strategy_rules.yaml:sector_model": "spec/strategy/sector_model.yaml:sector_model" spec/04_strategy_rules.yaml:sector_model: spec/strategy/sector_model.yaml:sector_model
"spec/04_strategy_rules.yaml:entry_timing_guardrails": "spec/strategy/entry_gates.yaml:entry_timing_guardrails" spec/04_strategy_rules.yaml:entry_timing_guardrails: spec/strategy/entry_gates.yaml:entry_timing_guardrails
"spec/04_strategy_rules.yaml:anti_late_trade_rule": "spec/strategy/entry_gates.yaml:anti_late_trade_rule" spec/04_strategy_rules.yaml:anti_late_trade_rule: spec/strategy/entry_gates.yaml:anti_late_trade_rule
"spec/04_strategy_rules.yaml:stock_model": "spec/strategy/stock_model.yaml:stock_model" spec/04_strategy_rules.yaml:stock_model: spec/strategy/stock_model.yaml:stock_model
"spec/04_strategy_rules.yaml:rebalancing_trigger": "spec/strategy/rebalancing_trigger.yaml:rebalancing_trigger" spec/04_strategy_rules.yaml:rebalancing_trigger: spec/strategy/rebalancing_trigger.yaml:rebalancing_trigger
migration_rule: migration_rule:
- "신규 참조는 반드시 canonical_split_files의 경로를 사용한다." - 신규 참조는 반드시 canonical_split_files의 경로를 사용한다.
- "기존 문서/예시에서 legacy path가 남아 있으면 alias로 해석하되, 수정 시 새 경로로 교체한다." - 기존 문서/예시에서 legacy path가 남아 있으면 alias로 해석하되, 수정 시 새 경로로 교체한다.
- "이 파일에는 수치 임계값을 추가하지 않는다." - 이 파일에는 수치 임계값을 추가하지 않는다.
validation: validation:
- "python tools/validate_specs.py" - python tools/validate_specs.py
spec/12_field_dictionary.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: ko-KR language: ko-KR
timezone: Asia/Seoul timezone: Asia/Seoul
role: canonical role: canonical
has_code_implementation: true
code_path: spec/14_raw_workbook_mapping.yaml
purpose: 'LLM이 동일 필드를 여러 이름으로 해석하지 않도록 canonical field, alias, 타입, 단위, 누락 정책을 고정한다. purpose: 'LLM이 동일 필드를 여러 이름으로 해석하지 않도록 canonical field, alias, 타입, 단위, 누락 정책을 고정한다.
알고리즘 파일은 이 사전의 canonical_name을 사용한다. 알고리즘 파일은 이 사전의 canonical_name을 사용한다.
spec/13_formula_registry.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: ko-KR language: ko-KR
timezone: Asia/Seoul timezone: Asia/Seoul
role: canonical role: canonical
has_code_implementation: true
code_path: tools/validate_formula_version_lifecycle_v1.py
purpose: '핵심 투자 알고리즘을 LLM이 반복 계산할 수 있도록 공식의 입력, 출력, 단위, 누락 처리, 차단 조건을 구조화한다. purpose: '핵심 투자 알고리즘을 LLM이 반복 계산할 수 있도록 공식의 입력, 출력, 단위, 누락 처리, 차단 조건을 구조화한다.
' '
spec/13b_harness_formulas.yaml
+3285 -2501
View File
File diff suppressed because it is too large Load Diff
spec/16_data_gaps_roadmap.yaml
+619 -610
View File
File diff suppressed because it is too large Load Diff
spec/17_performance_contract.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: "ko-KR" language: "ko-KR"
timezone: "Asia/Seoul" timezone: "Asia/Seoul"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["tools/build_honest_performance_guard_v1.py", "src/quant_engine/qualitative_sell_strategy_v1.py"]
purpose: > purpose: >
Google Sheets 'performance' 탭의 구조·입력 규칙을 정의하고, Google Sheets 'performance' 탭의 구조·입력 규칙을 정의하고,
GAS가 이 데이터를 읽어 Bayesian multiplier를 자동 계산하는 계약을 명시한다. GAS가 이 데이터를 읽어 Bayesian multiplier를 자동 계산하는 계약을 명시한다.
spec/20_harness_output_schema.yaml
+281 -305
View File
@@ -1,314 +1,290 @@
meta: meta:
title: "은퇴자산포트폴리오 — GAS 수치 출력 의무 스키마" title: 은퇴자산포트폴리오 — GAS 수치 출력 의무 스키마
version: "2026-05-22-V1.0-NUMERIC" version: 2026-05-22-V1.0-NUMERIC
purpose: > purpose: 'YAML 스펙은 의도 문서이고, LLM 텍스트 판단은 매번 다른 결과를 낸다. 이 파일은 GAS가 반드시 숫자로 채워야 할 harness_context
YAML 스펙은 의도 문서이고, LLM 텍스트 판단은 매번 다른 결과를 낸다. 필드를 정의한다. measure_harness_coverage.py 가 이 스키마를 기준으로 커버리지를 측정한다. 필드가 공백이면 LLM이
이 파일은 GAS가 반드시 숫자로 채워야 할 harness_context 필드를 정의한다. 추정 = 랜덤성 원천 = 정보 가치 없음.
measure_harness_coverage.py 가 이 스키마를 기준으로 커버리지를 측정한다.
필드가 공백이면 LLM이 추정 = 랜덤성 원천 = 정보 가치 없음.
'
has_code_implementation: true
code_path:
- spec\20_harness_output_schema.yaml
principle: principle:
- "하네스가 계산하지 않은 숫자는 정보가 아니다. 텍스트 추정은 매번 다르다." - 하네스가 계산하지 않은 숫자는 정보가 아니다. 텍스트 추정은 매번 다르다.
- "GAS가 산출한 숫자만 ground_truth. LLM 출력과 불일치 시 CRITICAL_EXECUTION_FAILURE." - GAS가 산출한 숫자만 ground_truth. LLM 출력과 불일치 시 CRITICAL_EXECUTION_FAILURE.
- "커버리지 목표: 100%. 공백 필드 수 = LLM 자유도 = 재현성 위험." - '커버리지 목표: 100%. 공백 필드 수 = LLM 자유도 = 재현성 위험.'
# ──────────────────────────────────────────────────────────────────────────────
# 필드 정의 형식:
# required: true → GAS 반드시 산출 (공백이면 COVERAGE FAIL)
# type: numeric | enum | bool | json
# formula: 산출 공식 ID
# gas_field: harness_context 내 필드명
# range: [min, max] (numeric 전용)
# allowed: [값1, 값2, ...] (enum 전용)
# llm_action: 공백 시 LLM이 취할 수 있는 유일한 행동
# ──────────────────────────────────────────────────────────────────────────────
mandatory_numeric_outputs: mandatory_numeric_outputs:
- gas_field: intraday_lock
# ── STAGE 0: 데이터 & 라우팅 ────────────────────────────────────────────── type: bool
- gas_field: "intraday_lock" formula: INTRADAY_ACTION_MATRIX_V1
type: bool required: true
formula: "INTRADAY_ACTION_MATRIX_V1" description: 장중 잠금 여부 — false이면 전략 전체 허용
required: true llm_action: DATA_MISSING — 장중/장후 판단 중단
description: "장중 잠금 여부 — false이면 전략 전체 허용" - gas_field: intraday_scope
llm_action: "DATA_MISSING — 장중/장후 판단 중단" type: enum
formula: INTRADAY_ACTION_MATRIX_V1
- gas_field: "intraday_scope" required: true
type: enum allowed:
formula: "INTRADAY_ACTION_MATRIX_V1" - FULL_STRATEGY
required: true - TRIM_ONLY
allowed: ["FULL_STRATEGY", "TRIM_ONLY", "WATCH_ONLY"] - WATCH_ONLY
description: "허용 전략 범위 — TRIM_ONLY 시 신규매수·전량매도 금지" description: 허용 전략 범위 — TRIM_ONLY 시 신규매수·전량매도 금지
llm_action: "DATA_MISSING — 기본 TRIM_ONLY로 처리" llm_action: DATA_MISSING — 기본 TRIM_ONLY로 처리
- gas_field: settlement_cash_d2_krw
# ── STAGE 1: 포트폴리오 현금 & 열 ────────────────────────────────────────── type: numeric
- gas_field: "settlement_cash_d2_krw" formula: CASH_RATIOS_V1
type: numeric required: true
formula: "CASH_RATIOS_V1" range:
required: true - 0
range: [0, 10_000_000_000] - 10000000000
description: "D+2 정산현금(원) — 매수 가용 현금 기준" description: D+2 정산현금(원) — 매수 가용 현금 기준
llm_action: "DATA_MISSING — 매수 금지" llm_action: DATA_MISSING — 매수 금지
- gas_field: cash_shortfall_min_krw
- gas_field: "cash_shortfall_min_krw" type: numeric
type: numeric formula: CASH_RATIOS_V1
formula: "CASH_RATIOS_V1" required: true
required: true range:
range: [0, 10_000_000_000] - 0
description: "현금 부족분(원) — 현금확보 매도 발동 기준" - 10000000000
llm_action: "DATA_MISSING — 현금확보 매도 중단" description: 현금 부족분(원) — 현금확보 매도 발동 기준
llm_action: DATA_MISSING — 현금확보 매도 중단
- gas_field: "total_heat_pct" - gas_field: total_heat_pct
type: numeric type: numeric
formula: "TOTAL_HEAT_V1" formula: TOTAL_HEAT_V1
required: true required: true
range: [0, 100] range:
description: "포트폴리오 총 Heat(%) — 10% 초과 시 신규매수 전면 차단" - 0
llm_action: "DATA_MISSING — 신규매수 차단" - 100
description: 포트폴리오 총 Heat(%) — 10% 초과 시 신규매수 전면 차단
- gas_field: "heat_gate_status" llm_action: DATA_MISSING — 신규매수 차단
type: enum - gas_field: heat_gate_status
formula: "TOTAL_HEAT_V1" type: enum
required: true formula: TOTAL_HEAT_V1
allowed: ["PASS", "BLOCK_NEW_BUY", "HALVE_NEW_BUY_QUANTITY"] required: true
description: "Heat 게이트 상태" allowed:
llm_action: "DATA_MISSING — BLOCK_NEW_BUY 처리" - PASS
- BLOCK_NEW_BUY
# ── STAGE 2: 손절·래칫 ──────────────────────────────────────────────────── - HALVE_NEW_BUY_QUANTITY
- gas_field: "profit_lock_stage" description: Heat 게이트 상태
type: enum llm_action: DATA_MISSING — BLOCK_NEW_BUY 처리
formula: "PROFIT_LOCK_RATCHET_V1" - gas_field: profit_lock_stage
required: true type: enum
per_ticker: true formula: PROFIT_LOCK_RATCHET_V1
allowed: required: true
- "NORMAL" per_ticker: true
- "BREAKEVEN_RATCHET" allowed:
- "PROFIT_LOCK_10" - NORMAL
- "PROFIT_LOCK_20" - BREAKEVEN_RATCHET
- "PROFIT_LOCK_30" - PROFIT_LOCK_10
- "APEX_TRAILING" - PROFIT_LOCK_20
- "APEX_SUPER" - PROFIT_LOCK_30
- "SECULAR_LEADER_DEFERRED" - APEX_TRAILING
description: "수익 구간 단계 — APEX_SUPER(+60%)이면 trailing_stop 병기 필수" - APEX_SUPER
llm_action: "DATA_MISSING — trailing_stop 병기 불가" - SECULAR_LEADER_DEFERRED
criticality: "HIGH — APEX_SUPER 미판정 시 +60% 수익 종목에 보유유지만 서술하게 됨" description: 수익 구간 단계 — APEX_SUPER(+60%)이면 trailing_stop 병기 필수
llm_action: DATA_MISSING — trailing_stop 병기 불가
- gas_field: "auto_trailing_stop_v2" criticality: HIGH — APEX_SUPER 미판정 시 +60% 수익 종목에 보유유지만 서술하게 됨
type: numeric - gas_field: auto_trailing_stop_v2
formula: "PROFIT_RATCHET_TIERED_V2" type: numeric
required: false formula: PROFIT_RATCHET_TIERED_V2
per_ticker: true required: false
description: "ATR×1.2 기반 APEX_SUPER 자동 trailing stop(원)" per_ticker: true
note: "profit_lock_stage >= PROFIT_LOCK_20 일 때만 산출 (null이면 적용 안함)" description: ATR×1.2 기반 APEX_SUPER 자동 trailing stop(원)
llm_action: "DATA_MISSING — trailing_stop 병기 불가. 보유유지 단독 서술 허용되어 수익 보호 실패" note: profit_lock_stage >= PROFIT_LOCK_20 일 때만 산출 (null이면 적용 안함)
criticality: "CRITICAL — 삼성전자 +61.5% 사례(E3)에서 미산출로 수익 보호 실패" llm_action: DATA_MISSING — trailing_stop 병기 불가. 보유유지 단독 서술 허용되어 수익 보호 실패
criticality: CRITICAL — 삼성전자 +61.5% 사례(E3)에서 미산출로 수익 보호 실패
# ── STAGE 3: 설거지 감지 ────────────────────────────────────────────────── - gas_field: distribution_sell_detector_status
- gas_field: "distribution_sell_detector_status" type: enum
type: enum formula: DISTRIBUTION_SELL_DETECTOR_V1
formula: "DISTRIBUTION_SELL_DETECTOR_V1" required: true
required: true per_ticker: true
per_ticker: true allowed:
allowed: ["DISTRIBUTION_CONFIRMED", "DISTRIBUTION_WARNING", "DISTRIBUTION_CLEAR"] - DISTRIBUTION_CONFIRMED
description: "설거지 6신호 합산 감지 상태 — CONFIRMED 시 BUY 완전 차단" - DISTRIBUTION_WARNING
llm_action: "DATA_MISSING — '오를 것 같다' 주관 판단으로 매수 → 설거지 진입 위험" - DISTRIBUTION_CLEAR
criticality: "HIGH" description: 설거지 6신호 합산 감지 상태 — CONFIRMED 시 BUY 완전 차단
llm_action: DATA_MISSING — '오를 것 같다' 주관 판단으로 매수 → 설거지 진입 위험
# ── STAGE 4: 매수 게이트 ────────────────────────────────────────────────── criticality: HIGH
- gas_field: "anti_chasing_verdict" - gas_field: anti_chasing_verdict
type: enum type: enum
formula: "ANTI_CHASING_VELOCITY_V1" formula: ANTI_CHASING_VELOCITY_V1
required: true required: true
per_ticker: true per_ticker: true
allowed: ["BLOCK_CHASE", "PULLBACK_WAIT", "CLEAR"] allowed:
description: "당일 속도 기반 뒷박 추격 차단 — BLOCK_CHASE 시 당일 BUY 금지" - BLOCK_CHASE
llm_action: "DATA_MISSING — velocity_1d 미계산으로 뒷박 추격 매수 허용" - PULLBACK_WAIT
criticality: "CRITICAL — 뒷박 매수는 진입 당일 고점. 실패의 주원인." - CLEAR
description: 당일 속도 기반 뒷박 추격 차단 — BLOCK_CHASE 시 당일 BUY 금지
- gas_field: "pullback_entry_trigger_price" llm_action: DATA_MISSING — velocity_1d 미계산으로 뒷박 추격 매수 허용
type: numeric criticality: CRITICAL — 뒷박 매수는 진입 당일 고점. 실패의 주원인.
formula: "PULLBACK_ENTRY_TRIGGER_V1" - gas_field: pullback_entry_trigger_price
required: false type: numeric
per_ticker: true formula: PULLBACK_ENTRY_TRIGGER_V1
description: "눌림목 허용 기준가(원) = MA20 - 0.5×ATR20, tick 정규화" required: false
note: "PULLBACK_WAIT 상태일 때만 유효" per_ticker: true
llm_action: "DATA_MISSING — '가격이 괜찮아 보이면' 즉시 매수 → 눌림목 미확인 진입" description: 눌림목 허용 기준가(원) = MA20 - 0.5×ATR20, tick 정규화
criticality: "HIGH" note: PULLBACK_WAIT 상태일 때만 유효
llm_action: DATA_MISSING — '가격이 괜찮아 보이면' 즉시 매수 → 눌림목 미확인 진입
# ── STAGE 5: 현금확보 매도 ──────────────────────────────────────────────── criticality: HIGH
- gas_field: "cash_recovery_plan_json" - gas_field: cash_recovery_plan_json
type: json type: json
formula: "CASH_RECOVERY_OPTIMIZER_V1" formula: CASH_RECOVERY_OPTIMIZER_V1
required: true required: true
condition: "cash_shortfall_min_krw > 0" condition: cash_shortfall_min_krw > 0
description: "현금부족 최적 매도조합 JSON — H2 우선순위 기반 결정론적 산출" description: 현금부족 최적 매도조합 JSON — H2 우선순위 기반 결정론적 산출
schema: schema:
sell_sequence: "array of {ticker, qty, limit_price, expected_krw}" sell_sequence: array of {ticker, qty, limit_price, expected_krw}
expected_total_krw: "numeric" expected_total_krw: numeric
shortfall_met: "boolean" shortfall_met: boolean
llm_action: "DATA_MISSING — LLM이 '삼성E&A 100주+한화에어로 50주' 즉석 계산 → HS011 위반" llm_action: DATA_MISSING — LLM이 '삼성E&A 100주+한화에어로 50주' 즉석 계산 → HS011 위반
criticality: "CRITICAL — 현금확보 매도 조합이 LLM마다 달라짐" criticality: CRITICAL — 현금확보 매도 조합이 LLM마다 달라짐
- gas_field: waterfall_plan_json
- gas_field: "waterfall_plan_json" type: json
type: json formula: SELL_WATERFALL_ENGINE_V1
formula: "SELL_WATERFALL_ENGINE_V1" required: true
required: true condition: cash_shortfall_min_krw > 0
condition: "cash_shortfall_min_krw > 0" description: 4단계 폭포수 매도 계획 JSON
description: "4단계 폭포수 매도 계획 JSON" schema:
schema: current_stage: int 1~4
current_stage: "int 1~4" stage_label: enum [IMMEDIATE_TRIM,REBOUND_WAIT,CASCADING_TRIM,EMERGENCY_EXIT]
stage_label: "enum [IMMEDIATE_TRIM,REBOUND_WAIT,CASCADING_TRIM,EMERGENCY_EXIT]" sell_sequence: array of {ticker, stage, qty, limit_price, rebound_trigger_price}
sell_sequence: "array of {ticker, stage, qty, limit_price, rebound_trigger_price}" llm_action: DATA_MISSING — stage 순서 없이 즉흥 매도 → 주식가치 훼손
llm_action: "DATA_MISSING — stage 순서 없이 즉흥 매도 → 주식가치 훼손" criticality: HIGH
criticality: "HIGH" - gas_field: preservation_verdict
type: enum
- gas_field: "preservation_verdict" formula: SELL_VALUE_PRESERVATION_TIERED_V2
type: enum required: true
formula: "SELL_VALUE_PRESERVATION_TIERED_V2" per_ticker: true
required: true condition: Final_Action in [SELL_READY, TRIM]
per_ticker: true allowed:
condition: "Final_Action in [SELL_READY, TRIM]" - EMERGENCY_EXIT
allowed: - OVERSOLD_REBOUND_SELL
- "EMERGENCY_EXIT" - APEX_TRIM
- "OVERSOLD_REBOUND_SELL" - STAGED_EXIT
- "APEX_TRIM" - PRESERVE_TIERED
- "STAGED_EXIT" - HOLD
- "PRESERVE_TIERED" description: 주식가치 보호 매도 결정 — HOLD 외에는 구체 계획 필수
- "HOLD" llm_action: DATA_MISSING — 무작위 매도 스타일 서술
description: "주식가치 보호 매도 결정 — HOLD 외에는 구체 계획 필수" criticality: MEDIUM
llm_action: "DATA_MISSING — 무작위 매도 스타일 서술" - gas_field: sell_price_sanity_status
criticality: "MEDIUM" type: enum
formula: SELL_PRICE_SANITY_V1
# ── STAGE 6: 가격 검증 ──────────────────────────────────────────────────── required: true
- gas_field: "sell_price_sanity_status" per_ticker: true
type: enum condition: Final_Action in [SELL_READY, TRIM, EXIT_100]
formula: "SELL_PRICE_SANITY_V1" allowed:
required: true - PASS
per_ticker: true - INVALID_PRICE_INVERSION
condition: "Final_Action in [SELL_READY, TRIM, EXIT_100]" - INVALID_UNREALISTIC_PRICE
allowed: ["PASS", "INVALID_PRICE_INVERSION", "INVALID_UNREALISTIC_PRICE", "INVALID_TICK"] - INVALID_TICK
description: "매도가 역전·비현실가 검증 — INVALID 시 HTS 주문표 제거" description: 매도가 역전·비현실가 검증 — INVALID 시 HTS 주문표 제거
llm_action: "DATA_MISSING — LS Electric 사례처럼 역전 가격이 HTS 주문표에 그대로 들어감" llm_action: DATA_MISSING — LS Electric 사례처럼 역전 가격이 HTS 주문표에 그대로 들어감
criticality: "CRITICAL — 실제 손실 오류 E1의 직접 원인" criticality: CRITICAL — 실제 손실 오류 E1의 직접 원인
- gas_field: prices_json
# ── STAGE 6: HTS 주문 잠금 ──────────────────────────────────────────────── type: json
- gas_field: "prices_json" formula: PRICES_LOCK
type: json required: true
formula: "PRICES_LOCK" description: 종목별 stop_price, tp1_price, tp2_price JSON — LLM 재계산 금지
required: true schema:
description: "종목별 stop_price, tp1_price, tp2_price JSON — LLM 재계산 금지" stop_price: numeric KRW
schema: tp1_price: numeric KRW or null
stop_price: "numeric KRW" tp2_price: numeric KRW or null
tp1_price: "numeric KRW or null" profit_lock_stage: enum
tp2_price: "numeric KRW or null" llm_action: DATA_MISSING — LLM이 차트 지지선으로 손절가 임의 추정 → 매번 다른 값
profit_lock_stage: "enum" criticality: CRITICAL — 수량·가격 기반 모든 주문이 불확실해짐
llm_action: "DATA_MISSING — LLM이 차트 지지선으로 손절가 임의 추정 → 매번 다른 값" - gas_field: sell_quantities_json
criticality: "CRITICAL — 수량·가격 기반 모든 주문이 불확실해짐" type: json
formula: QUANTITIES_LOCK
- gas_field: "sell_quantities_json" required: true
type: json description: 종목별 매도 수량 잠금 JSON
formula: "QUANTITIES_LOCK" llm_action: DATA_MISSING — LLM이 '적절한 수량으로' 즉흥 계산
required: true criticality: CRITICAL
description: "종목별 매도 수량 잠금 JSON" - gas_field: order_blueprint_json
llm_action: "DATA_MISSING — LLM이 '적절한 수량으로' 즉흥 계산" type: json
criticality: "CRITICAL" formula: ORDER_BLUEPRINT
required: true
- gas_field: "order_blueprint_json" description: HTS 주문 청사진 JSON — validation_status=PASS만 HTS 입력 허용
type: json llm_action: DATA_MISSING — Shadow Ledger / HTS 주문표 분리 불가
formula: "ORDER_BLUEPRINT" criticality: CRITICAL
required: true - gas_field: rs_verdict
description: "HTS 주문 청사진 JSON — validation_status=PASS만 HTS 입력 허용" type: enum
llm_action: "DATA_MISSING — Shadow Ledger / HTS 주문표 분리 불가" formula: RS_VERDICT_V2
criticality: "CRITICAL" required: true
per_ticker: true
# ── STAGE 7: RS 판정 ────────────────────────────────────────────────────── allowed:
- gas_field: "rs_verdict" - LEADER
type: enum - NEUTRAL
formula: "RS_VERDICT_V2" - LAGGARD
required: true - BROKEN
per_ticker: true description: 최종 상대강도 판정 — BROKEN 시 매도 우선순위 최상위
allowed: ["LEADER", "NEUTRAL", "LAGGARD", "BROKEN"] llm_action: DATA_MISSING — '차트가 좋아 보이면 LEADER' 주관 판단
description: "최종 상대강도 판정 — BROKEN 시 매도 우선순위 최상위" criticality: HIGH
llm_action: "DATA_MISSING — '차트가 좋아 보이면 LEADER' 주관 판단" - gas_field: trade_quality_json
criticality: "HIGH" type: json
formula: TRADE_QUALITY_SCORER_V1
# ── MONTHLY BATCH ────────────────────────────────────────────────────────── required: false
- gas_field: "trade_quality_json" batch_only: true
type: json description: T+5/T+20 거래 품질 채점 결과 — POOR/CRITICAL 누적 블랙리스트 발동
formula: "TRADE_QUALITY_SCORER_V1" schema:
required: false ticker: string
batch_only: true score: int 0~100
description: "T+5/T+20 거래 품질 채점 결과 — POOR/CRITICAL 누적 블랙리스트 발동" grade: enum [EXCELLENT,GOOD,ACCEPTABLE,POOR,CRITICAL]
schema: feedback_tag: enum
ticker: "string" llm_action: DATA_MISSING — 'POOR 매매였지만 이번엔 다르다' 무근거 판단
score: "int 0~100" criticality: MEDIUM — 반복 실수 패턴 차단 불가
grade: "enum [EXCELLENT,GOOD,ACCEPTABLE,POOR,CRITICAL]"
feedback_tag: "enum"
llm_action: "DATA_MISSING — 'POOR 매매였지만 이번엔 다르다' 무근거 판단"
criticality: "MEDIUM — 반복 실수 패턴 차단 불가"
# ──────────────────────────────────────────────────────────────────────────────
# 커버리지 임계값
# ──────────────────────────────────────────────────────────────────────────────
coverage_thresholds: coverage_thresholds:
critical_fields_target_pct: 100 # CRITICAL 필드는 100% 필수 critical_fields_target_pct: 100
overall_target_pct: 85 # 전체 목표 커버리지 overall_target_pct: 85
llm_freedom_score_max: 15 # LLM 자유도 15% 이하 목표 llm_freedom_score_max: 15
grade_table: grade_table:
100: {grade: "DETERMINISTIC", label: "완전 결정론적 — 이상적 상태"} 100:
85_99: {grade: "NEAR_FULL", label: "거의 결정론적 — 배치 필드만 미계산"} grade: DETERMINISTIC
60_84: {grade: "PARTIAL", label: "부분 결정론적 — GAS 구현 필요"} label: 완전 결정론적 — 이상적 상태
0_59: {grade: "LLM_DEPENDENT", label: "LLM 의존 — 결과 재현 불가"} 8599:
grade: NEAR_FULL
# ────────────────────────────────────────────────────────────────────────────── label: 거의 결정론적 — 배치 필드만 미계산
# 현재 GAS 구현 상태 (2026-05-22 기준) 6084:
# ────────────────────────────────────────────────────────────────────────────── grade: PARTIAL
label: 부분 결정론적 — GAS 구현 필요
0_59:
grade: LLM_DEPENDENT
label: LLM 의존 — 결과 재현 불가
current_state: current_state:
gas_version: "2026-05-19-X4R1" gas_version: 2026-05-19-X4R1
overall_coverage_pct: 30 # 실측값 — measure_harness_coverage.py 참조 overall_coverage_pct: 30
llm_freedom_score: 70 # 100 - 30 = 70% → LLM 의존도 매우 높음 llm_freedom_score: 70
grade: "LLM_DEPENDENT" grade: LLM_DEPENDENT
critical_gaps: critical_gaps:
- field: "prices_json" - field: prices_json
status: "EMPTY" status: EMPTY
impact: "stop_price/tp_price 전부 LLM 추정 → 매 호출마다 다른 손절가" impact: stop_price/tp_price 전부 LLM 추정 → 매 호출마다 다른 손절가
- field: sell_quantities_json
- field: "sell_quantities_json" status: EMPTY
status: "EMPTY" impact: 매도 수량 LLM 추정 → 매 호출마다 다른 수량
impact: "매도 수량 LLM 추정 → 매 호출마다 다른 수량" - field: order_blueprint_json
status: EMPTY
- field: "order_blueprint_json" impact: HTS 주문 청사진 없음 → Shadow Ledger 분리 불가
status: "EMPTY" - field: anti_chasing_verdict
impact: "HTS 주문 청사진 없음 → Shadow Ledger 분리 불가" status: MISSING
impact: 뒷박 추격 매수 차단 미작동 → 진입 당일 고점 손실 반복
- field: "anti_chasing_verdict" - field: sell_price_sanity_status
status: "MISSING" status: MISSING
impact: "뒷박 추격 매수 차단 미작동 → 진입 당일 고점 손실 반복" impact: LS Electric 사례(E1) 재발 — 역전 매도가 HTS 입력 허용
- field: auto_trailing_stop_v2
- field: "sell_price_sanity_status" status: MISSING
status: "MISSING" impact: 삼성전자 +61.5% 사례(E3) 재발 — APEX_SUPER trailing_stop 미병기
impact: "LS Electric 사례(E1) 재발 — 역전 매도가 HTS 입력 허용" - field: rs_verdict
status: MISSING
- field: "auto_trailing_stop_v2" impact: RS_VERDICT_V2 미산출 → H2 매도 우선순위 BROKEN 판정 불가
status: "MISSING" - field: cash_recovery_plan_json
impact: "삼성전자 +61.5% 사례(E3) 재발 — APEX_SUPER trailing_stop 미병기" status: MISSING
impact: 현금확보 매도조합 LLM 즉석 계산 → HS011 위반 반복 (E2)
- field: "rs_verdict"
status: "MISSING"
impact: "RS_VERDICT_V2 미산출 → H2 매도 우선순위 BROKEN 판정 불가"
- field: "cash_recovery_plan_json"
status: "MISSING"
impact: "현금확보 매도조합 LLM 즉석 계산 → HS011 위반 반복 (E2)"
next_gas_implementation_priority: next_gas_implementation_priority:
1: "prices_json — stop_price, tp_price 실제 계산 및 채우기" 1: prices_json — stop_price, tp_price 실제 계산 및 채우기
2: "sell_quantities_json — Sell_Qty 실제 수량 채우기" 2: sell_quantities_json — Sell_Qty 실제 수량 채우기
3: "order_blueprint_json — HTS 주문 청사진 생성" 3: order_blueprint_json — HTS 주문 청사진 생성
4: "anti_chasing_verdict — velocity_1d 계산 + 차단 판정" 4: anti_chasing_verdict — velocity_1d 계산 + 차단 판정
5: "sell_price_sanity_status — 역전/비현실가 검증" 5: sell_price_sanity_status — 역전/비현실가 검증
6: "auto_trailing_stop_v2 — ATR×1.2 APEX_SUPER trailing" 6: auto_trailing_stop_v2 — ATR×1.2 APEX_SUPER trailing
7: "rs_verdict — RS_VERDICT_V2 실제 산출" 7: rs_verdict — RS_VERDICT_V2 실제 산출
8: "cash_recovery_plan_json — H2 순서 누적 매도조합" 8: cash_recovery_plan_json — H2 순서 누적 매도조합
spec/21_harness_governance_contract.yaml
+54 -54
View File
@@ -1,70 +1,70 @@
meta: meta:
title: "Harness Governance Contract" title: Harness Governance Contract
version: "2026-05-22-v1" version: 2026-05-22-v1
purpose: "하네스 준수 강제: 문서 지침 + 검증기 + 실행 게이트의 3중 잠금" purpose: '하네스 준수 강제: 문서 지침 + 검증기 + 실행 게이트의 3중 잠금'
has_code_implementation: true
code_path:
- spec\21_harness_governance_contract.yaml
governance: governance:
required_layers: required_layers:
- name: "static_guide" - name: static_guide
required_files: required_files:
- "AGENTS.md" - AGENTS.md
- "spec/07_output_schema.yaml" - spec/07_output_schema.yaml
- "spec/19_harness_contract.yaml" - spec/19_harness_contract.yaml
- name: "machine_validation" - name: machine_validation
required_validators: required_validators:
- "tools/validate_specs.py" - tools/validate_specs.py
- "tools/validate_harness_context.py" - tools/validate_harness_context.py
- "tools/validate_report_quality.py" - tools/validate_report_quality.py
- name: "execution_gate" - name: execution_gate
required_runners: required_runners:
- "tools/validate_engine_harness_gate.py" - tools/validate_engine_harness_gate.py
- "tools/run_engine_harness_gate.ps1" - tools/run_engine_harness_gate.ps1
- "tools/run_yolo_full_cycle.ps1" - tools/run_yolo_full_cycle.ps1
hardlocks: hardlocks:
- id: "HG001" - id: HG001
rule: "coverage_strict_100_required" rule: coverage_strict_100_required
fail_condition: "measure_harness_coverage --strict-100 미통과" fail_condition: measure_harness_coverage --strict-100 미통과
- id: "HG002" - id: HG002
rule: "watch_transparency_required" rule: watch_transparency_required
fail_condition: "WATCH_LEDGER_OK 미충족" fail_condition: WATCH_LEDGER_OK 미충족
- id: "HG003" - id: HG003
rule: "satellite_proposal_sheet_required" rule: satellite_proposal_sheet_required
fail_condition: "SATELLITE_PROPOSAL_SHEET_OK 미충족" fail_condition: SATELLITE_PROPOSAL_SHEET_OK 미충족
- id: "HG004" - id: HG004
rule: "strategy_harness_required" rule: strategy_harness_required
fail_condition: "STRATEGY_HARNESS_V2_OK 미충족" fail_condition: STRATEGY_HARNESS_V2_OK 미충족
gate_validity_rules: gate_validity_rules:
NON_VACUOUS_PASS_GUARD_V1: NON_VACUOUS_PASS_GUARD_V1:
formula_id: NON_VACUOUS_PASS_GUARD_V1 formula_id: NON_VACUOUS_PASS_GUARD_V1
rationale: > rationale: 'row_count=0 또는 sample_n < min_samples 인데 gate=PASS인 항목은 점수 분자를 부풀린다.
row_count=0 또는 sample_n < min_samples 인데 gate=PASS인 항목은 effective_n 미달 게이트는 WATCH_PENDING_SAMPLE로 강제 강등.
점수 분자를 부풀린다. effective_n 미달 게이트는 WATCH_PENDING_SAMPLE로 강제 강등.
'
min_samples_default: 30 min_samples_default: 30
min_samples_exceptions: min_samples_exceptions:
rebound_efficiency_score: 30 # 이전 4 → 30으로 상향 rebound_efficiency_score: 30
late_rebound_bucket_score: 30 late_rebound_bucket_score: 30
enforcement: enforcement:
- "effective_n < min_samples 이면 gate를 PASS로 둘 수 없다" - effective_n < min_samples 이면 gate를 PASS로 둘 수 없다
- "강등된 게이트는 release/pass_100 집계 분자(PASS count)에 포함 금지" - 강등된 게이트는 release/pass_100 집계 분자(PASS count)에 포함 금지
- "강등 라벨: WATCH_PENDING_SAMPLE" - '강등 라벨: WATCH_PENDING_SAMPLE'
- "보고서 해당 셀에 '[PASS_INVALID_LOW_N: n={effective_n} < {min}]' 라벨 부착" - '보고서 해당 셀에 ''[PASS_INVALID_LOW_N: n={effective_n} < {min}]'' 라벨 부착'
effective_n_fields: effective_n_fields:
- sample_count - sample_count
- row_count - row_count
- evaluated_count - evaluated_count
- samples - samples
- n - n
output: output:
- Temp/vacuous_pass_audit_v1.json - Temp/vacuous_pass_audit_v1.json
- operational_report.json.summary.vacuous_pass_gate_count - operational_report.json.summary.vacuous_pass_gate_count
python_tool: tools/build_vacuous_pass_audit_v1.py python_tool: tools/build_vacuous_pass_audit_v1.py
gs_coverage: "gas_apex_runtime_core.gs:guardNonVacuousPass_()" gs_coverage: gas_apex_runtime_core.gs:guardNonVacuousPass_()
validator: "tools/validate_harness_governance_contract.py --check non_vacuous_pass" validator: tools/validate_harness_governance_contract.py --check non_vacuous_pass
operations: operations:
release_policy: release_policy:
- "failed_checks 비어있지 않으면 배포/실행 차단" - failed_checks 비어있지 않으면 배포/실행 차단
- "gap_alert=true 이면 배포/실행 차단" - gap_alert=true 이면 배포/실행 차단
- "vacuous_pass_gate_count > 0 이면 배포/실행 차단 (NON_VACUOUS_PASS_GUARD_V1)" - vacuous_pass_gate_count > 0 이면 배포/실행 차단 (NON_VACUOUS_PASS_GUARD_V1)
spec/23_low_capability_llm_pipeline_todo.yaml
+64 -60
View File
@@ -1,66 +1,70 @@
meta:
has_code_implementation: true
code_path:
- spec\23_low_capability_llm_pipeline_todo.yaml
low_capability_llm_pipeline_todo: low_capability_llm_pipeline_todo:
formula_id: LOW_CAPABILITY_LLM_PIPELINE_TODO_V2 formula_id: LOW_CAPABILITY_LLM_PIPELINE_TODO_V2
objective: 저성능 LLM을 위한 기계적 복사 보고 절차 규정 objective: 저성능 LLM을 위한 기계적 복사 보고 절차 규정
ordered_steps: ordered_steps:
- step_id: STEP_01 - step_id: STEP_01
action: "AGENTS.md 읽기" action: AGENTS.md 읽기
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_02 - step_id: STEP_02
action: "active manifest 읽기" action: active manifest 읽기
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_03 - step_id: STEP_03
action: "final_context 읽기" action: final_context 읽기
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_04 - step_id: STEP_04
action: "engine gate status 확인" action: engine gate status 확인
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_05 - step_id: STEP_05
action: "blockers 먼저 출력" action: blockers 먼저 출력
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_06 - step_id: STEP_06
action: "allowed/blocked actions 복사" action: allowed/blocked actions 복사
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_07 - step_id: STEP_07
action: "shadow ledger 복사" action: shadow ledger 복사
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_08 - step_id: STEP_08
action: "data_missing 복사" action: data_missing 복사
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_09 - step_id: STEP_09
action: "숫자 provenance 확인" action: 숫자 provenance 확인
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_10 - step_id: STEP_10
action: "자유 계산 제거" action: 자유 계산 제거
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_11 - step_id: STEP_11
action: "report contract 검증" action: report contract 검증
ambiguous: false ambiguous: false
calculation: false calculation: false
- step_id: STEP_12 - step_id: STEP_12
action: "실패 시 DATA_MISSING 또는 REVIEW_ONLY로 종료" action: 실패 시 DATA_MISSING 또는 REVIEW_ONLY로 종료
ambiguous: false ambiguous: false
calculation: false calculation: false
forbidden_actions: forbidden_actions:
- do not set --skip-validate as default resolution - do not set --skip-validate as default resolution
- do not remove validate-engine-strict from release gate - do not remove validate-engine-strict from release gate
- do not mark success without engine_harness_gate_result.status=OK - do not mark success without engine_harness_gate_result.status=OK
completion_criteria: completion_criteria:
- Temp/engine_harness_gate_result.json.status == OK - Temp/engine_harness_gate_result.json.status == OK
- len(Temp/engine_harness_gate_result.json.failed_checks) == 0 - len(Temp/engine_harness_gate_result.json.failed_checks) == 0
- Temp/formula_runtime_registry_v1.json.runtime_adjusted_coverage_pct == 100.0 - Temp/formula_runtime_registry_v1.json.runtime_adjusted_coverage_pct == 100.0
- Temp/formula_runtime_registry_v1.json.unmapped_formula_count == 0 - Temp/formula_runtime_registry_v1.json.unmapped_formula_count == 0
- Temp/data_quality_reconciliation_v1.json.schema_presence_score == 100.0 - Temp/data_quality_reconciliation_v1.json.schema_presence_score == 100.0
- Temp/data_quality_reconciliation_v1.json.quality_conflict_flag in [true, false] - Temp/data_quality_reconciliation_v1.json.quality_conflict_flag in [true, false]
- Temp/operational_alpha_calibration_v2.json.formula_id == OPERATIONAL_ALPHA_CALIBRATION_V2 - Temp/operational_alpha_calibration_v2.json.formula_id == OPERATIONAL_ALPHA_CALIBRATION_V2
- Temp/pipeline_runtime_profile_v1.json.mode in [release, quick, package-only] - Temp/pipeline_runtime_profile_v1.json.mode in [release, quick, package-only]
- Temp/pipeline_runtime_profile_v1.json.gate_status == OK - Temp/pipeline_runtime_profile_v1.json.gate_status == OK
spec/24_strategy_hardening_todo_v1.yaml
+250 -236
View File
@@ -1,3 +1,7 @@
meta:
has_code_implementation: true
code_path:
- spec\24_strategy_hardening_todo_v1.yaml
strategy_hardening_todo_v1: strategy_hardening_todo_v1:
formula_id: STRATEGY_HARDENING_TODO_V1 formula_id: STRATEGY_HARDENING_TODO_V1
objective: objective:
@@ -7,7 +11,7 @@ strategy_hardening_todo_v1:
target_metrics: target_metrics:
engine_gate_status: OK engine_gate_status: OK
failed_checks_count: 0 failed_checks_count: 0
formula_total: 170 # 실제값 170 (163→170 갱신 2026-05-30) formula_total: 170
declared_runtime_count: 170 declared_runtime_count: 170
runtime_adjusted_coverage_pct: 100.0 runtime_adjusted_coverage_pct: 100.0
unmapped_formula_count: 0 unmapped_formula_count: 0
@@ -17,255 +21,269 @@ strategy_hardening_todo_v1:
outcome_quality_score_v1_min: 60.0 outcome_quality_score_v1_min: 60.0
value_damage_pct_avg_max: 10.0 value_damage_pct_avg_max: 10.0
llm_freedom_pct: 0.0 llm_freedom_pct: 0.0
# BCH-V1 행위기반 커버리지 (P1 달성 2026-05-30)
behavioral_coverage_pct: 100.0 behavioral_coverage_pct: 100.0
implementation_divergence_count: 0 implementation_divergence_count: 0
# P2 임계값 보정 레지스트리
unregistered_threshold_count: 0 unregistered_threshold_count: 0
overclaimed_calibration_count: 0 overclaimed_calibration_count: 0
# P3 LLM 자유도 측정
llm_freedom_pct_measured: 0.0 llm_freedom_pct_measured: 0.0
ungrounded_number_count: 0 ungrounded_number_count: 0
# P4 정직 성과증빙
design_score_as_proof_violations: 0 design_score_as_proof_violations: 0
ordered_todo: ordered_todo:
- id: T01_COVERAGE_AUDIT - id: T01_COVERAGE_AUDIT
command: python tools/harness_coverage_auditor.py command: python tools/harness_coverage_auditor.py
expect: expect:
true_missing_count: 0 true_missing_count: 0
coverage_pct_min: 95.0 coverage_pct_min: 95.0
fail_code: HARNESS_COVERAGE_AUDIT_FAIL fail_code: HARNESS_COVERAGE_AUDIT_FAIL
- id: T02_RUNTIME_REGISTRY_BUILD - id: T02_RUNTIME_REGISTRY_BUILD
command: python tools/build_formula_runtime_registry_v1.py --audit Temp/harness_coverage_audit.json --out Temp/formula_runtime_registry_v1.json command: python tools/build_formula_runtime_registry_v1.py --audit Temp/harness_coverage_audit.json
expect: --out Temp/formula_runtime_registry_v1.json
formula_total: 163 expect:
declared_runtime_count: 163 formula_total: 163
runtime_adjusted_coverage_pct: 100.0 declared_runtime_count: 163
unmapped_formula_count: 0 runtime_adjusted_coverage_pct: 100.0
fail_code: FORMULA_IMPLEMENTATION_REGISTRY_V1_FAIL unmapped_formula_count: 0
- id: T03_RUNTIME_REGISTRY_VALIDATE fail_code: FORMULA_IMPLEMENTATION_REGISTRY_V1_FAIL
command: python tools/validate_formula_runtime_registry_v1.py --json Temp/formula_runtime_registry_v1.json --target-coverage 100 - id: T03_RUNTIME_REGISTRY_VALIDATE
expect: command: python tools/validate_formula_runtime_registry_v1.py --json Temp/formula_runtime_registry_v1.json
status_token: FORMULA_IMPLEMENTATION_REGISTRY_V1_OK --target-coverage 100
fail_code: FORMULA_IMPLEMENTATION_REGISTRY_V1_FAIL expect:
- id: T04_DQ_RECON_BUILD status_token: FORMULA_IMPLEMENTATION_REGISTRY_V1_OK
command: python tools/build_data_quality_reconciliation_v1.py --json GatherTradingData.json --integrity Temp/data_integrity_score_v1.json --out Temp/data_quality_reconciliation_v1.json fail_code: FORMULA_IMPLEMENTATION_REGISTRY_V1_FAIL
expect: - id: T04_DQ_RECON_BUILD
formula_id: DATA_QUALITY_RECONCILIATION_V1 command: python tools/build_data_quality_reconciliation_v1.py --json GatherTradingData.json
fail_code: DATA_QUALITY_RECONCILIATION_V1_BUILD_FAIL --integrity Temp/data_integrity_score_v1.json --out Temp/data_quality_reconciliation_v1.json
- id: T05_DQ_RECON_VALIDATE expect:
command: python tools/validate_data_quality_reconciliation_v1.py --json Temp/data_quality_reconciliation_v1.json --min-schema-score 100 --min-investment-quality-score 90 formula_id: DATA_QUALITY_RECONCILIATION_V1
expect: fail_code: DATA_QUALITY_RECONCILIATION_V1_BUILD_FAIL
status_token: DATA_QUALITY_RECONCILIATION_V1_OK - id: T05_DQ_RECON_VALIDATE
fail_code: DATA_QUALITY_RECONCILIATION_V1_FAIL command: python tools/validate_data_quality_reconciliation_v1.py --json Temp/data_quality_reconciliation_v1.json
note: "실데이터 부족 구간은 임시로 FAIL 허용하지 않고 WARN 원장으로 기록 후 원인 해결" --min-schema-score 100 --min-investment-quality-score 90
- id: T06_ENGINE_GATE expect:
command: python tools/validate_engine_harness_gate.py --json GatherTradingData.json --report Temp/operational_report.md --harness-json Temp/prediction_improvement_harness.json --result-json Temp/engine_harness_gate_result.json --rule-lifecycle-json Temp/rule_lifecycle_policy.json --strategy-harness-json Temp/strategy_harness_v2.json status_token: DATA_QUALITY_RECONCILIATION_V1_OK
expect: fail_code: DATA_QUALITY_RECONCILIATION_V1_FAIL
status: OK note: 실데이터 부족 구간은 임시로 FAIL 허용하지 않고 WARN 원장으로 기록 후 원인 해결
failed_checks_count: 0 - id: T06_ENGINE_GATE
required_checks: command: python tools/validate_engine_harness_gate.py --json GatherTradingData.json
- CHECK_81_FORMULA_RUNTIME_REGISTRY_V1 --report Temp/operational_report.md --harness-json Temp/prediction_improvement_harness.json
- CHECK_82_DATA_QUALITY_RECONCILIATION_V1 --result-json Temp/engine_harness_gate_result.json --rule-lifecycle-json Temp/rule_lifecycle_policy.json
fail_code: ENGINE_HARNESS_GATE_FAIL --strategy-harness-json Temp/strategy_harness_v2.json
- id: T07_RELEASE_PACKAGE expect:
command: npm run prepare-upload-zip -- --validation-mode release --profile status: OK
expect: failed_checks_count: 0
gate_status: OK required_checks:
profile_exists: Temp/pipeline_runtime_profile_v1.json - CHECK_81_FORMULA_RUNTIME_REGISTRY_V1
package_exists: ../data_feed.zip - CHECK_82_DATA_QUALITY_RECONCILIATION_V1
fail_code: PREPARE_UPLOAD_ZIP_FAIL fail_code: ENGINE_HARNESS_GATE_FAIL
# ── P1: 행위기반 커버리지 하네스 (BCH-V1) ───────────────────────────────── - id: T07_RELEASE_PACKAGE
- id: B01_BCH_CONTRACT command: npm run prepare-upload-zip -- --validation-mode release --profile
command: "# spec/26_behavioral_coverage_contract.yaml 작성 완료" expect:
expect: {file_exists: "spec/26_behavioral_coverage_contract.yaml", decision_critical_count: 40} gate_status: OK
status: DONE_2026_05_30 profile_exists: Temp/pipeline_runtime_profile_v1.json
- id: B02_GOLDEN_AUTHOR package_exists: ../data_feed.zip
command: "# spec/formula_golden_cases_v2.yaml 손계산 골든케이스 작성 완료" fail_code: PREPARE_UPLOAD_ZIP_FAIL
expect: {cases_total_min: 18, provenance: "HAND_COMPUTED or SPEC_DERIVED only"} - id: B01_BCH_CONTRACT
status: DONE_2026_05_30 command: '# spec/26_behavioral_coverage_contract.yaml 작성 완료'
- id: B03_PY_MIRROR expect:
command: python tools/run_formula_golden_cases_v2.py file_exists: spec/26_behavioral_coverage_contract.yaml
expect: decision_critical_count: 40
status_token: BEHAVIORAL_COVERAGE_PY_OK status: DONE_2026_05_30
behavioral_coverage_pct: 100.0 - id: B02_GOLDEN_AUTHOR
python_fail: 0 command: '# spec/formula_golden_cases_v2.yaml 손계산 골든케이스 작성 완료'
fail_code: BCH_PY_MIRROR_FAIL expect:
- id: B04_GAS_PARITY cases_total_min: 18
command: node tools/run_gas_golden_parity.js provenance: HAND_COMPUTED or SPEC_DERIVED only
expect: status: DONE_2026_05_30
status_token: GAS_PARITY_OK - id: B03_PY_MIRROR
gas_fail: 0 command: python tools/run_formula_golden_cases_v2.py
fail_code: BCH_GAS_PARITY_FAIL expect:
- id: B05_3WAY_VALIDATE status_token: BEHAVIORAL_COVERAGE_PY_OK
command: python tools/validate_behavioral_coverage_v1.py --strict behavioral_coverage_pct: 100.0
expect: python_fail: 0
status_token: BEHAVIORAL_COVERAGE_V1_OK fail_code: BCH_PY_MIRROR_FAIL
behavioral_coverage_pct: 100.0 - id: B04_GAS_PARITY
implementation_divergence_count: 0 command: node tools/run_gas_golden_parity.js
fail_code: BEHAVIORAL_COVERAGE_V1_FAIL expect:
- id: B06_DIVERGENCE_FIX status_token: GAS_PARITY_OK
command: "# normalize_tick round→floor 수정, PROFIT_LOCK_STAGE GAS 단계명 정정 완료" gas_fail: 0
expect: {divergence_count: 0} fail_code: BCH_GAS_PARITY_FAIL
status: DONE_2026_05_30 - id: B05_3WAY_VALIDATE
- id: B07_WIRE_FULLGATE command: python tools/validate_behavioral_coverage_v1.py --strict
command: npm run validate-behavioral-coverage expect:
expect: {exit_code: 0} status_token: BEHAVIORAL_COVERAGE_V1_OK
fail_code: BCH_WIRING_FAIL behavioral_coverage_pct: 100.0
# ── P2: 임계값 보정 레지스트리 (CALIB-V1) ──────────────────────────────── implementation_divergence_count: 0
- id: P2_REGISTRY_BUILD fail_code: BEHAVIORAL_COVERAGE_V1_FAIL
command: "# spec/calibration_registry.yaml 69개 임계값 등록 완료" - id: B06_DIVERGENCE_FIX
expect: {total_thresholds_min: 60} command: '# normalize_tick round→floor 수정, PROFIT_LOCK_STAGE GAS 단계명 정정 완료'
status: DONE_2026_05_30 expect:
- id: P2_REGISTRY_VALIDATE divergence_count: 0
command: python tools/validate_calibration_registry_v1.py status: DONE_2026_05_30
expect: - id: B07_WIRE_FULLGATE
overclaimed_count: 0 command: npm run validate-behavioral-coverage
unregistered_threshold_count: 0 expect:
status_token: "CALIBRATION_REGISTRY_WARN or CALIBRATION_REGISTRY_OK" exit_code: 0
fail_code: CALIBRATION_REGISTRY_FAIL fail_code: BCH_WIRING_FAIL
- id: P2_PRIORITY_BUILD - id: P2_REGISTRY_BUILD
command: python tools/build_calibration_priority_v1.py command: '# spec/calibration_registry.yaml 69개 임계값 등록 완료'
expect: expect:
status_token: CALIBRATION_PRIORITY_OK total_thresholds_min: 60
priority_count_min: 5 status: DONE_2026_05_30
fail_code: CALIBRATION_PRIORITY_FAIL - id: P2_REGISTRY_VALIDATE
# ── P3: LLM 자유도 측정·폐쇄 (LFM-V1) ────────────────────────────────── command: python tools/validate_calibration_registry_v1.py
- id: P3_FREEDOM_VALIDATE expect:
command: python tools/validate_number_provenance_v1.py overclaimed_count: 0
expect: unregistered_threshold_count: 0
status_token: LFM_V1_OK status_token: CALIBRATION_REGISTRY_WARN or CALIBRATION_REGISTRY_OK
llm_freedom_pct: 0.0 fail_code: CALIBRATION_REGISTRY_FAIL
fail_code: LFM_V1_FAIL - id: P2_PRIORITY_BUILD
- id: P3_NARRATIVE_LOCK command: python tools/build_calibration_priority_v1.py
command: python tools/build_llm_narrative_template_lock_v1.py expect:
expect: status_token: CALIBRATION_PRIORITY_OK
gate: PASS priority_count_min: 5
total_violations: 0 fail_code: CALIBRATION_PRIORITY_FAIL
softening_violations: 0 - id: P3_FREEDOM_VALIDATE
fail_code: LLM_NARRATIVE_LOCK_FAIL command: python tools/validate_number_provenance_v1.py
# ── P4: 정직 성과증빙 + 보정루프 (HONEST-V1) ───────────────────────────── expect:
- id: P4_HONEST_GUARD status_token: LFM_V1_OK
command: python tools/build_honest_performance_guard_v1.py llm_freedom_pct: 0.0
expect: fail_code: LFM_V1_FAIL
status_token: "HONEST_PERFORMANCE_V1_OK or HONEST_PERFORMANCE_V1_WARN" - id: P3_NARRATIVE_LOCK
design_score_note: "UNVALIDATED_DESIGN_SCORE 표기 필수 (samples<30)" command: python tools/build_llm_narrative_template_lock_v1.py
fail_code: HONEST_PERFORMANCE_V1_FAIL expect:
# ── 반도체 집중 허용 하네스 ──────────────────────────────────────────────── gate: PASS
- id: SEMI_CONCENTRATION_POLICY total_violations: 0
command: "# spec/strategy/semiconductor_concentration_policy.yaml 작성 완료" softening_violations: 0
expect: {file_exists: "spec/strategy/semiconductor_concentration_policy.yaml"} fail_code: LLM_NARRATIVE_LOCK_FAIL
status: DONE_2026_05_30 - id: P4_HONEST_GUARD
command: python tools/build_honest_performance_guard_v1.py
- id: SEMI_CLUSTER_GATE_UPDATE expect:
command: "# gas_data_feed.gs calcSemiconductorClusterGate_ → MARKET_WEIGHT_AWARE_CLUSTER_GATE_V1 업데이트" status_token: HONEST_PERFORMANCE_V1_OK or HONEST_PERFORMANCE_V1_WARN
expect: {formula_id: "MARKET_WEIGHT_AWARE_CLUSTER_GATE_V1", kospi_weight_settings_driven: true} design_score_note: UNVALIDATED_DESIGN_SCORE 표기 필수 (samples<30)
status: DONE_2026_05_30 fail_code: HONEST_PERFORMANCE_V1_FAIL
- id: SEMI_CONCENTRATION_POLICY
- id: LEADER_CAP_UPDATE command: '# spec/strategy/semiconductor_concentration_policy.yaml 작성 완료'
command: "# gas_data_feed.gs calcSinglePositionWeightCap_ → LEADER_POSITION_WEIGHT_CAP_V1 업데이트" expect:
expect: {samsung_risk_on_cap: 40, hynix_risk_on_cap: 22, kospi_weight_settings_driven: true} file_exists: spec/strategy/semiconductor_concentration_policy.yaml
status: DONE_2026_05_30 status: DONE_2026_05_30
- id: SEMI_CLUSTER_GATE_UPDATE
- id: SECULAR_LEADER_AUTO_DETECT command: '# gas_data_feed.gs calcSemiconductorClusterGate_ → MARKET_WEIGHT_AWARE_CLUSTER_GATE_V1
command: "# gas_data_feed.gs calcSecularLeaderAutoDetect_ 함수 신설" 업데이트'
expect: {formula_id: "SECULAR_LEADER_AUTO_DETECT_V1", threshold: 6} expect:
status: DONE_2026_05_30 formula_id: MARKET_WEIGHT_AWARE_CLUSTER_GATE_V1
kospi_weight_settings_driven: true
- id: SEMI_INJECT_UPDATE status: DONE_2026_05_30
command: "# tools/inject_computed_harness.py 클러스터/개별 게이트 함수 KOSPI 비중 반영" - id: LEADER_CAP_UPDATE
expect: {settings_kospi_semi_weight_pct: true, gate_overwrite_direct: true} command: '# gas_data_feed.gs calcSinglePositionWeightCap_ → LEADER_POSITION_WEIGHT_CAP_V1
status: DONE_2026_05_30 업데이트'
expect:
- id: AGENTS_O1_O2_UPDATE samsung_risk_on_cap: 40
command: "# AGENTS.md Direction O1/O2 새 공식명·차등한도로 업데이트" hynix_risk_on_cap: 22
expect: {o1_formula: "LEADER_POSITION_WEIGHT_CAP_V1", o2_formula: "MARKET_WEIGHT_AWARE_CLUSTER_GATE_V1"} kospi_weight_settings_driven: true
status: DONE_2026_05_30 status: DONE_2026_05_30
- id: SECULAR_LEADER_AUTO_DETECT
# ── 통합 게이트 ────────────────────────────────────────────────────────── command: '# gas_data_feed.gs calcSecularLeaderAutoDetect_ 함수 신설'
- id: INTEGRATED_ENGINE_INTEGRITY expect:
command: npm run validate-engine-integrity formula_id: SECULAR_LEADER_AUTO_DETECT_V1
expect: threshold: 6
behavioral_coverage_pct: 100.0 status: DONE_2026_05_30
implementation_divergence_count: 0 - id: SEMI_INJECT_UPDATE
overclaimed_count: 0 command: '# tools/inject_computed_harness.py 클러스터/개별 게이트 함수 KOSPI 비중 반영'
unregistered_threshold_count: 0 expect:
llm_freedom_pct: 0.0 settings_kospi_semi_weight_pct: true
softening_violations: 0 gate_overwrite_direct: true
fail_code: ENGINE_INTEGRITY_FAIL status: DONE_2026_05_30
- id: AGENTS_O1_O2_UPDATE
# ── CAPITAL_STYLE_ALLOCATION_V1 (Section 3B) ──────────────────────────── command: '# AGENTS.md Direction O1/O2 새 공식명·차등한도로 업데이트'
- id: C1_BUILD_CAPITAL_STYLE_ALLOC expect:
command: python tools/build_capital_style_allocation_v1.py o1_formula: LEADER_POSITION_WEIGHT_CAP_V1
expect: {gate: PASS, ticker_count_min: 1, conviction_range: "[0,100]"} o2_formula: MARKET_WEIGHT_AWARE_CLUSTER_GATE_V1
fail_code: CAPITAL_ALLOC_BUILD_FAIL status: DONE_2026_05_30
status: DONE_2026_05_30 - id: INTEGRATED_ENGINE_INTEGRITY
command: npm run validate-engine-integrity
- id: C2_VALIDATE_CAPITAL_STYLE_ALLOC expect:
command: python tools/validate_capital_style_allocation_v1.py behavioral_coverage_pct: 100.0
expect: {status_token: CAPITAL_ALLOC_OK, violations: 0} implementation_divergence_count: 0
fail_code: CAPITAL_ALLOC_VALIDATE_FAIL overclaimed_count: 0
status: DONE_2026_05_30 unregistered_threshold_count: 0
llm_freedom_pct: 0.0
- id: C3_CALIB_W_STYLE_REGISTER softening_violations: 0
command: python tools/validate_calibration_registry_v1.py fail_code: ENGINE_INTEGRITY_FAIL
expect: {total_thresholds_min: 130, unregistered: 0, overclaimed: 0} - id: C1_BUILD_CAPITAL_STYLE_ALLOC
status: DONE_2026_05_30 command: python tools/build_capital_style_allocation_v1.py
expect:
- id: C4_GOLDEN_CASE_CAPITAL_STYLE gate: PASS
command: npm run validate-behavioral-coverage ticker_count_min: 1
expect: {status_token: BEHAVIORAL_COVERAGE_V1_OK, behavioral_coverage_pct: 100.0} conviction_range: '[0,100]'
fail_code: BCH_CAPITAL_STYLE_FAIL fail_code: CAPITAL_ALLOC_BUILD_FAIL
status: DONE_2026_05_30 status: DONE_2026_05_30
- id: C2_VALIDATE_CAPITAL_STYLE_ALLOC
- id: C5_WIRE_FULL_GATE_CAPITAL_STYLE command: python tools/validate_capital_style_allocation_v1.py
command: npm run full-gate expect:
expect: {exit_code: 0} status_token: CAPITAL_ALLOC_OK
fail_code: WIRE_FAIL violations: 0
status: DONE_2026_05_30 fail_code: CAPITAL_ALLOC_VALIDATE_FAIL
status: DONE_2026_05_30
- id: S1_AGENTS_MD_DIRECTION_S1 - id: C3_CALIB_W_STYLE_REGISTER
command: "# AGENTS.md Direction S1 추가" command: python tools/validate_calibration_registry_v1.py
expect: {direction_s1_exists: true} expect:
status: DONE_2026_05_30 total_thresholds_min: 130
unregistered: 0
- id: S2_INJECT_RENDER_CAPITAL overclaimed: 0
command: npm run render-report-json status: DONE_2026_05_30
expect: {capital_style_conviction_section_exists: true} - id: C4_GOLDEN_CASE_CAPITAL_STYLE
status: DONE_2026_05_30 command: npm run validate-behavioral-coverage
expect:
status_token: BEHAVIORAL_COVERAGE_V1_OK
behavioral_coverage_pct: 100.0
fail_code: BCH_CAPITAL_STYLE_FAIL
status: DONE_2026_05_30
- id: C5_WIRE_FULL_GATE_CAPITAL_STYLE
command: npm run full-gate
expect:
exit_code: 0
fail_code: WIRE_FAIL
status: DONE_2026_05_30
- id: S1_AGENTS_MD_DIRECTION_S1
command: '# AGENTS.md Direction S1 추가'
expect:
direction_s1_exists: true
status: DONE_2026_05_30
- id: S2_INJECT_RENDER_CAPITAL
command: npm run render-report-json
expect:
capital_style_conviction_section_exists: true
status: DONE_2026_05_30
evidence_artifacts: evidence_artifacts:
- Temp/harness_coverage_audit.json - Temp/harness_coverage_audit.json
- Temp/formula_runtime_registry_v1.json - Temp/formula_runtime_registry_v1.json
- Temp/data_quality_reconciliation_v1.json - Temp/data_quality_reconciliation_v1.json
- Temp/engine_harness_gate_result.json - Temp/engine_harness_gate_result.json
- Temp/pipeline_runtime_profile_v1.json - Temp/pipeline_runtime_profile_v1.json
# BCH-V1 추가 (2026-05-30) - Temp/formula_behavioral_coverage_v1.json
- Temp/formula_behavioral_coverage_v1.json - Temp/formula_gas_parity_v1.json
- Temp/formula_gas_parity_v1.json - Temp/formula_behavioral_coverage_summary_v1.json
- Temp/formula_behavioral_coverage_summary_v1.json - Temp/calibration_registry_v1.json
- Temp/calibration_registry_v1.json - Temp/calibration_priority_v1.json
- Temp/calibration_priority_v1.json - Temp/llm_freedom_v1.json
- Temp/llm_freedom_v1.json - Temp/honest_performance_guard_v1.json
- Temp/honest_performance_guard_v1.json
completion_definition: completion_definition:
hard_requirements: hard_requirements:
- "모든 숫자 산출은 하네스 JSON 근거가 있어야 한다" - 모든 숫자 산출은 하네스 JSON 근거가 있어야 한다
- "HTS 주문표와 WATCH 원장을 물리적으로 분리해야 한다" - HTS 주문표와 WATCH 원장을 물리적으로 분리해야 한다
- "runtime_adjusted_coverage_pct 100%를 숫자로 증빙해야 한다" - runtime_adjusted_coverage_pct 100%를 숫자로 증빙해야 한다
- "데이터 품질 충돌은 숨기지 않고 quality_conflict_flag로 보고해야 한다" - 데이터 품질 충돌은 숨기지 않고 quality_conflict_flag로 보고해야 한다
reject_conditions: reject_conditions:
- "정량 근거 없이 100% 완료 문구 사용" - 정량 근거 없이 100% 완료 문구 사용
- "llm 추정값으로 가격/수량 생성" - llm 추정값으로 가격/수량 생성
- "engine_harness_gate_result.status!=OK 인데 완료 선언" - engine_harness_gate_result.status!=OK 인데 완료 선언
current_status: current_status:
as_of: "2026-05-30" as_of: '2026-05-30'
T01_T03: PASS (coverage 100%, formula_total=168) T01_T03: PASS (coverage 100%, formula_total=168)
T04: PASS (data_quality_reconciliation built) T04: PASS (data_quality_reconciliation built)
T05: FAIL_WARN (investment_quality=13% - 펀더멘털 미수집, 데이터 수집만이 해결) T05: FAIL_WARN (investment_quality=13% - 펀더멘털 미수집, 데이터 수집만이 해결)
T06: STATUS=OK (engine_gate 1개 WARN_ONLY fail) T06: STATUS=OK (engine_gate 1개 WARN_ONLY fail)
T07: 미실행 T07: 미실행
# BCH-V1 4-기둥 추가 완료 (2026-05-30)
B01_B07: PASS (behavioral_coverage_pct=100%, divergence=0, GAS pass=45/45) B01_B07: PASS (behavioral_coverage_pct=100%, divergence=0, GAS pass=45/45)
B06_FIX: normalize_tick round→floor 수정 + PROFIT_LOCK_STAGE 단계명 7개 spec 일치 정정 B06_FIX: normalize_tick round→floor 수정 + PROFIT_LOCK_STAGE 단계명 7개 spec 일치 정정
P2_CALIB: overclaimed=0, unregistered=0, 69개 임계값 EXPERT_PRIOR 정직 공시 P2_CALIB: overclaimed=0, unregistered=0, 69개 임계값 EXPERT_PRIOR 정직 공시
@@ -290,9 +308,7 @@ strategy_hardening_todo_v1:
leader_position_weight_cap: LEADER_POSITION_WEIGHT_CAP_V1 leader_position_weight_cap: LEADER_POSITION_WEIGHT_CAP_V1
secular_leader_auto_detect: calcSecularLeaderAutoDetect_ 신설 secular_leader_auto_detect: calcSecularLeaderAutoDetect_ 신설
kospi_weights: settings 시트 입력값 반영 (하드코딩 금지) kospi_weights: settings 시트 입력값 반영 (하드코딩 금지)
# CAPITAL_STYLE_ALLOCATION_V1 (Section 3B)
capital_style_allocation: DONE_2026_05_30 (C1~C5+S1+S2 완료) capital_style_allocation: DONE_2026_05_30 (C1~C5+S1+S2 완료)
final_completion_2026_05_30_extended: final_completion_2026_05_30_extended:
outcome_quality: 85.23 (PASS) outcome_quality: 85.23 (PASS)
guidance_proof: 99.26 (PASS) guidance_proof: 99.26 (PASS)
@@ -300,7 +316,6 @@ strategy_hardening_todo_v1:
t5_op_rate: 73.24% (CALIBRATED) t5_op_rate: 73.24% (CALIBRATED)
canonical_conflicts: 0 (score=100) canonical_conflicts: 0 (score=100)
all_major_targets: ACHIEVED all_major_targets: ACHIEVED
# 추가 달성 (2026-05-30 2차)
behavioral_coverage_pct: 100.0 (PASS) behavioral_coverage_pct: 100.0 (PASS)
implementation_divergence_count: 0 (PASS) implementation_divergence_count: 0 (PASS)
gas_parity_cases: 45/45 (PASS) gas_parity_cases: 45/45 (PASS)
@@ -309,7 +324,6 @@ strategy_hardening_todo_v1:
llm_freedom_pct_measured: 0.0 (PASS) llm_freedom_pct_measured: 0.0 (PASS)
softening_violations: 0 (PASS) softening_violations: 0 (PASS)
design_score_as_proof_violations: 1 (WARN - UNVALIDATED 라벨 추가로 정직 처리) design_score_as_proof_violations: 1 (WARN - UNVALIDATED 라벨 추가로 정직 처리)
# CAPITAL_STYLE_ALLOCATION_V1 (Section 3B) 달성 (2026-05-30 최종)
capital_style_allocation_gate: PASS (11종목, CAPITAL_ALLOC_OK) capital_style_allocation_gate: PASS (11종목, CAPITAL_ALLOC_OK)
capital_style_calibration: total=134 thresholds, overclaimed=0, unregistered=0 capital_style_calibration: total=134 thresholds, overclaimed=0, unregistered=0
capital_style_report_section: operational_report에 conviction 표 렌더링 완료 capital_style_report_section: operational_report에 conviction 표 렌더링 완료
spec/27_bch_calibration_runbook.yaml
+320 -431
View File
@@ -1,528 +1,417 @@
# spec/27_bch_calibration_runbook.yaml meta:
# BCH-V1 + CALIB-V1 실행 런북 (저성능 LLM 완전 재현 가이드) has_code_implementation: true
# ──────────────────────────────────────────────────────────────────────────── code_path:
# 이 파일을 처음부터 끝까지 순서대로 따라 실행하면 - spec\27_bch_calibration_runbook.yaml
# 저성능 LLM도 동일한 결과(behavioral_coverage_pct=100%, divergence=0,
# llm_freedom_pct=0.0%)를 얻을 수 있다.
#
# 진정한 작업완료 기준:
# - BEHAVIORAL_COVERAGE_V1_OK (behavioral_coverage_pct=100%, divergence=0)
# - CALIBRATION_REGISTRY_WARN/OK (overclaimed=0, unregistered=0)
# - LFM_V1_OK (llm_freedom_pct=0.0%)
# - LLM_NARRATIVE_LOCK gate=PASS (softening_violations=0)
# - HONEST_PERFORMANCE_V1_WARN/OK (design_score_as_proof ≤1건)
# - full-gate EXIT=0 (53단계 파이프라인 전부 통과)
# ────────────────────────────────────────────────────────────────────────────
runbook_id: BCH_CALIBRATION_RUNBOOK_V1 runbook_id: BCH_CALIBRATION_RUNBOOK_V1
version: "2026-05-30" version: '2026-05-30'
objective: | objective: 'yaml 지침(spec/13_formula_registry.yaml)의 공식이 GAS(.gs) 및 Python 구현과
yaml 지침(spec/13_formula_registry.yaml)의 공식이 GAS(.gs) 및 Python 구현과
행위 수준에서 100% 일치하는지 검증하고, 하드코딩 임계값을 정직하게 관리하며, 행위 수준에서 100% 일치하는지 검증하고, 하드코딩 임계값을 정직하게 관리하며,
LLM의 가격·수량 자유계산 여지를 0으로 측정·폐쇄한다. LLM의 가격·수량 자유계산 여지를 0으로 측정·폐쇄한다.
"거짓 100%"를 제거하고 수치로 증빙 가능한 진짜 100%를 달성한다. "거짓 100%"를 제거하고 수치로 증빙 가능한 진짜 100%를 달성한다.
# ════════════════════════════════════════════════════════════════════════════ '
# PHASE 0 — 전제조건 확인
# ════════════════════════════════════════════════════════════════════════════
phase_0_prerequisites: phase_0_prerequisites:
description: "이 단계를 모두 만족해야 Phase 1을 시작할 수 있다." description: 이 단계를 모두 만족해야 Phase 1을 시작할 수 있다.
checks: checks:
- id: P0_1 - id: P0_1
command: "python --version" command: python --version
expect: "Python 3.10+" expect: Python 3.10+
note: "yaml, json, math, re, pathlib 사용. 외부 패키지: pyyaml(pip install pyyaml)" note: 'yaml, json, math, re, pathlib 사용. 외부 패키지: pyyaml(pip install pyyaml)'
- id: P0_2
- id: P0_2 command: node --version
command: "node --version" expect: v18+
expect: "v18+" note: GAS 패리티 러너(run_gas_golden_parity.js)에 필요
note: "GAS 패리티 러너(run_gas_golden_parity.js)에 필요" - id: P0_3
command: python -c "import yaml; print('yaml OK')"
- id: P0_3 expect: yaml OK
command: "python -c \"import yaml; print('yaml OK')\"" fail_action: pip install pyyaml
expect: "yaml OK" - id: P0_4
fail_action: "pip install pyyaml" command: ls spec/13_formula_registry.yaml spec/13b_harness_formulas.yaml
expect: 두 파일 모두 존재
- id: P0_4 note: 170개 공식 정의 파일
command: "ls spec/13_formula_registry.yaml spec/13b_harness_formulas.yaml" - id: P0_5
expect: "두 파일 모두 존재" command: ls gas_data_feed.gs gas_lib.gs gas_apex_alpha_watch.gs
note: "170개 공식 정의 파일" expect: 세 파일 모두 존재
note: GAS 구현 파일
- id: P0_5
command: "ls gas_data_feed.gs gas_lib.gs gas_apex_alpha_watch.gs"
expect: "세 파일 모두 존재"
note: "GAS 구현 파일"
# ════════════════════════════════════════════════════════════════════════════
# PHASE 1 — 행위기반 커버리지 하네스 (BCH-V1)
# ════════════════════════════════════════════════════════════════════════════
phase_1_behavioral_coverage: phase_1_behavioral_coverage:
description: | description: '"formula_id 문자열이 .gs에 등장한다" → "golden == Python미러 == GAS미러" 로 전환.
"formula_id 문자열이 .gs에 등장한다" → "golden == Python미러 == GAS미러" 로 전환.
발견된 분기(divergence)는 spec/13 기준으로 근본 정정. 발견된 분기(divergence)는 spec/13 기준으로 근본 정정.
'
ordered_steps: ordered_steps:
- id: S1_1_VERIFY_CONTRACT
name: 계약 파일 확인
command: cat spec/26_behavioral_coverage_contract.yaml | head -20
expect: 'behavioral_coverage_pct_min: 100.0'
note: 없으면 계약 파일 작성 필요 (spec/26_behavioral_coverage_contract.yaml)
- id: S1_2_VERIFY_GOLDEN_CASES
name: 골든케이스 파일 확인
command: "python -c \"\nimport yaml\nwith open('spec/formula_golden_cases_v2.yaml',\
\ encoding='utf-8') as f:\n d = yaml.safe_load(f)\nformulas = d.get('golden_cases_v2',\
\ [])\nprint(f'등록 공식 수: {len(formulas)}')\nfor f in formulas:\n cases = f.get('cases',\
\ [])\n n = sum(1 for c in cases if 'inputs' in c)\n print(f' {f[\\\"\
formula_id\\\"]}: {n}개 케이스')\n\"\n"
expect: 등록 공식 수 ≥ 22
fail_action: "spec/formula_golden_cases_v2.yaml 에 golden case 추가.\n각 case 형식:\n\
\ - id: 케이스ID\n inputs: {필드명: 값}\n expected: {출력필드: 기대값}\n tolerance:\
\ {수치필드: 허용오차}\n provenance: HAND_COMPUTED # 반드시 spec에서 손계산. .gs 역복사 금지.\n\
주의: expected 값을 .gs 출력에서 역복사하면 순환논리(REJECT).\n"
- id: S1_3_RUN_PY_MIRROR
name: Python 미러 검증
command: python tools/run_formula_golden_cases_v2.py
expect:
status_token: BEHAVIORAL_COVERAGE_PY_OK
behavioral_coverage_pct: 100.0
python_fail: 0
fail_code: BCH_PY_MIRROR_FAIL
fail_action: '출력에서 [FAIL] 공식을 찾아 python_function 로직을 spec/13 expression과 비교.
- id: S1_1_VERIFY_CONTRACT spec/13 expression이 맞고 Python이 틀린 경우 → Python 수정.
name: "계약 파일 확인"
command: "cat spec/26_behavioral_coverage_contract.yaml | head -20"
expect: "behavioral_coverage_pct_min: 100.0"
note: "없으면 계약 파일 작성 필요 (spec/26_behavioral_coverage_contract.yaml)"
- id: S1_2_VERIFY_GOLDEN_CASES Python이 맞고 golden expected가 틀린 경우 → golden case 수정(다시 손계산).
name: "골든케이스 파일 확인"
command: |
python -c "
import yaml
with open('spec/formula_golden_cases_v2.yaml', encoding='utf-8') as f:
d = yaml.safe_load(f)
formulas = d.get('golden_cases_v2', [])
print(f'등록 공식 수: {len(formulas)}')
for f in formulas:
cases = f.get('cases', [])
n = sum(1 for c in cases if 'inputs' in c)
print(f' {f[\"formula_id\"]}: {n}개 케이스')
"
expect: "등록 공식 수 ≥ 22"
fail_action: |
spec/formula_golden_cases_v2.yaml 에 golden case 추가.
각 case 형식:
- id: 케이스ID
inputs: {필드명: 값}
expected: {출력필드: 기대값}
tolerance: {수치필드: 허용오차}
provenance: HAND_COMPUTED # 반드시 spec에서 손계산. .gs 역복사 금지.
주의: expected 값을 .gs 출력에서 역복사하면 순환논리(REJECT).
- id: S1_3_RUN_PY_MIRROR 절대 "Python 출력 = expected" 방식의 역복사 금지.
name: "Python 미러 검증"
command: "python tools/run_formula_golden_cases_v2.py"
expect:
status_token: BEHAVIORAL_COVERAGE_PY_OK
behavioral_coverage_pct: 100.0
python_fail: 0
fail_code: BCH_PY_MIRROR_FAIL
fail_action: |
출력에서 [FAIL] 공식을 찾아 python_function 로직을 spec/13 expression과 비교.
spec/13 expression이 맞고 Python이 틀린 경우 → Python 수정.
Python이 맞고 golden expected가 틀린 경우 → golden case 수정(다시 손계산).
절대 "Python 출력 = expected" 방식의 역복사 금지.
- id: S1_4_RUN_GAS_PARITY '
name: "GAS 패리티 검증" - id: S1_4_RUN_GAS_PARITY
command: "node tools/run_gas_golden_parity.js" name: GAS 패리티 검증
expect: command: node tools/run_gas_golden_parity.js
status_token: GAS_PARITY_OK expect:
gas_fail: 0 status_token: GAS_PARITY_OK
fail_code: BCH_GAS_PARITY_FAIL gas_fail: 0
fail_action: | fail_code: BCH_GAS_PARITY_FAIL
출력에서 [GAS_FAIL] 또는 [GAS_CORRECT_PYTHON_WRONG] 확인. fail_action: '출력에서 [GAS_FAIL] 또는 [GAS_CORRECT_PYTHON_WRONG] 확인.
GAS_CORRECT_PYTHON_WRONG: GAS가 spec_correct → Python 수정 필요.
GAS_FAIL: GAS가 틀림 → gas_data_feed.gs / gas_lib.gs / gas_apex_alpha_watch.gs 수정.
수정 기준: 항상 spec/13_formula_registry.yaml의 expression.
- id: S1_5_3WAY_VALIDATE GAS_CORRECT_PYTHON_WRONG: GAS가 spec_correct → Python 수정 필요.
name: "3-way 동등성 게이트"
command: "python tools/validate_behavioral_coverage_v1.py --strict"
expect:
status_token: BEHAVIORAL_COVERAGE_V1_OK
behavioral_coverage_pct: 100.0
implementation_divergence_count: 0
fail_code: BEHAVIORAL_COVERAGE_V1_FAIL
fail_action: |
implementation_divergence_count > 0 이면:
- PYTHON_DIVERGES_FROM_SPEC: Python normalize_tick 등 → math.floor로 수정
- GAS_DIVERGES_FROM_GOLDEN: GAS 함수 → spec/13 expression 적용
divergence가 0이어도 coverage < 100%이면 golden case 부족 → S1_2로 돌아감.
- id: S1_6_WIRE_PIPELINE GAS_FAIL: GAS가 틀림 → gas_data_feed.gs / gas_lib.gs / gas_apex_alpha_watch.gs
name: "파이프라인 연결 확인" 수정.
command: "npm run validate-behavioral-coverage"
expect:
exit_code: 0
status_token: BEHAVIORAL_COVERAGE_V1_OK
fail_code: BCH_WIRING_FAIL
note: "package.json의 validate-behavioral-coverage 스크립트가 S1_3+S1_4+S1_5를 순서대로 실행"
수정 기준: 항상 spec/13_formula_registry.yaml의 expression.
'
- id: S1_5_3WAY_VALIDATE
name: 3-way 동등성 게이트
command: python tools/validate_behavioral_coverage_v1.py --strict
expect:
status_token: BEHAVIORAL_COVERAGE_V1_OK
behavioral_coverage_pct: 100.0
implementation_divergence_count: 0
fail_code: BEHAVIORAL_COVERAGE_V1_FAIL
fail_action: "implementation_divergence_count > 0 이면:\n - PYTHON_DIVERGES_FROM_SPEC:\
\ Python normalize_tick 등 → math.floor로 수정\n - GAS_DIVERGES_FROM_GOLDEN: GAS\
\ 함수 → spec/13 expression 적용\ndivergence가 0이어도 coverage < 100%이면 golden case\
\ 부족 → S1_2로 돌아감.\n"
- id: S1_6_WIRE_PIPELINE
name: 파이프라인 연결 확인
command: npm run validate-behavioral-coverage
expect:
exit_code: 0
status_token: BEHAVIORAL_COVERAGE_V1_OK
fail_code: BCH_WIRING_FAIL
note: package.json의 validate-behavioral-coverage 스크립트가 S1_3+S1_4+S1_5를 순서대로 실행
completion_gate: completion_gate:
command: "python tools/validate_behavioral_coverage_v1.py --strict" command: python tools/validate_behavioral_coverage_v1.py --strict
required_output: required_output:
behavioral_coverage_pct: "== 100.0" behavioral_coverage_pct: == 100.0
implementation_divergence_count: "== 0" implementation_divergence_count: == 0
evidence_artifact: "Temp/formula_behavioral_coverage_summary_v1.json" evidence_artifact: Temp/formula_behavioral_coverage_summary_v1.json
# ════════════════════════════════════════════════════════════════════════════
# PHASE 2 — 임계값 보정 레지스트리 (CALIB-V1)
# ════════════════════════════════════════════════════════════════════════════
phase_2_calibration_registry: phase_2_calibration_registry:
description: | description: '모든 하드코딩 임계값을 spec/calibration_registry.yaml 에 등록하고
모든 하드코딩 임계값을 spec/calibration_registry.yaml 에 등록하고
overclaimed(검증 안 된 값을 CALIBRATED로 위장) = 0 을 달성한다. overclaimed(검증 안 된 값을 CALIBRATED로 위장) = 0 을 달성한다.
'
ordered_steps: ordered_steps:
- id: S2_1_VERIFY_REGISTRY
- id: S2_1_VERIFY_REGISTRY name: 레지스트리 파일 확인
name: "레지스트리 파일 확인" command: "python -c \"\nimport yaml\nwith open('spec/calibration_registry.yaml',\
command: | \ encoding='utf-8') as f:\n d = yaml.safe_load(f)\nt = d.get('thresholds',\
python -c " \ [])\nprint(f'총 임계값: {len(t)}')\nby_src = {}\nfor e in t:\n s = e.get('source',\
import yaml \ 'EXPERT_PRIOR')\n by_src[s] = by_src.get(s, 0) + 1\nfor s, n in sorted(by_src.items()):\n\
with open('spec/calibration_registry.yaml', encoding='utf-8') as f: \ print(f' {s}: {n}')\n\"\n"
d = yaml.safe_load(f) expect: 총 임계값 ≥ 60
t = d.get('thresholds', []) - id: S2_2_RUN_REGISTRY_VALIDATE
print(f'총 임계값: {len(t)}') name: 레지스트리 검증
by_src = {} command: python tools/validate_calibration_registry_v1.py
for e in t: expect:
s = e.get('source', 'EXPERT_PRIOR') overclaimed_count: 0
by_src[s] = by_src.get(s, 0) + 1 unregistered_threshold_count: 0
for s, n in sorted(by_src.items()): status_token: CALIBRATION_REGISTRY_WARN or CALIBRATION_REGISTRY_OK
print(f' {s}: {n}') fail_code: CALIBRATION_REGISTRY_FAIL
" fail_action: "OVERCLAIMED: source=CALIBRATED 이면서 sample_n<30 → source를 PROVISIONAL\
expect: "총 임계값 ≥ 60" \ 로 변경.\n 절대 sample_n을 30으로 올려서 해결 금지 (실제 표본 없이 수치 조작).\nUNREGISTERED: .gs/.py\
\ 핫존에서 발견된 미등록 상수 → calibration_registry.yaml에 추가.\n 추가 형식:\n - id: 고유ID\n\
- id: S2_2_RUN_REGISTRY_VALIDATE \ value: 상수값\n unit: pct/ratio/count/etc\n source: EXPERT_PRIOR\
name: "레지스트리 검증" \ # 실측 없으면 반드시 EXPERT_PRIOR\n sample_n: 0\n owner_formula: 관련_FORMULA_ID\n\
command: "python tools/validate_calibration_registry_v1.py" \ gs_location: \"파일명:줄번호\"\n"
expect: - id: S2_3_BUILD_PRIORITY
overclaimed_count: 0 name: 보정 우선순위 연결
unregistered_threshold_count: 0 command: python tools/build_calibration_priority_v1.py
status_token: "CALIBRATION_REGISTRY_WARN or CALIBRATION_REGISTRY_OK" expect:
fail_code: CALIBRATION_REGISTRY_FAIL status_token: CALIBRATION_PRIORITY_OK
fail_action: | priority_count_min: 5
OVERCLAIMED: source=CALIBRATED 이면서 sample_n<30 → source를 PROVISIONAL 로 변경. fail_code: CALIBRATION_PRIORITY_FAIL
절대 sample_n을 30으로 올려서 해결 금지 (실제 표본 없이 수치 조작). note: alpha_feedback_loop_v2.json의 miss5_count 신호를 임계값 보정 우선순위에 연결
UNREGISTERED: .gs/.py 핫존에서 발견된 미등록 상수 → calibration_registry.yaml에 추가.
추가 형식:
- id: 고유ID
value: 상수값
unit: pct/ratio/count/etc
source: EXPERT_PRIOR # 실측 없으면 반드시 EXPERT_PRIOR
sample_n: 0
owner_formula: 관련_FORMULA_ID
gs_location: "파일명:줄번호"
- id: S2_3_BUILD_PRIORITY
name: "보정 우선순위 연결"
command: "python tools/build_calibration_priority_v1.py"
expect:
status_token: CALIBRATION_PRIORITY_OK
priority_count_min: 5
fail_code: CALIBRATION_PRIORITY_FAIL
note: "alpha_feedback_loop_v2.json의 miss5_count 신호를 임계값 보정 우선순위에 연결"
calibration_policy_enforcement: calibration_policy_enforcement:
- rule: "source=CALIBRATED 이려면 sample_n ≥ 30 AND backtest_doc이 있어야 한다" - rule: source=CALIBRATED 이려면 sample_n ≥ 30 AND backtest_doc이 있어야 한다
- rule: "실측 없는 임계값은 반드시 EXPERT_PRIOR 또는 PROVISIONAL" - rule: 실측 없는 임계값은 반드시 EXPERT_PRIOR 또는 PROVISIONAL
- rule: "overclaimed_count > 0 이면 CALIBRATION_REGISTRY_FAIL → 배포 차단" - rule: overclaimed_count > 0 이면 CALIBRATION_REGISTRY_FAIL → 배포 차단
calibration_path: calibration_path:
EXPERT_PRIOR: EXPERT_PRIOR:
description: "30년 현장경험 기반 초기값. 검증 없음." description: 30년 현장경험 기반 초기값. 검증 없음.
next_step: "표본 수집 → 30건 이상이면 PROVISIONAL 승격 심사" next_step: 표본 수집 → 30건 이상이면 PROVISIONAL 승격 심사
PROVISIONAL: PROVISIONAL:
description: "예비 검증(1-29건). 방향성 확인됨." description: 예비 검증(1-29건). 방향성 확인됨.
next_step: "30건 이상 + 실제 P&L 검증 → CALIBRATED" next_step: 30건 이상 + 실제 P&L 검증 → CALIBRATED
CALIBRATED: CALIBRATED:
description: "실측 표본 ≥30건 backtest 완료." description: 실측 표본 ≥30건 backtest 완료.
maintenance: "분기별 재검증. 시장 국면 변화 시 재보정" maintenance: 분기별 재검증. 시장 국면 변화 시 재보정
completion_gate: completion_gate:
command: "python tools/validate_calibration_registry_v1.py" command: python tools/validate_calibration_registry_v1.py
required_output: required_output:
overclaimed_count: "== 0" overclaimed_count: == 0
unregistered_threshold_count: "== 0" unregistered_threshold_count: == 0
evidence_artifact: "Temp/calibration_registry_v1.json" evidence_artifact: Temp/calibration_registry_v1.json
# ════════════════════════════════════════════════════════════════════════════
# PHASE 3 — LLM 자유도 측정·폐쇄 (LFM-V1)
# ════════════════════════════════════════════════════════════════════════════
phase_3_llm_freedom: phase_3_llm_freedom:
description: | description: '가격·수량을 LLM이 자유계산하는 여지를 0으로 측정하고,
가격·수량을 LLM이 자유계산하는 여지를 0으로 측정하고,
narrative 완화어휘(INVALID_SOFTENING)를 차단한다. narrative 완화어휘(INVALID_SOFTENING)를 차단한다.
'
ordered_steps: ordered_steps:
- id: S3_1_FREEDOM_VALIDATE
name: LLM 자유도 측정
command: python tools/validate_number_provenance_v1.py
expect:
status_token: LFM_V1_OK
llm_freedom_pct: 0.0
freedom_signals_count: 0
fail_code: LFM_V1_FAIL
fail_action: 'prompts/analysis_prompt.md 에서 ''직접 계산한다'' 문구를 찾아
- id: S3_1_FREEDOM_VALIDATE ''DATA_MISSING — 하네스 업데이트 필요'' 로 교체.
name: "LLM 자유도 측정"
command: "python tools/validate_number_provenance_v1.py"
expect:
status_token: LFM_V1_OK
llm_freedom_pct: 0.0
freedom_signals_count: 0
fail_code: LFM_V1_FAIL
fail_action: |
prompts/analysis_prompt.md 에서 '직접 계산한다' 문구를 찾아
'DATA_MISSING — 하네스 업데이트 필요' 로 교체.
harness 결측 시 LLM 직접계산 허용 문구 전면 삭제 (HS011).
- id: S3_2_NARRATIVE_LOCK harness 결측 시 LLM 직접계산 허용 문구 전면 삭제 (HS011).
name: "LLM 내러티브 잠금"
command: "python tools/build_llm_narrative_template_lock_v1.py"
expect:
gate: PASS
total_violations: 0
narrative_violations: 0
softening_violations: 0
fail_code: LLM_NARRATIVE_LOCK_FAIL
fail_action: |
[INVALID_NARRATIVE]: 금지어휘(같다/약간/괜찮다/곧 등) 섹션에서 제거.
[INVALID_SOFTENING]: BLOCK/SELL verdict 근방에서 완화어휘 제거.
완화어휘 패턴: 그래도/유연하게/장기관점재진입/고려가능/상황에따라/아직괜찮/지켜볼만
규칙: BLOCK verdict가 있으면 완화 해석 문장을 완전히 삭제.
'
- id: S3_2_NARRATIVE_LOCK
name: LLM 내러티브 잠금
command: python tools/build_llm_narrative_template_lock_v1.py
expect:
gate: PASS
total_violations: 0
narrative_violations: 0
softening_violations: 0
fail_code: LLM_NARRATIVE_LOCK_FAIL
fail_action: "[INVALID_NARRATIVE]: 금지어휘(같다/약간/괜찮다/곧 등) 섹션에서 제거.\n[INVALID_SOFTENING]:\
\ BLOCK/SELL verdict 근방에서 완화어휘 제거.\n 완화어휘 패턴: 그래도/유연하게/장기관점재진입/고려가능/상황에따라/아직괜찮/지켜볼만\n\
\ 규칙: BLOCK verdict가 있으면 완화 해석 문장을 완전히 삭제.\n"
completion_gate: completion_gate:
command: "python tools/validate_number_provenance_v1.py && python tools/build_llm_narrative_template_lock_v1.py" command: python tools/validate_number_provenance_v1.py && python tools/build_llm_narrative_template_lock_v1.py
required_output: required_output:
llm_freedom_pct: "== 0.0" llm_freedom_pct: == 0.0
softening_violations: "== 0" softening_violations: == 0
evidence_artifact: "Temp/llm_freedom_v1.json" evidence_artifact: Temp/llm_freedom_v1.json
# ════════════════════════════════════════════════════════════════════════════
# PHASE 4 — 정직 성과증빙 (HONEST-V1)
# ════════════════════════════════════════════════════════════════════════════
phase_4_honest_performance: phase_4_honest_performance:
description: | description: '설계점수(design_score)와 실측점수(actual_score)를 물리적으로 분리.
설계점수(design_score)와 실측점수(actual_score)를 물리적으로 분리.
sample_n < 30 이면 반드시 UNVALIDATED_DESIGN_SCORE 라벨. sample_n < 30 이면 반드시 UNVALIDATED_DESIGN_SCORE 라벨.
'
ordered_steps: ordered_steps:
- id: S4_1_HONEST_GUARD
name: 정직 성과증빙 하네스
command: python tools/build_honest_performance_guard_v1.py
expect:
status_token: HONEST_PERFORMANCE_V1_OK or HONEST_PERFORMANCE_V1_WARN
note: WARN은 표본 부족(sample<30)을 정직하게 공시하는 정상 상태
fail_code: HONEST_PERFORMANCE_V1_FAIL
fail_action: "design_score_as_proof 위반이 있으면:\n 해당 score를 보고서에 표시할 때 '[UNVALIDATED_DESIGN_SCORE:\
\ n=N]' 주석 필수.\n 'score=97.12(검증됨)' 형식 절대 금지.\nT+1/T+5 KPI가 목표 미달이면:\n 보정루프\
\ 로드맵(build_calibration_priority_v1.py 결과) 참조.\n '목표 달성' 선언 금지 — 수치 그대로 공시.\n"
- id: S4_2_KPI_TRACKING
name: T+1/T+5 KPI 추적
command: "python -c \"\nimport json\nwith open('Temp/honest_performance_guard_v1.json',\
\ encoding='utf-8') as f:\n d = json.load(f)\nfor k in d.get('kpi_tracker',\
\ []):\n status = 'OK' if k['status'] != 'BELOW_TARGET' else 'BELOW_TARGET'\n\
\ print(f\\\"{k['metric']}: {k['current']}% (target={k['target_min']}%) [{status}]\\\
\")\n\"\n"
expect: 출력 확인
note: 'T+5 35.86% → 50% 목표: 보정루프 4단계
- id: S4_1_HONEST_GUARD Step1: 표본 누적(30건)
name: "정직 성과증빙 하네스"
command: "python tools/build_honest_performance_guard_v1.py"
expect:
status_token: "HONEST_PERFORMANCE_V1_OK or HONEST_PERFORMANCE_V1_WARN"
note: "WARN은 표본 부족(sample<30)을 정직하게 공시하는 정상 상태"
fail_code: HONEST_PERFORMANCE_V1_FAIL
fail_action: |
design_score_as_proof 위반이 있으면:
해당 score를 보고서에 표시할 때 '[UNVALIDATED_DESIGN_SCORE: n=N]' 주석 필수.
'score=97.12(검증됨)' 형식 절대 금지.
T+1/T+5 KPI가 목표 미달이면:
보정루프 로드맵(build_calibration_priority_v1.py 결과) 참조.
'목표 달성' 선언 금지 — 수치 그대로 공시.
- id: S4_2_KPI_TRACKING Step2: ALEG_V2_GATE1_BLOCK_PCT 3% → 실측 최적값 PROVISIONAL 승격
name: "T+1/T+5 KPI 추적"
command: |
python -c "
import json
with open('Temp/honest_performance_guard_v1.json', encoding='utf-8') as f:
d = json.load(f)
for k in d.get('kpi_tracker', []):
status = 'OK' if k['status'] != 'BELOW_TARGET' else 'BELOW_TARGET'
print(f\"{k['metric']}: {k['current']}% (target={k['target_min']}%) [{status}]\")
"
expect: "출력 확인"
note: |
T+5 35.86% → 50% 목표: 보정루프 4단계
Step1: 표본 누적(30건)
Step2: ALEG_V2_GATE1_BLOCK_PCT 3% → 실측 최적값 PROVISIONAL 승격
Step3: DSD_V1 가중치 logistic regression 최적화
Step4: K2 분할비율 backtest → CALIBRATED
Step3: DSD_V1 가중치 logistic regression 최적화
Step4: K2 분할비율 backtest → CALIBRATED
'
completion_gate: completion_gate:
command: "python tools/build_honest_performance_guard_v1.py" command: python tools/build_honest_performance_guard_v1.py
required_output: required_output:
violation_count: "== 0 (UNVALIDATED 라벨 추가로 해소)" violation_count: == 0 (UNVALIDATED 라벨 추가로 해소)
evidence_artifact: "Temp/honest_performance_guard_v1.json" evidence_artifact: Temp/honest_performance_guard_v1.json
# ════════════════════════════════════════════════════════════════════════════
# PHASE 5 — 통합 게이트 + 파이프라인 확인
# ════════════════════════════════════════════════════════════════════════════
phase_5_integration: phase_5_integration:
description: "4-기둥 통합 실행 후 full-gate / daily-feedback-report 최종 통과 확인" description: 4-기둥 통합 실행 후 full-gate / daily-feedback-report 최종 통과 확인
ordered_steps: ordered_steps:
- id: S5_1_ENGINE_INTEGRITY
name: 통합 엔진 무결성
command: npm run validate-engine-integrity
expect:
exit_code: 0
behavioral_coverage_pct: 100.0
overclaimed_count: 0
llm_freedom_pct: 0.0
softening_violations: 0
fail_code: ENGINE_INTEGRITY_FAIL
note: 'validate-behavioral-coverage &&
- id: S5_1_ENGINE_INTEGRITY validate-calibration-registry &&
name: "통합 엔진 무결성"
command: "npm run validate-engine-integrity"
expect:
exit_code: 0
behavioral_coverage_pct: 100.0
overclaimed_count: 0
llm_freedom_pct: 0.0
softening_violations: 0
fail_code: ENGINE_INTEGRITY_FAIL
note: |
validate-behavioral-coverage &&
validate-calibration-registry &&
validate-llm-freedom &&
validate-narrative-lock &&
build-honest-performance-guard &&
build-calibration-priority
- id: S5_2_FULL_GATE validate-llm-freedom &&
name: "전체 파이프라인 확인"
command: "npm run full-gate"
expect:
exit_code: 0
note: "53단계 전부 통과. WARN_ONLY 항목(펀더멘털 미수집)은 허용."
fail_code: FULL_GATE_FAIL
fail_action: |
실패 단계를 단독 실행해 원인 파악:
npm run <실패_단계명>
HARNESS CONTEXT FAIL → validate_harness_context.py의 허용 enum 확인
validate-specs FAIL → RetirementAssetPortfolio.yaml spec_files에 신규 파일 등록
- id: S5_3_DAILY_FEEDBACK validate-narrative-lock &&
name: "일일 피드백 리포트 확인"
command: "npm run daily-feedback-report"
expect:
exit_code: 0
fail_code: DAILY_FEEDBACK_FAIL
build-honest-performance-guard &&
build-calibration-priority
'
- id: S5_2_FULL_GATE
name: 전체 파이프라인 확인
command: npm run full-gate
expect:
exit_code: 0
note: 53단계 전부 통과. WARN_ONLY 항목(펀더멘털 미수집)은 허용.
fail_code: FULL_GATE_FAIL
fail_action: "실패 단계를 단독 실행해 원인 파악:\n npm run <실패_단계명>\nHARNESS CONTEXT FAIL →\
\ validate_harness_context.py의 허용 enum 확인\nvalidate-specs FAIL → RetirementAssetPortfolio.yaml\
\ spec_files에 신규 파일 등록\n"
- id: S5_3_DAILY_FEEDBACK
name: 일일 피드백 리포트 확인
command: npm run daily-feedback-report
expect:
exit_code: 0
fail_code: DAILY_FEEDBACK_FAIL
completion_gate: completion_gate:
command: "npm run full-gate && npm run daily-feedback-report" command: npm run full-gate && npm run daily-feedback-report
required_output: required_output:
exit_code: "== 0 (both)" exit_code: == 0 (both)
evidence_artifact: "Temp/formula_behavioral_coverage_summary_v1.json" evidence_artifact: Temp/formula_behavioral_coverage_summary_v1.json
# ════════════════════════════════════════════════════════════════════════════
# PHASE 6 — 보정루프 (표본 누적 후 반복 실행)
# ════════════════════════════════════════════════════════════════════════════
phase_6_calibration_loop: phase_6_calibration_loop:
description: | description: '매 거래일 T+5 결과 수집 후 실행. 표본이 누적될수록
매 거래일 T+5 결과 수집 후 실행. 표본이 누적될수록
임계값을 EXPERT_PRIOR → PROVISIONAL → CALIBRATED 로 승격한다. 임계값을 EXPERT_PRIOR → PROVISIONAL → CALIBRATED 로 승격한다.
trigger: "매 거래일 장마감 후 (15:30 이후)" '
trigger: 매 거래일 장마감 후 (15:30 이후)
ordered_steps: ordered_steps:
- id: L1_UPDATE_HISTORY
- id: L1_UPDATE_HISTORY name: 평가 이력 업데이트
name: "평가 이력 업데이트" command: npm run update-evaluation-history
command: "npm run update-evaluation-history" note: proposal_evaluation_history.json 에 T+5 결과 추가
note: "proposal_evaluation_history.json 에 T+5 결과 추가" - id: L2_CHECK_SAMPLE_COUNT
name: 표본 수 확인
- id: L2_CHECK_SAMPLE_COUNT command: "python -c \"\nimport json\nwith open('Temp/calibration_priority_v1.json',\
name: "표본 수 확인" \ encoding='utf-8') as f:\n d = json.load(f)\nprint('cases_analyzed:', d.get('cases_analyzed',\
command: | \ 0))\nprint('miss5_count:', d.get('miss5_count', 0))\ntop3 = d.get('priority_list',\
python -c " \ [])[:3]\nfor p in top3:\n print(f' [{p[\\\"urgency_score\\\"]}] {p[\\\"\
import json calibration_id\\\"]}: value={p[\\\"current_value\\\"]} n={p[\\\"sample_n\\\"\
with open('Temp/calibration_priority_v1.json', encoding='utf-8') as f: ]}')\n\"\n"
d = json.load(f) note: cases_analyzed ≥ 30이면 최우선 임계값 PROVISIONAL 승격 심사
print('cases_analyzed:', d.get('cases_analyzed', 0)) - id: L3_CALIBRATION_CANDIDATE_REVIEW
print('miss5_count:', d.get('miss5_count', 0)) name: 보정 후보 심사 (cases ≥ 30 시)
top3 = d.get('priority_list', [])[:3] trigger_condition: cases_analyzed >= 30
for p in top3: manual_action: "1. ALEG_V2_GATE1_BLOCK_PCT(3%) 검증:\n - late_chase_attribution_v1.json\
print(f' [{p[\"urgency_score\"]}] {p[\"calibration_id\"]}: value={p[\"current_value\"]} n={p[\"sample_n\"]}') \ 의 chase_entry_rate 확인\n - velocity_1d ≥ 3%에서 진입한 케이스의 T+5 승률 계산\n - 현재\
" \ 3%보다 낮은 임계값이 더 효과적이면 새 값 제안\n2. 새 값 제안 후:\n - calibration_registry.yaml의\
note: "cases_analyzed ≥ 30이면 최우선 임계값 PROVISIONAL 승격 심사" \ source를 PROVISIONAL로 변경\n - sample_n에 실제 표본 수 기재\n - last_calibrated:\
\ 오늘 날짜\n3. 변경 후 반드시 npm run validate-engine-integrity 재실행\n"
- id: L3_CALIBRATION_CANDIDATE_REVIEW - id: L4_RUN_FULL_GATE
name: "보정 후보 심사 (cases ≥ 30 시)" name: 변경 후 전체 검증
trigger_condition: "cases_analyzed >= 30" command: npm run full-gate
manual_action: | expect:
1. ALEG_V2_GATE1_BLOCK_PCT(3%) 검증: exit_code: 0
- late_chase_attribution_v1.json 의 chase_entry_rate 확인
- velocity_1d ≥ 3%에서 진입한 케이스의 T+5 승률 계산
- 현재 3%보다 낮은 임계값이 더 효과적이면 새 값 제안
2. 새 값 제안 후:
- calibration_registry.yaml의 source를 PROVISIONAL로 변경
- sample_n에 실제 표본 수 기재
- last_calibrated: 오늘 날짜
3. 변경 후 반드시 npm run validate-engine-integrity 재실행
- id: L4_RUN_FULL_GATE
name: "변경 후 전체 검증"
command: "npm run full-gate"
expect: {exit_code: 0}
calibration_escalation_criteria: calibration_escalation_criteria:
PROVISIONAL: PROVISIONAL:
condition: "sample_n >= 10 AND 방향성 확인" condition: sample_n >= 10 AND 방향성 확인
CALIBRATED: CALIBRATED:
condition: "sample_n >= 30 AND 실제 P&L backtest 완료 AND 이전 임계값 대비 명확한 개선" condition: sample_n >= 30 AND 실제 P&L backtest 완료 AND 이전 임계값 대비 명확한 개선
required_doc: "backtest 결과 노트 (날짜, 표본 수, 이전값, 신규값, 성과 비교)" required_doc: backtest 결과 노트 (날짜, 표본 수, 이전값, 신규값, 성과 비교)
# ════════════════════════════════════════════════════════════════════════════
# 거부 조건 (Reject Conditions) — 어떤 상황에서도 적용
# ════════════════════════════════════════════════════════════════════════════
reject_conditions: reject_conditions:
- "behavioral_coverage_pct < 100% 인데 '커버리지 100% 달성' 선언" - behavioral_coverage_pct < 100% 인데 '커버리지 100% 달성' 선언
- "golden expected 값을 .gs 출력에서 역복사 (순환논리)" - golden expected 값을 .gs 출력에서 역복사 (순환논리)
- "임계값을 실측 없이 source=CALIBRATED로 기재 (overclaimed)" - 임계값을 실측 없이 source=CALIBRATED로 기재 (overclaimed)
- "LLM이 가격/수량을 spec 등록 공식 없이 즉석 계산" - LLM이 가격/수량을 spec 등록 공식 없이 즉석 계산
- "rebound_efficiency_score 등 설계점수를 '검증된 성과'로 서술 (UNVALIDATED 라벨 없이)" - rebound_efficiency_score 등 설계점수를 '검증된 성과'로 서술 (UNVALIDATED 라벨 없이)
- "T+1/T+5 목표 미달 상태에서 '예측 정확도 100%' 선언" - T+1/T+5 목표 미달 상태에서 '예측 정확도 100%' 선언
- "divergence_count > 0 상태에서 '구현 일치' 선언" - divergence_count > 0 상태에서 '구현 일치' 선언
- "sample_n < 30인 임계값을 '보정완료'로 처리" - sample_n < 30인 임계값을 '보정완료'로 처리
# ════════════════════════════════════════════════════════════════════════════
# 현재 달성 현황 (2026-06-21 재검증 — WBS-7.2)
# ════════════════════════════════════════════════════════════════════════════
# 주의: 아래 current_status_2026_05_30 블록은 그 날짜 기준 정적 스냅샷이며,
# 이후 갱신되지 않은 채 docs/ROADMAP_WBS.md 등에서 "현재 상태"로 인용되어
# 서로 다른 시점의 T+5 수치(54.76%/35.86%)가 혼재하는 문제를 일으켰다.
# Temp/honest_performance_guard_v1.json(생성: 2026-06-14)과
# Temp/prediction_accuracy_harness_v2.json(생성: 2026-06-21, 7일 더 최신)을
# 직접 재확인한 결과는 다음과 같다 — 이 블록을 단일 진실원천으로 삼는다.
current_status_2026_06_21: current_status_2026_06_21:
source_of_truth: "Temp/prediction_accuracy_harness_v2.json (as_of_date=2026-06-21, 가장 최신)" source_of_truth: Temp/prediction_accuracy_harness_v2.json (as_of_date=2026-06-21,
t1_match_rate_pct: 52.94 # sample=68, decisive_sample=53, rate_decisive=67.92 가장 최신)
t5_match_rate_pct: null # sample=0 — INSUFFICIENT_SAMPLES. honest_performance_guard_v1.json(2026-06-14)의 t1_match_rate_pct: 52.94
# 35.86%는 7일 전 스냅샷이며 표본이 0으로 줄어 더 이상 유효하지 않음. t5_match_rate_pct: null
t5_sample_regression_note: > t5_sample_regression_note: 'cases_analyzed가 141건(2026-05-30 기준)에서 t5_sample=0(2026-06-21)으로
cases_analyzed가 141건(2026-05-30 기준)에서 t5_sample=0(2026-06-21)으로 감소했다. 감소했다. evaluation_methodology가 ACTIVE_PASSIVE_SPLIT_V1_INCONCLUSIVE_EXCLUDED로 변경되며
evaluation_methodology가 ACTIVE_PASSIVE_SPLIT_V1_INCONCLUSIVE_EXCLUDED로 변경되며
inconclusive/replay 표본이 제외된 것으로 추정 — 근본 원인은 별도 조사 필요(WBS-7.2 잔여 항목). inconclusive/replay 표본이 제외된 것으로 추정 — 근본 원인은 별도 조사 필요(WBS-7.2 잔여 항목).
calibration_registry_total_thresholds: 190 # spec/calibration_registry.yaml 직접 집계 (구문서의 70은 stale)
'
calibration_registry_total_thresholds: 190
calibration_registry_expert_prior_count: 59 calibration_registry_expert_prior_count: 59
calibration_registry_calibrated_count: 0 calibration_registry_calibrated_count: 0
rule: "이 문서를 인용할 때는 항상 as_of_date를 동반 표기하고, 아래 5/30 스냅샷을 '현재'로 인용하지 않는다." rule: 이 문서를 인용할 때는 항상 as_of_date를 동반 표기하고, 아래 5/30 스냅샷을 '현재'로 인용하지 않는다.
# ════════════════════════════════════════════════════════════════════════════
# 과거 달성 현황 (2026-05-30, 역사적 스냅샷 — "현재"로 인용 금지)
# ════════════════════════════════════════════════════════════════════════════
current_status_2026_05_30: current_status_2026_05_30:
phase_1_bch: COMPLETE phase_1_bch: COMPLETE
behavioral_coverage_pct: 100.0 behavioral_coverage_pct: 100.0
gas_parity_cases: 63 gas_parity_cases: 63
implementation_divergence_count: 0 implementation_divergence_count: 0
bugs_fixed: bugs_fixed:
- "normalize_tick: round() → math.floor() (Python-GAS divergence 제거)" - 'normalize_tick: round() → math.floor() (Python-GAS divergence 제거)'
- "PROFIT_LOCK_STAGE 단계명 7개 spec 일치 정정 (GAS calcPrices_)" - PROFIT_LOCK_STAGE 단계명 7개 spec 일치 정정 (GAS calcPrices_)
- "validate_harness_context.py VALID_PROFIT_LOCK_STAGES 신규 명칭 추가" - validate_harness_context.py VALID_PROFIT_LOCK_STAGES 신규 명칭 추가
- "RetirementAssetPortfolio.yaml spec_files 신규 3파일 등록" - RetirementAssetPortfolio.yaml spec_files 신규 3파일 등록
phase_2_calib: COMPLETE phase_2_calib: COMPLETE
total_thresholds: 70 total_thresholds: 70
overclaimed_count: 0 overclaimed_count: 0
unregistered_count: 0 unregistered_count: 0
expert_prior_count: 61 # 정직하게 공시됨 expert_prior_count: 61
phase_3_lfm: COMPLETE phase_3_lfm: COMPLETE
llm_freedom_pct: 0.0 llm_freedom_pct: 0.0
softening_violations: 0 softening_violations: 0
prompt_freedom_risks_removed: 4 prompt_freedom_risks_removed: 4
phase_4_honest: COMPLETE phase_4_honest: COMPLETE
design_score_labeled_unvalidated: true design_score_labeled_unvalidated: true
t1_match_rate_pct: 47.28 t1_match_rate_pct: 47.28
t5_match_rate_pct: 35.86 t5_match_rate_pct: 35.86
target_t5: 55.0 target_t5: 55.0
phase_5_integration: COMPLETE phase_5_integration: COMPLETE
full_gate_exit: 0 full_gate_exit: 0
daily_feedback_exit: 0 daily_feedback_exit: 0
phase_6_calibration_loop: IN_PROGRESS phase_6_calibration_loop: IN_PROGRESS
cases_analyzed: 141 cases_analyzed: 141
miss5_count: 51 miss5_count: 51
next_milestone: "cases_analyzed=30 달성 후 ALEG_V2_GATE1_BLOCK_PCT 보정 심사" next_milestone: cases_analyzed=30 달성 후 ALEG_V2_GATE1_BLOCK_PCT 보정 심사
automation_entrypoints: automation_entrypoints:
gitea_schedule: ".gitea/workflows/calibration_backlog.yml" gitea_schedule: .gitea/workflows/calibration_backlog.yml
npm_script: "npm run ops:calibration-backlog" npm_script: npm run ops:calibration-backlog
generated_artifacts: generated_artifacts:
- Temp/calibration_priority_v1.json - Temp/calibration_priority_v1.json
- Temp/calibration_change_ledger_v4.json - Temp/calibration_change_ledger_v4.json
- Temp/calibration_review_report_v1.json - Temp/calibration_review_report_v1.json
- Temp/calibration_review_report_v1.md - Temp/calibration_review_report_v1.md
- Temp/calibration_approval_list_v1.json - Temp/calibration_approval_list_v1.json
- Temp/calibration_approval_list_v1.md - Temp/calibration_approval_list_v1.md
- Temp/calibration_registry_v1.json - Temp/calibration_registry_v1.json
promotion_rules: promotion_rules:
provisional: "sample_n >= 10 AND direction confirmed AND change_ledger entry exists" provisional: sample_n >= 10 AND direction confirmed AND change_ledger entry
calibrated: "sample_n >= 30 AND backtest_doc exists AND validator overclaimed_count == 0" exists
calibrated: sample_n >= 30 AND backtest_doc exists AND validator overclaimed_count
== 0
spec/29_backtest_harness_contract.yaml
+100 -105
View File
@@ -1,105 +1,94 @@
# spec/29 — 백테스트 · Walk-forward 하네스 계약 (BACKTEST_HARNESS_V1)
#
# 목적: 전략이 과거 데이터에만 맞춰진 과최적화인지, 실제 운용 중 의미있는 예측력이 있는지
# 수치로 검증한다. 미충족 항목은 추정·날조하지 않고 insufficient_data로 표기한다.
#
# 계층: 감사·진단 계약(spec/28과 동급). GAS 런타임·주문 생성에 개입 없음.
# 구현: Python 계층(tools/*.py). 실측 누적 데이터가 없는 구간은 채울 수 없음을 명시.
meta: meta:
formula_id: BACKTEST_HARNESS_V1 formula_id: BACKTEST_HARNESS_V1
version: "2026-05-31" version: '2026-05-31'
python_tool: "tools/build_yaml_code_coverage_v1.py (커버리지), tools/build_engine_audit_v1.py (집계)" python_tool: tools/build_yaml_code_coverage_v1.py (커버리지), tools/build_engine_audit_v1.py
(집계)
sources: sources:
- Temp/prediction_accuracy_harness_v2.json # T+1/T+5/T+20 정확도 - Temp/prediction_accuracy_harness_v2.json
- Temp/outcome_quality_score_v1.json # 운용 성과 질 점수 - Temp/outcome_quality_score_v1.json
- Temp/operational_alpha_calibration_v2.json # 알파 보정 - Temp/operational_alpha_calibration_v2.json
- Temp/proposal_evaluation_history.json # 제안-결과 이력 - Temp/proposal_evaluation_history.json
has_code_implementation: true
# ── 현재 실측 가능 지표 (2026-05-31 기준) ──────────────────────────────── code_path:
- spec\29_backtest_harness_contract.yaml
current_metrics: current_metrics:
direction_accuracy: direction_accuracy:
t1_op_rate: t1_op_rate:
value: 50.37 value: 50.37
n_sample: 546 n_sample: 546
unit: percent unit: percent
interpretation: "동전던지기(50%) 수준 — 단기 방향 예측력 불충분" interpretation: 동전던지기(50%) 수준 — 단기 방향 예측력 불충분
t5_op_rate: t5_op_rate:
value: 73.24 value: 73.24
n_sample: 161 n_sample: 161
unit: percent unit: percent
method: "decisive 케이스만(passive/ambiguous 제외). PREDICTION_ACCURACY_HARNESS_V2" method: decisive 케이스만(passive/ambiguous 제외). PREDICTION_ACCURACY_HARNESS_V2
interpretation: "T+5 능동 결정 케이스 73%. 전체 포함 레거시=31.94% — 표본 정의 혼용 금지" interpretation: T+5 능동 결정 케이스 73%. 전체 포함 레거시=31.94% — 표본 정의 혼용 금지
t5_legacy_rate: t5_legacy_rate:
value: 31.94 value: 31.94
n_sample: "not_available" n_sample: not_available
unit: percent unit: percent
interpretation: "전체 평가 윈도우 비율(거시이벤트 미제외). t5_op_rate와 다른 지표." interpretation: 전체 평가 윈도우 비율(거시이벤트 미제외). t5_op_rate와 다른 지표.
t20_op_rate: t20_op_rate:
value: insufficient_data value: insufficient_data
n_sample: 0 n_sample: 0
unit: percent unit: percent
interpretation: "T+20 실현 표본 0건 — 장기 예측력 검증 불가. t5_operational_proxy=73.24 사용 중(추정)" interpretation: T+20 실현 표본 0건 — 장기 예측력 검증 불가. t5_operational_proxy=73.24 사용
중(추정)
window_90d_rate: window_90d_rate:
value: 31.94 value: 31.94
n_sample: "not_available" n_sample: not_available
unit: percent unit: percent
interpretation: "최근 90일 창 일치율. 낮음." interpretation: 최근 90일 창 일치율. 낮음.
outcome_quality: outcome_quality:
score: 84.43 score: 84.43
gate: CAUTION_MODE gate: CAUTION_MODE
t20_effective_rate: 73.24 t20_effective_rate: 73.24
t20_source: t5_operational_proxy # 실측 아님 — estimated=true t20_source: t5_operational_proxy
t5_decisive_count: 161 t5_decisive_count: 161
basis_note: "t20는 실측이 아니라 t5 proxy. 실측 T+20 누적 전까지 estimated." basis_note: t20는 실측이 아니라 t5 proxy. 실측 T+20 누적 전까지 estimated.
walk_forward: walk_forward:
status: insufficient_data status: insufficient_data
reason: > reason: 'Walk-forward 검증을 위해 필요한 in-sample/out-of-sample 분리, 기간별 성과 비교, slippage/cost
Walk-forward 검증을 위해 필요한 in-sample/out-of-sample 분리, 반영 데이터가 없음. backfill_eod_replay_history.py를 통해 이력 재현 시 채울 수 있음.
기간별 성과 비교, slippage/cost 반영 데이터가 없음.
backfill_eod_replay_history.py를 통해 이력 재현 시 채울 수 있음.
# ── 정의되어야 하나 현재 측정 불가한 지표 ───────────────────────────────── '
missing_metrics: missing_metrics:
CAGR: CAGR:
status: insufficient_data status: insufficient_data
required_data: "1년 이상 완전 실현 손익 이력" required_data: 1년 이상 완전 실현 손익 이력
sharpe_ratio: sharpe_ratio:
status: insufficient_data status: insufficient_data
required_data: "일별 수익률 시계열 + 무위험수익률" required_data: 일별 수익률 시계열 + 무위험수익률
sortino_ratio: sortino_ratio:
status: insufficient_data status: insufficient_data
required_data: "일별 하락 편차 시계열" required_data: 일별 하락 편차 시계열
max_drawdown: max_drawdown:
status: insufficient_data status: insufficient_data
required_data: "계좌 고점 추적 이력. portfolio_peak_krw 필드 존재하나 historical 없음" required_data: 계좌 고점 추적 이력. portfolio_peak_krw 필드 존재하나 historical 없음
calmar_ratio: calmar_ratio:
status: insufficient_data status: insufficient_data
required_data: "CAGR / MDD" required_data: CAGR / MDD
win_rate: win_rate:
status: insufficient_data status: insufficient_data
required_data: "청산 완료 거래 이력. backdata에 MAE/MFE/pnl 모두 공란" required_data: 청산 완료 거래 이력. backdata에 MAE/MFE/pnl 모두 공란
profit_factor: profit_factor:
status: insufficient_data status: insufficient_data
required_data: "총 이익 / 총 손실 (실현 기준)" required_data: 총 이익 / 총 손실 (실현 기준)
average_win_loss_ratio: average_win_loss_ratio:
status: insufficient_data status: insufficient_data
required_data: "실현 수익/손실 건별 데이터" required_data: 실현 수익/손실 건별 데이터
slippage_impact: slippage_impact:
status: insufficient_data status: insufficient_data
required_data: "체결 가격 vs 지정가 괴리 이력" required_data: 체결 가격 vs 지정가 괴리 이력
transaction_cost_impact: transaction_cost_impact:
status: insufficient_data status: insufficient_data
required_data: "수수료·세금 반영 순수익 이력" required_data: 수수료·세금 반영 순수익 이력
hit_rate_by_horizon: hit_rate_by_horizon:
scalp: insufficient_data scalp: insufficient_data
short_term: insufficient_data short_term: insufficient_data
mid_term: insufficient_data mid_term: insufficient_data
long_term: insufficient_data long_term: insufficient_data
# ── 측정 가능한 회귀 지표 (현재 구현됨) ──────────────────────────────────
measurable_now: measurable_now:
yaml_to_code_coverage_ratio: yaml_to_code_coverage_ratio:
value: 1.0 value: 1.0
@@ -107,43 +96,37 @@ measurable_now:
golden_test_coverage_ratio: golden_test_coverage_ratio:
value: 0.2337 value: 0.2337
source: Temp/yaml_code_coverage_v1.json source: Temp/yaml_code_coverage_v1.json
note: "43/184 공식 — golden 테스트 확대 필요" note: 43/184 공식 — golden 테스트 확대 필요
decision_reproducibility_score: decision_reproducibility_score:
value: 1.0 value: 1.0
method: "build_engine_audit_v1.py 10회 실행 byte-identical" method: build_engine_audit_v1.py 10회 실행 byte-identical
llm_dependency_ratio: llm_dependency_ratio:
value: 0.0 value: 0.0
source: Temp/llm_freedom_v1.json source: Temp/llm_freedom_v1.json
schema_validity_score: schema_validity_score:
value: 95.5 value: 95.5
source: Temp/data_quality_reconciliation_v1.json source: Temp/data_quality_reconciliation_v1.json
# ── 목표치 (충족 시 PASS 판정) ──────────────────────────────────────────
targets: targets:
t1_op_rate_min: 55 # 현재 50.37 — 미달 t1_op_rate_min: 55
t5_op_rate_min: 60 # 현재 73.24 — 충족(주의: decisive 케이스 기준) t5_op_rate_min: 60
t20_op_rate_min: 55 # 현재 insufficient_data t20_op_rate_min: 55
win_rate_min: 50 # 현재 insufficient_data win_rate_min: 50
max_drawdown_max_pct: 20 # 현재 측정 불가 max_drawdown_max_pct: 20
yaml_to_code_coverage: 1.0 # 충족 yaml_to_code_coverage: 1.0
golden_coverage_min: 0.5 # 현재 0.23 — 미달 golden_coverage_min: 0.5
# ── 과최적화 경계 지표 ────────────────────────────────────────────────────
overfit_risk: overfit_risk:
in_sample_vs_oos_gap: in_sample_vs_oos_gap:
status: insufficient_data status: insufficient_data
note: "in-sample / out-of-sample 분리 없음" note: in-sample / out-of-sample 분리 없음
regime_dependency: regime_dependency:
note: > note: '현재 포트폴리오는 RISK_ON 국면에 집중(SHORT 71.4% vs 25% 한도 위반). 단일 국면 의존도 과다 — regime
현재 포트폴리오는 RISK_ON 국면에 집중(SHORT 71.4% vs 25% 한도 위반). 다양화 필요.
단일 국면 의존도 과다 — regime 다양화 필요.
sample_size_warning:
t1: "n=546 — 통계적으로 유의하나 변동 큼"
t5: "n=161 — 최소 수준. 더 많은 누적 필요"
t20: "n=0 — 미충족"
# ── 실측 표본 백필 의무화 (OPERATIONAL_SAMPLE_BACKFILL_V1) ───────────────── '
# [SCAFFOLDED_PENDING_LIVE_DATA: operational_t5_sample_count=0, target>=30] sample_size_warning:
t1: n=546 — 통계적으로 유의하나 변동 큼
t5: n=161 — 최소 수준. 더 많은 누적 필요
t20: n=0 — 미충족
operational_sample_backfill: operational_sample_backfill:
formula_id: OPERATIONAL_SAMPLE_BACKFILL_V1 formula_id: OPERATIONAL_SAMPLE_BACKFILL_V1
status: SCAFFOLDED_PENDING_LIVE_DATA status: SCAFFOLDED_PENDING_LIVE_DATA
@@ -152,42 +135,54 @@ operational_sample_backfill:
current_replay_sample_count: 510 current_replay_sample_count: 510
target_operational_t5_sample: 30 target_operational_t5_sample: 30
target_operational_t20_sample: 30 target_operational_t20_sample: 30
rationale: > rationale: 'live=0, paper=0, op_t20=0. REPLAY 510건은 예측력 증거가 못 된다(미래정보 누수 위험). 실제
live=0, paper=0, op_t20=0. REPLAY 510건은 예측력 증거가 못 된다(미래정보 누수 위험). 제안→실측 결과 연결 고리가 끊겨 있다.
실제 제안→실측 결과 연결 고리가 끊겨 있다.
implementation_steps:
- step: 1
desc: "proposal_evaluation_history.json의 각 과거 제안(BUY/SELL/TRIM)에 entry_date, entry_price를 고정 기록"
status: NOT_STARTED
- step: 2
desc: "T+5/T+20 경과 시 data_feed의 종가로 realized_return_pct 채움 (미래 데이터 사용 금지: 평가일 ≤ 오늘)"
status: NOT_STARTED
- step: 3
desc: "origin 태그를 LIVE/PAPER/REPLAY로 명확히 분리. 예측력 지표는 LIVE+PAPER만 집계"
status: NOT_STARTED
- step: 4
desc: "operational_t5_sample_count, operational_t20_sample_count 매 사이클 갱신"
status: NOT_STARTED
- step: 5
desc: "표본 < 30인 동안 모든 예측력 지표에 '[UNVALIDATED_LIVE: n={n}]' 라벨 부착, PASS 금지(WATCH)"
status: ACTIVE_GUARD
outputs:
- live_trade_outcome_ledger_v1.json # LIVE/PAPER 채움
- prediction_accuracy_harness_v5.json.operational_t5_sample
numeric_acceptance:
operational_t5_sample_count: {op: ">=", target: 30, current: 0, blocking: true}
replay_contamination: {op: "==", target: 0, current: 0, note: "REPLAY 표본 예측지표 집계 혼입 금지"}
future_leak: {op: "==", target: 0, note: "모든 realized_return 평가일 <= capture_date"}
python_tools:
- tools/backfill_eod_replay_history.py
- tools/build_live_trade_outcome_ledger_v1.py
gs_coverage: "gas_apex_runtime_core.gs:evaluateOperationalOutcomeBatch_()"
validator: "tools/validate_outcome_eval_window.py --no-future-leak --min-live 30"
unvalidated_label: "[UNVALIDATED_LIVE: n=0 < 30]"
# ── 금지 사항 ───────────────────────────────────────────────────────────── '
implementation_steps:
- step: 1
desc: proposal_evaluation_history.json의 각 과거 제안(BUY/SELL/TRIM)에 entry_date, entry_price를
고정 기록
status: NOT_STARTED
- step: 2
desc: 'T+5/T+20 경과 시 data_feed의 종가로 realized_return_pct 채움 (미래 데이터 사용 금지: 평가일
≤ 오늘)'
status: NOT_STARTED
- step: 3
desc: origin 태그를 LIVE/PAPER/REPLAY로 명확히 분리. 예측력 지표는 LIVE+PAPER만 집계
status: NOT_STARTED
- step: 4
desc: operational_t5_sample_count, operational_t20_sample_count 매 사이클 갱신
status: NOT_STARTED
- step: 5
desc: '표본 < 30인 동안 모든 예측력 지표에 ''[UNVALIDATED_LIVE: n={n}]'' 라벨 부착, PASS 금지(WATCH)'
status: ACTIVE_GUARD
outputs:
- live_trade_outcome_ledger_v1.json
- prediction_accuracy_harness_v5.json.operational_t5_sample
numeric_acceptance:
operational_t5_sample_count:
op: '>='
target: 30
current: 0
blocking: true
replay_contamination:
op: ==
target: 0
current: 0
note: REPLAY 표본 예측지표 집계 혼입 금지
future_leak:
op: ==
target: 0
note: 모든 realized_return 평가일 <= capture_date
python_tools:
- tools/backfill_eod_replay_history.py
- tools/build_live_trade_outcome_ledger_v1.py
gs_coverage: gas_apex_runtime_core.gs:evaluateOperationalOutcomeBatch_()
validator: tools/validate_outcome_eval_window.py --no-future-leak --min-live 30
unvalidated_label: '[UNVALIDATED_LIVE: n=0 < 30]'
prohibitions: prohibitions:
- "insufficient_data 지표를 추정값으로 대체해 투자 판단에 사용 금지(AGENTS.md §0.3)" - insufficient_data 지표를 추정값으로 대체해 투자 판단에 사용 금지(AGENTS.md §0.3)
- "t5_op_rate(73%)와 window_90d_rate(31%)를 동일 지표로 혼용 금지" - t5_op_rate(73%)와 window_90d_rate(31%)를 동일 지표로 혼용 금지
- "t20_op_rate=t5_operational_proxy를 실측 T+20으로 표기 금지 (estimated=true 필수)" - t20_op_rate=t5_operational_proxy를 실측 T+20으로 표기 금지 (estimated=true 필수)
- "CAGR/Sharpe/MDD가 없는 상태에서 '검증된 전략' 단정 금지" - CAGR/Sharpe/MDD가 없는 상태에서 '검증된 전략' 단정 금지
spec/31_low_capability_llm_response_contract.yaml
+18 -18
View File
@@ -1,27 +1,27 @@
meta:
has_code_implementation: true
code_path:
- spec\31_low_capability_llm_response_contract.yaml
schema_version: low_capability_llm_response_contract.v1 schema_version: low_capability_llm_response_contract.v1
formula_id: LLM_NARRATIVE_TEMPLATE_LOCK_V1 formula_id: LLM_NARRATIVE_TEMPLATE_LOCK_V1
purpose: > purpose: '저성능 LLM도 동일한 JSON 입력에서 동일한 서술 결과를 내도록 응답 섹션과 금지 섹션을 고정한다.
저성능 LLM도 동일한 JSON 입력에서 동일한 서술 결과를 내도록
응답 섹션과 금지 섹션을 고정한다.
'
required_sections: required_sections:
- source_summary - source_summary
- fail_codes - fail_codes
- allowed_actions - allowed_actions
- blocked_actions - blocked_actions
- todo_yaml - todo_yaml
- no_order_notice - no_order_notice
forbidden_sections_when_blocked: forbidden_sections_when_blocked:
- hts_order_table - hts_order_table
- new_buy_recommendation - new_buy_recommendation
- freeform_target_price - freeform_target_price
copy_exact_rules: copy_exact_rules:
- "All numeric values must be copied from JSON with json_path." - All numeric values must be copied from JSON with json_path.
- "No unregistered formula names may be invented." - No unregistered formula names may be invented.
- "If HTS_READY is false, render shadow ledger only." - If HTS_READY is false, render shadow ledger only.
output_constraints: output_constraints:
language: ko-KR language: ko-KR
allow_freeform_numbers: false allow_freeform_numbers: false
spec/33_execution_precedence_lock.yaml
+6 -2
View File
@@ -1,6 +1,10 @@
meta:
has_code_implementation: true
code_path:
- spec\33_execution_precedence_lock.yaml
schema_version: 2026-06-03-execution-precedence-lock-v1 schema_version: 2026-06-03-execution-precedence-lock-v1
formula_id: FINAL_EXECUTION_PRECEDENCE_LOCK_V1 formula_id: FINAL_EXECUTION_PRECEDENCE_LOCK_V1
purpose: 최종 HTS 권한과 child internal allowed 분리. purpose: 최종 HTS 권한과 child internal allowed 분리.
rules: rules:
- global_execution_gate != HTS_READY 이면 child execution_allowed는 THEORETICAL_ONLY - global_execution_gate != HTS_READY 이면 child execution_allowed는 THEORETICAL_ONLY
- HTS_READY일 때만 order_blueprint_json.validation_status=PASS 를 주문표로 렌더링 - HTS_READY일 때만 order_blueprint_json.validation_status=PASS 를 주문표로 렌더링
spec/34_architecture_boundaries.yaml
+6 -2
View File
@@ -1,6 +1,10 @@
meta:
has_code_implementation: true
code_path:
- spec\34_architecture_boundaries.yaml
schema_version: 2026-06-03-architecture-boundaries-v1 schema_version: 2026-06-03-architecture-boundaries-v1
formula_id: ARCHITECTURE_BOUNDARIES_V1 formula_id: ARCHITECTURE_BOUNDARIES_V1
purpose: data -> feature -> decision -> execution -> report 단방향 원칙. purpose: data -> feature -> decision -> execution -> report 단방향 원칙.
contracts: contracts:
- renderer_calculation_count: 0 - renderer_calculation_count: 0
- reverse_dependency_count: 0 - reverse_dependency_count: 0
spec/35_rule_lifecycle_governance_v3.yaml
+8 -4
View File
@@ -1,8 +1,12 @@
meta:
has_code_implementation: true
code_path:
- spec\35_rule_lifecycle_governance_v3.yaml
schema_version: 2026-06-03-rule-lifecycle-governance-v3 schema_version: 2026-06-03-rule-lifecycle-governance-v3
formula_id: RULE_LIFECYCLE_POLICY_V3 formula_id: RULE_LIFECYCLE_POLICY_V3
purpose: rule registry governance for hardening todo completion. purpose: rule registry governance for hardening todo completion.
required_fields: required_fields:
- rule_key - rule_key
- owner - owner
- retirement_condition - retirement_condition
- expected_metric - expected_metric
spec/40_final_decision_packet_contract.yaml
+11 -7
View File
@@ -1,14 +1,18 @@
meta:
has_code_implementation: true
code_path:
- spec\40_final_decision_packet_contract.yaml
schema_version: final_decision_packet_contract.v4 schema_version: final_decision_packet_contract.v4
formula_id: FINAL_DECISION_PACKET_V4 formula_id: FINAL_DECISION_PACKET_V4
purpose: Single packet source of truth for renderer and low-capability LLM output. purpose: Single packet source of truth for renderer and low-capability LLM output.
sections: sections:
- executive - executive
- portfolio - portfolio
- ticker - ticker
- risk - risk
- execution - execution
- performance - performance
- data_quality - data_quality
provenance_requirements: provenance_requirements:
source_path: required source_path: required
json_pointer: required json_pointer: required
spec/43_quant_factor_taxonomy.yaml
+25 -21
View File
@@ -1,28 +1,32 @@
meta:
has_code_implementation: true
code_path:
- spec\43_quant_factor_taxonomy.yaml
schema_version: quant_factor_taxonomy.v1 schema_version: quant_factor_taxonomy.v1
purpose: Classify factors by horizon, decay, and conflict policy. purpose: Classify factors by horizon, decay, and conflict policy.
factor_horizons: factor_horizons:
- scalping - scalping
- short - short
- mid - mid
- long - long
required_lifecycle_fields: required_lifecycle_fields:
- factor_id - factor_id
- hypothesis - hypothesis
- market_regime_applicability - market_regime_applicability
- horizon - horizon
- decay_half_life - decay_half_life
- input_fields - input_fields
- formula_id - formula_id
- data_quality_requirements - data_quality_requirements
- expected_edge_formula - expected_edge_formula
- conflict_precedence - conflict_precedence
- position_sizing_impact - position_sizing_impact
- exit_impact - exit_impact
- golden_cases - golden_cases
- shadow_start_date - shadow_start_date
- activation_threshold - activation_threshold
- retirement_condition - retirement_condition
- owner - owner
factor_retirement_policy: factor_retirement_policy:
no_edge_improvement_lookback_days: 90 no_edge_improvement_lookback_days: 90
shadow_only_before_activation: true shadow_only_before_activation: true
spec/44_live_replay_separation.yaml
+8 -4
View File
@@ -1,8 +1,12 @@
meta:
has_code_implementation: true
code_path:
- spec\44_live_replay_separation.yaml
schema_version: live_replay_separation.v2 schema_version: live_replay_separation.v2
purpose: Separate live, paper, backtest, and replay evidence. purpose: Separate live, paper, backtest, and replay evidence.
source_types: source_types:
- live - live
- paper - paper
- backtest - backtest
- replay - replay
promotion_rule: live_t20_count_gte_30 promotion_rule: live_t20_count_gte_30
spec/45_number_provenance_contract.yaml
+9 -5
View File
@@ -1,8 +1,12 @@
meta:
has_code_implementation: true
code_path:
- spec\45_number_provenance_contract.yaml
schema_version: number_provenance_contract.v1 schema_version: number_provenance_contract.v1
purpose: Attach provenance to every report number. purpose: Attach provenance to every report number.
required_fields: required_fields:
- source_path - source_path
- json_pointer - json_pointer
- formula_id - formula_id
- input_hash - input_hash
- freshness_status - freshness_status
spec/46_low_capability_execution_pack.yaml
+10 -6
View File
@@ -1,9 +1,13 @@
meta:
has_code_implementation: true
code_path:
- spec\46_low_capability_execution_pack.yaml
schema_version: low_capability_execution_pack.v1 schema_version: low_capability_execution_pack.v1
purpose: Fixed-order packet-only context for low capability LLMs. purpose: Fixed-order packet-only context for low capability LLMs.
required_sections: required_sections:
- executive - executive
- blockers - blockers
- action_table - action_table
- shadow_ledger - shadow_ledger
- data_missing - data_missing
- education_notes - education_notes
spec/47_packaging_policy.yaml
+8 -4
View File
@@ -1,7 +1,11 @@
meta:
has_code_implementation: true
code_path:
- spec\47_packaging_policy.yaml
schema_version: packaging_policy.v1 schema_version: packaging_policy.v1
purpose: Define upload bundle inclusion and exclusion policy. purpose: Define upload bundle inclusion and exclusion policy.
required_rules: required_rules:
- source_required - source_required
- runtime_required - runtime_required
- report_required - report_required
- test_required - test_required
spec/48_module_io_contract_registry.yaml
+30 -19
View File
@@ -1,33 +1,44 @@
meta:
has_code_implementation: true
code_path:
- spec\48_module_io_contract_registry.yaml
schema_version: module_io_contract_registry.v1 schema_version: module_io_contract_registry.v1
modules: modules:
core_engine: core_engine:
id: core_engine id: core_engine
owner: architect owner: architect
inputs: ["GatherTradingData.json"] inputs:
outputs: ["Temp/final_decision_packet_active.json"] - GatherTradingData.json
schema: "schemas/final_decision_packet_v3.schema.json" outputs:
artifact_path: "src/quant_engine/compute_formula_outputs.py" - Temp/final_decision_packet_active.json
schema: schemas/final_decision_packet_v3.schema.json
artifact_path: src/quant_engine/compute_formula_outputs.py
risk_manager: risk_manager:
id: risk_manager id: risk_manager
owner: risk_officer owner: risk_officer
inputs: ["GatherTradingData.json"] inputs:
outputs: ["Temp/strategy_decision_result_v3.json"] - GatherTradingData.json
schema: "schemas/strategy_decision_result.schema.json" outputs:
artifact_path: "src/quant_engine/exit_decisions.py" - Temp/strategy_decision_result_v3.json
schema: schemas/strategy_decision_result.schema.json
artifact_path: src/quant_engine/exit_decisions.py
reporting_renderer: reporting_renderer:
id: reporting_renderer id: reporting_renderer
owner: pm owner: pm
inputs: ["Temp/final_decision_packet_active.json"] inputs:
outputs: ["Temp/operational_report.json", "Temp/operational_report.md"] - Temp/final_decision_packet_active.json
schema: "schemas/operational_report.schema.json" outputs:
artifact_path: "tools/render_operational_report.py" - Temp/operational_report.json
- Temp/operational_report.md
schema: schemas/operational_report.schema.json
artifact_path: tools/render_operational_report.py
context_builder: context_builder:
id: context_builder id: context_builder
owner: pm owner: pm
inputs: ["Temp/final_decision_packet_v4.json", "runtime/active_artifact_manifest.yaml"] inputs:
outputs: ["Temp/final_context_for_llm_v5.yaml"] - Temp/final_decision_packet_v4.json
schema: "spec/46_low_capability_execution_pack.yaml" - runtime/active_artifact_manifest.yaml
artifact_path: "tools/build_low_capability_context_pack_v5.py" outputs:
- Temp/final_context_for_llm_v5.yaml
schema: spec/46_low_capability_execution_pack.yaml
artifact_path: tools/build_low_capability_context_pack_v5.py
spec/49_refactor_methodology_contract.yaml
+25 -14
View File
@@ -1,19 +1,30 @@
meta:
has_code_implementation: true
code_path:
- spec\49_refactor_methodology_contract.yaml
schema_version: refactor_methodology_contract.v1 schema_version: refactor_methodology_contract.v1
principles: principles:
- QEDD: Quant Evidence-Driven Deterministic Development - QEDD: Quant Evidence-Driven Deterministic Development
- Contract-First: YAML/Schema before implementation - Contract-First: YAML/Schema before implementation
- Python Canonical: Logic in src/quant_engine, GAS is thin adapter - Python Canonical: Logic in src/quant_engine, GAS is thin adapter
- No LLM Math: LLM copies harness values with provenance - No LLM Math: LLM copies harness values with provenance
authority_order: authority_order:
- spec/*.yaml - spec/*.yaml
- runtime/active_artifact_manifest.yaml - runtime/active_artifact_manifest.yaml
- Temp/final_decision_packet_active.json - Temp/final_decision_packet_active.json
- governance/rules/*.yaml - governance/rules/*.yaml
file_policy: file_policy:
source: [md, yaml, py, gs] source:
runtime: [json, jsonl, schema.json] - md
- yaml
- py
- gs
runtime:
- json
- jsonl
- schema.json
do_done: do_done:
- Release DAG PASS - Release DAG PASS
- No SKIPPED in release mode - No SKIPPED in release mode
- Zero architecture boundary violations - Zero architecture boundary violations
- 100% number provenance coverage - 100% number provenance coverage
spec/51_formula_lifecycle_registry.yaml
+4
View File
@@ -1,3 +1,7 @@
meta:
has_code_implementation: true
code_path:
- spec\51_formula_lifecycle_registry.yaml
schema_version: formula_lifecycle_registry.v1 schema_version: formula_lifecycle_registry.v1
updated_at: '2026-06-13' updated_at: '2026-06-13'
purpose: '모든 ACTIVE 공식의 lifecycle 상태를 단일 레지스트리로 관리한다. spec/13_formula_registry.yaml(149개) purpose: '모든 ACTIVE 공식의 lifecycle 상태를 단일 레지스트리로 관리한다. spec/13_formula_registry.yaml(149개)
spec/anti_late_entry_contract.yaml
+23 -18
View File
@@ -1,21 +1,26 @@
meta:
has_code_implementation: true
code_path:
- spec\anti_late_entry_contract.yaml
schema_version: anti_late_entry_contract.v2 schema_version: anti_late_entry_contract.v2
goal: Define rules to block late chasing of leading stocks and prevent buying in distribution phases. goal: Define rules to block late chasing of leading stocks and prevent buying in distribution
phases.
metrics: metrics:
- id: breakout_quality - id: breakout_quality
description: "20D high status and volume acceleration at breakout" description: 20D high status and volume acceleration at breakout
- id: flow_acceleration - id: flow_acceleration
description: "Foreign/Institutional net buying acceleration indicator" description: Foreign/Institutional net buying acceleration indicator
- id: distribution_risk - id: distribution_risk
description: "Risk score reflecting high-volume churn without price progression" description: Risk score reflecting high-volume churn without price progression
- id: entry_timing_decile - id: entry_timing_decile
description: "Decile rank of entry timing relative to recent price progression" description: Decile rank of entry timing relative to recent price progression
rules: rules:
- id: RULE_OVERHEATED_BLOCK_BUY - id: RULE_OVERHEATED_BLOCK_BUY
condition: "entry_timing_decile >= 8 (Overheated zone)" condition: entry_timing_decile >= 8 (Overheated zone)
action: "BLOCK new buys until a pullback trigger is confirmed" action: BLOCK new buys until a pullback trigger is confirmed
- id: RULE_DISTRIBUTION_DOWNGRADE - id: RULE_DISTRIBUTION_DOWNGRADE
condition: "distribution_risk is HIGH" condition: distribution_risk is HIGH
action: "Downgrade BUY to HOLD/WAIT" action: Downgrade BUY to HOLD/WAIT
- id: RULE_LATE_CHASE_ATTRIBUTION - id: RULE_LATE_CHASE_ATTRIBUTION
condition: "T+5/T+20 operational outcomes are updated to attribute entry quality" condition: T+5/T+20 operational outcomes are updated to attribute entry quality
action: "Update attribution metrics" action: Update attribution metrics
spec/data_quality/expectations.yaml
+17 -7
View File
@@ -1,14 +1,24 @@
meta:
has_code_implementation: true
code_path:
- spec\data_quality\expectations.yaml
schema_version: data_quality_expectations.v1 schema_version: data_quality_expectations.v1
groups: groups:
account: account:
- not_null: [total_asset_krw, cash_krw] - not_null:
- total_asset_krw
- cash_krw
price: price:
- freshness_minutes: 390 - freshness_minutes: 390
- not_null: [current_price_krw] - not_null:
- current_price_krw
fundamentals: fundamentals:
- not_null: [roe, revenue_growth] - not_null:
- roe
- revenue_growth
external_context: external_context:
- allowed_values: [CONTEXT_ONLY, ORDER_JUDGMENT] - allowed_values:
- CONTEXT_ONLY
- ORDER_JUDGMENT
cross_field: cross_field:
- rule: current_price_krw > 0 - rule: current_price_krw > 0
spec/execution_authority_matrix_v2.yaml
+16 -12
View File
@@ -1,14 +1,18 @@
meta:
has_code_implementation: true
code_path:
- spec\execution_authority_matrix_v2.yaml
schema_version: execution_authority_matrix.v2 schema_version: execution_authority_matrix.v2
purpose: "Final execution authority map for low-capability LLM rendering." purpose: Final execution authority map for low-capability LLM rendering.
source_of_truth: source_of_truth:
- Temp/final_execution_decision_v1.json - Temp/final_execution_decision_v1.json
- Temp/operational_report.json - Temp/operational_report.json
- Temp/execution_authority_matrix_v1.json - Temp/execution_authority_matrix_v1.json
authority_precedence: authority_precedence:
- BLOCK_EXECUTION - BLOCK_EXECUTION
- AUDIT_ONLY - AUDIT_ONLY
- EXPLAIN_ONLY - EXPLAIN_ONLY
- HTS_READY - HTS_READY
field_owner_runtime: field_owner_runtime:
price: GAS_OR_PY price: GAS_OR_PY
qty: GAS_OR_PY qty: GAS_OR_PY
@@ -18,10 +22,10 @@ field_owner_runtime:
gate: GAS_OR_PY gate: GAS_OR_PY
narrative: LLM_ONLY narrative: LLM_ONLY
rules: rules:
- "If global_execution_gate != HTS_READY, suppress HTS order table." - If global_execution_gate != HTS_READY, suppress HTS order table.
- "Candidate rows with validation_status != PASS must render only as shadow ledger." - Candidate rows with validation_status != PASS must render only as shadow ledger.
- "Numeric fields must be copied from JSON; do not infer or interpolate values." - Numeric fields must be copied from JSON; do not infer or interpolate values.
- "If any authority conflict appears, emit AUDIT_ONLY and record the conflict in JSON." - If any authority conflict appears, emit AUDIT_ONLY and record the conflict in JSON.
acceptance: acceptance:
authority_conflict_count: 0 authority_conflict_count: 0
hidden_order_leak_count: 0 hidden_order_leak_count: 0
spec/exit/take_profit.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: "ko-KR" language: "ko-KR"
timezone: "Asia/Seoul" timezone: "Asia/Seoul"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["spec/strategy/action_matrix.yaml"]
migration_status: "canonical_split_active" migration_status: "canonical_split_active"
take_profit: take_profit:
spec/factor_lifecycle_registry.yaml
+4
View File
@@ -1,3 +1,7 @@
meta:
has_code_implementation: true
code_path:
- spec\factor_lifecycle_registry.yaml
schema_version: factor_lifecycle_registry.v1 schema_version: factor_lifecycle_registry.v1
description: Lifecycle states (draft, shadow, candidate, active, retired) for all description: Lifecycle states (draft, shadow, candidate, active, retired) for all
registered quant factors registered quant factors
spec/fields/field_dictionary.yaml
+4 -1
View File
@@ -1,7 +1,10 @@
meta:
has_code_implementation: true
code_path:
- spec\fields\field_dictionary.yaml
schema_version: field_dictionary.v1 schema_version: field_dictionary.v1
source_of_truth: spec/12_field_dictionary.yaml source_of_truth: spec/12_field_dictionary.yaml
policy: policy:
canonical_name_required: true canonical_name_required: true
unknown_field_action: DATA_MISSING unknown_field_action: DATA_MISSING
unit_conflict_action: DATA_CONFLICT unit_conflict_action: DATA_CONFLICT
spec/formula_golden_cases_nf.yaml
+3
View File
@@ -2,6 +2,9 @@ meta:
title: NF1~NF5 Python-harness 보조 공식 명세 golden cases title: NF1~NF5 Python-harness 보조 공식 명세 golden cases
note: 'GAS_REFERENCE_ONLY: Python 미러 없음. behavioral_coverage 체크 대상 외.' note: 'GAS_REFERENCE_ONLY: Python 미러 없음. behavioral_coverage 체크 대상 외.'
source_registry: spec/13_formula_registry.yaml:NF1~NF5 source_registry: spec/13_formula_registry.yaml:NF1~NF5
has_code_implementation: true
code_path:
- spec\formula_golden_cases_nf.yaml
golden_cases_nf: golden_cases_nf:
- formula_id: REGIME_CONDITIONAL_MACRO_FACTOR_V1 - formula_id: REGIME_CONDITIONAL_MACRO_FACTOR_V1
gas_function: GAS_REFERENCE_ONLY gas_function: GAS_REFERENCE_ONLY
spec/formula_golden_cases_v2.yaml
+4
View File
@@ -1,3 +1,7 @@
meta:
has_code_implementation: true
code_path:
- spec\formula_golden_cases_v2.yaml
golden_cases_v2: golden_cases_v2:
- formula_id: TICK_NORMALIZER_V1 - formula_id: TICK_NORMALIZER_V1
gas_function: tickNormalize_ gas_function: tickNormalize_
spec/formula_golden_cases_v3.yaml
+122 -118
View File
@@ -1,122 +1,126 @@
meta:
has_code_implementation: true
code_path:
- spec\formula_golden_cases_v3.yaml
schema_version: formula_golden_cases.v3 schema_version: formula_golden_cases.v3
source: formula_golden_cases_v2 source: formula_golden_cases_v2
note: deterministic bridge spec for coverage expansion note: deterministic bridge spec for coverage expansion
golden_cases: golden_cases:
- formula_id: POSITION_SIZE_V1 - formula_id: POSITION_SIZE_V1
bridge_only: true bridge_only: true
- formula_id: TAKE_PROFIT_LADDER_V1 - formula_id: TAKE_PROFIT_LADDER_V1
bridge_only: true bridge_only: true
- formula_id: TAKE_PROFIT_LADDER_V2 - formula_id: TAKE_PROFIT_LADDER_V2
bridge_only: true bridge_only: true
- formula_id: PEG_SCORE_V1 - formula_id: PEG_SCORE_V1
bridge_only: true bridge_only: true
- formula_id: PORTFOLIO_BAND_STATUS_V1 - formula_id: PORTFOLIO_BAND_STATUS_V1
bridge_only: true bridge_only: true
- formula_id: OVERSOLD_DELAY_V1 - formula_id: OVERSOLD_DELAY_V1
bridge_only: true bridge_only: true
- formula_id: SECTOR_ROTATION_RADAR_V1 - formula_id: SECTOR_ROTATION_RADAR_V1
bridge_only: true bridge_only: true
- formula_id: SEA_TIMING_V1 - formula_id: SEA_TIMING_V1
bridge_only: true bridge_only: true
- formula_id: ECP_RISK_SCALE_V1 - formula_id: ECP_RISK_SCALE_V1
bridge_only: true bridge_only: true
- formula_id: FOLLOW_THROUGH_DAY_CONFIRM_V1 - formula_id: FOLLOW_THROUGH_DAY_CONFIRM_V1
bridge_only: true bridge_only: true
- formula_id: EXECUTION_QUALITY_SCORE_V1 - formula_id: EXECUTION_QUALITY_SCORE_V1
bridge_only: true bridge_only: true
- formula_id: COMPOSITE_VERDICT_V1 - formula_id: COMPOSITE_VERDICT_V1
bridge_only: true bridge_only: true
- formula_id: SATELLITE_FAILURE_GATE_V1 - formula_id: SATELLITE_FAILURE_GATE_V1
bridge_only: true bridge_only: true
- formula_id: BENCHMARK_RELATIVE_TIMESERIES_V1 - formula_id: BENCHMARK_RELATIVE_TIMESERIES_V1
bridge_only: true bridge_only: true
- formula_id: RS_VERDICT_V2 - formula_id: RS_VERDICT_V2
bridge_only: true bridge_only: true
- formula_id: CASH_CREATION_PURPOSE_LOCK_V1 - formula_id: CASH_CREATION_PURPOSE_LOCK_V1
bridge_only: true bridge_only: true
- formula_id: ALPHA_EVALUATION_WINDOW_V1 - formula_id: ALPHA_EVALUATION_WINDOW_V1
bridge_only: true bridge_only: true
- formula_id: HARNESS_DATA_FRESHNESS_GATE_V1 - formula_id: HARNESS_DATA_FRESHNESS_GATE_V1
bridge_only: true bridge_only: true
- formula_id: PORTFOLIO_CORRELATION_GATE_V1 - formula_id: PORTFOLIO_CORRELATION_GATE_V1
bridge_only: true bridge_only: true
- formula_id: ALPHA_FEEDBACK_LOOP_V1 - formula_id: ALPHA_FEEDBACK_LOOP_V1
bridge_only: true bridge_only: true
- formula_id: DISTRIBUTION_SELL_DETECTOR_V1 - formula_id: DISTRIBUTION_SELL_DETECTOR_V1
bridge_only: true bridge_only: true
- formula_id: SELL_EXECUTION_TIMING_V1 - formula_id: SELL_EXECUTION_TIMING_V1
bridge_only: true bridge_only: true
- formula_id: DETERMINISTIC_ROUTING_ENGINE_V1 - formula_id: DETERMINISTIC_ROUTING_ENGINE_V1
bridge_only: true bridge_only: true
- formula_id: SELL_VALUE_PRESERVATION_TIERED_V2 - formula_id: SELL_VALUE_PRESERVATION_TIERED_V2
bridge_only: true bridge_only: true
- formula_id: TRADE_QUALITY_SCORER_V1 - formula_id: TRADE_QUALITY_SCORER_V1
bridge_only: true bridge_only: true
- formula_id: PATTERN_BLACKLIST_AUTO_V1 - formula_id: PATTERN_BLACKLIST_AUTO_V1
bridge_only: true bridge_only: true
- formula_id: HORIZON_ALLOCATION_LOCK_V1 - formula_id: HORIZON_ALLOCATION_LOCK_V1
bridge_only: true bridge_only: true
- formula_id: SMART_MONEY_LIQUIDITY_GATE_V1 - formula_id: SMART_MONEY_LIQUIDITY_GATE_V1
bridge_only: true bridge_only: true
- formula_id: FUNDAMENTAL_MULTI_FACTOR_SCORE_V2 - formula_id: FUNDAMENTAL_MULTI_FACTOR_SCORE_V2
bridge_only: true bridge_only: true
- formula_id: EARNINGS_GROWTH_QUALITY_GATE_V1 - formula_id: EARNINGS_GROWTH_QUALITY_GATE_V1
bridge_only: true bridge_only: true
- formula_id: MARKET_SHARE_MOMENTUM_PROXY_V1 - formula_id: MARKET_SHARE_MOMENTUM_PROXY_V1
bridge_only: true bridge_only: true
- formula_id: CASHFLOW_STABILITY_GATE_V1 - formula_id: CASHFLOW_STABILITY_GATE_V1
bridge_only: true bridge_only: true
- formula_id: ROUTING_DECISION_EXPLAIN_LOCK_V1 - formula_id: ROUTING_DECISION_EXPLAIN_LOCK_V1
bridge_only: true bridge_only: true
- formula_id: RATCHET_TRAILING_GENERAL_V1 - formula_id: RATCHET_TRAILING_GENERAL_V1
bridge_only: true bridge_only: true
- formula_id: EJCE_VIEW_RENDERER_V1 - formula_id: EJCE_VIEW_RENDERER_V1
bridge_only: true bridge_only: true
- formula_id: ROUTING_EXECUTION_LOG_TABLE_V1 - formula_id: ROUTING_EXECUTION_LOG_TABLE_V1
bridge_only: true bridge_only: true
- formula_id: HORIZON_CLASSIFICATION_V1 - formula_id: HORIZON_CLASSIFICATION_V1
bridge_only: true bridge_only: true
- formula_id: SMART_MONEY_FLOW_SIGNAL_V2 - formula_id: SMART_MONEY_FLOW_SIGNAL_V2
bridge_only: true bridge_only: true
- formula_id: EARNINGS_QUALITY_SIGNAL_V1 - formula_id: EARNINGS_QUALITY_SIGNAL_V1
bridge_only: true bridge_only: true
- formula_id: TRADE_QUALITY_FROM_T5_V1 - formula_id: TRADE_QUALITY_FROM_T5_V1
bridge_only: true bridge_only: true
- formula_id: PREDICTION_ACCURACY_HARNESS_V2 - formula_id: PREDICTION_ACCURACY_HARNESS_V2
bridge_only: true bridge_only: true
- formula_id: MACRO_EVENT_TICKER_IMPACT_V1 - formula_id: MACRO_EVENT_TICKER_IMPACT_V1
bridge_only: true bridge_only: true
- formula_id: EJCE_DIVERGENCE_AUDIT_V1 - formula_id: EJCE_DIVERGENCE_AUDIT_V1
bridge_only: true bridge_only: true
- formula_id: PREDICTIVE_ALPHA_REPORT_LOCK_V2 - formula_id: PREDICTIVE_ALPHA_REPORT_LOCK_V2
bridge_only: true bridge_only: true
- formula_id: VERDICT_CONSISTENCY_LOCK_V1 - formula_id: VERDICT_CONSISTENCY_LOCK_V1
bridge_only: true bridge_only: true
- formula_id: INVESTMENT_QUALITY_HEADLINE_V1 - formula_id: INVESTMENT_QUALITY_HEADLINE_V1
bridge_only: true bridge_only: true
- formula_id: SELL_PRICE_SANITY_V2 - formula_id: SELL_PRICE_SANITY_V2
bridge_only: true bridge_only: true
- formula_id: EXPORT_GATE_V2 - formula_id: EXPORT_GATE_V2
bridge_only: true bridge_only: true
- formula_id: PROACTIVE_SELL_RADAR_V2 - formula_id: PROACTIVE_SELL_RADAR_V2
bridge_only: true bridge_only: true
- formula_id: ANTI_LATE_ENTRY_GATE_V3 - formula_id: ANTI_LATE_ENTRY_GATE_V3
bridge_only: true bridge_only: true
- formula_id: PRICE_HIERARCHY_LOCK_V1 - formula_id: PRICE_HIERARCHY_LOCK_V1
bridge_only: true bridge_only: true
- formula_id: DATA_QUALITY_GATE_V2 - formula_id: DATA_QUALITY_GATE_V2
bridge_only: true bridge_only: true
- formula_id: CASH_RECOVERY_DISPLAY_LOCK_V1 - formula_id: CASH_RECOVERY_DISPLAY_LOCK_V1
bridge_only: true bridge_only: true
- formula_id: ALPHA_FEEDBACK_LOOP_V2 - formula_id: ALPHA_FEEDBACK_LOOP_V2
bridge_only: true bridge_only: true
- formula_id: ALPHA_LEAD_THRESHOLD_OPTIMIZER_V1 - formula_id: ALPHA_LEAD_THRESHOLD_OPTIMIZER_V1
bridge_only: true bridge_only: true
- formula_id: DYNAMIC_VALUE_PRESERVATION_SELL_V6 - formula_id: DYNAMIC_VALUE_PRESERVATION_SELL_V6
bridge_only: true bridge_only: true
- formula_id: PREDICTIVE_ALPHA_DIALECTIC_ENGINE_V2 - formula_id: PREDICTIVE_ALPHA_DIALECTIC_ENGINE_V2
bridge_only: true bridge_only: true
- formula_id: CAPITAL_STYLE_TIME_STOP_V1 - formula_id: CAPITAL_STYLE_TIME_STOP_V1
bridge_only: true bridge_only: true
- formula_id: EXECUTION_INTEGRITY_GATE_V1 - formula_id: EXECUTION_INTEGRITY_GATE_V1
bridge_only: true bridge_only: true
spec/formula_golden_cases_v4.yaml
+1208 -854
View File
File diff suppressed because it is too large Load Diff
spec/formula_lifecycle_index.yaml
+14 -10
View File
@@ -1,17 +1,21 @@
meta:
has_code_implementation: true
code_path:
- spec\formula_lifecycle_index.yaml
schema_version: formula_lifecycle_index.v1 schema_version: formula_lifecycle_index.v1
classifications: classifications:
active: active:
description: "Formulas currently used in the primary execution order" description: Formulas currently used in the primary execution order
match_rule: "in_primary_execution_order == true" match_rule: in_primary_execution_order == true
shadow: shadow:
description: "New formulas under evaluation" description: New formulas under evaluation
match_rule: "status == 'shadow'" match_rule: status == 'shadow'
experimental: experimental:
description: "Formulas for research only" description: Formulas for research only
match_rule: "status == 'experimental'" match_rule: status == 'experimental'
deprecated: deprecated:
description: "Formulas pending retirement" description: Formulas pending retirement
match_rule: "status == 'deprecated'" match_rule: status == 'deprecated'
runtime_supplement: runtime_supplement:
description: "Helper formulas for internal logic" description: Helper formulas for internal logic
match_rule: "role == 'helper'" match_rule: role == 'helper'
spec/formulas/manifest.yaml
+4
View File
@@ -1,3 +1,7 @@
meta:
has_code_implementation: true
code_path:
- spec\formulas\manifest.yaml
schema_version: formula_domain_manifest.v1 schema_version: formula_domain_manifest.v1
source: C:\Temp\data_feed\spec\13_formula_registry.yaml source: C:\Temp\data_feed\spec\13_formula_registry.yaml
domains: domains:
spec/gas_adapter_contract.yaml
+280 -276
View File
@@ -1,278 +1,282 @@
meta:
has_code_implementation: true
code_path:
- spec\gas_adapter_contract.yaml
schema_version: gas_adapter_contract.v1 schema_version: gas_adapter_contract.v1
exports: exports:
- function_name: "runDataFeed" - function_name: runDataFeed
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "sector_flow" sheet_key: sector_flow
status: "active" status: active
- function_name: "runDataFeed" - function_name: runDataFeed
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "macro" sheet_key: macro
status: "active" status: active
- function_name: "runDataFeed" - function_name: runDataFeed
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "core_satellite" sheet_key: core_satellite
status: "active" status: active
- function_name: "logDailyAssetHistory_" - function_name: logDailyAssetHistory_
min_arity: 2 min_arity: 2
max_arity: 2 max_arity: 2
return_shape: "void" return_shape: void
sheet_key: "daily_history" sheet_key: daily_history
status: "active" status: active
- function_name: "ensureAccountSnapshotConfirmModeSetting_" - function_name: ensureAccountSnapshotConfirmModeSetting_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "void" return_shape: void
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "upsertOperationalWarningSetting_" - function_name: upsertOperationalWarningSetting_
min_arity: 2 min_arity: 2
max_arity: 2 max_arity: 2
return_shape: "void" return_shape: void
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "getCoreSatelliteUniverse" - function_name: getCoreSatelliteUniverse
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "array" return_shape: array
sheet_key: "universe" sheet_key: universe
status: "active" status: active
- function_name: "parseAccountSnapshot_" - function_name: parseAccountSnapshot_
min_arity: 3 min_arity: 3
max_arity: 3 max_arity: 3
return_shape: "object" return_shape: object
sheet_key: "account_snapshot" sheet_key: account_snapshot
status: "active" status: active
- function_name: "parseAccountSnapshot_" - function_name: parseAccountSnapshot_
min_arity: 3 min_arity: 3
max_arity: 3 max_arity: 3
return_shape: "object" return_shape: object
sheet_key: "macro" sheet_key: macro
status: "active" status: active
- function_name: "getActiveTickers_" - function_name: getActiveTickers_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "array" return_shape: array
sheet_key: "account_snapshot" sheet_key: account_snapshot
status: "active" status: active
- function_name: "getActiveTickers_" - function_name: getActiveTickers_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "array" return_shape: array
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "checkAccountSnapshotFreshness_" - function_name: checkAccountSnapshotFreshness_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "account_snapshot" sheet_key: account_snapshot
status: "active" status: active
- function_name: "readAccountSnapshotHeat_" - function_name: readAccountSnapshotHeat_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "object" return_shape: object
sheet_key: "data_feed" sheet_key: data_feed
status: "active" status: active
- function_name: "getAccountSnapshotConfirmStats_" - function_name: getAccountSnapshotConfirmStats_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "object" return_shape: object
sheet_key: "account_snapshot" sheet_key: account_snapshot
status: "active" status: active
- function_name: "readMacroRegime_" - function_name: readMacroRegime_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "object" return_shape: object
sheet_key: "macro" sheet_key: macro
status: "active" status: active
- function_name: "parseAccuracy_" - function_name: parseAccuracy_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "object" return_shape: object
sheet_key: "monthly_history" sheet_key: monthly_history
status: "active" status: active
- function_name: "parseAccuracy_" - function_name: parseAccuracy_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "object" return_shape: object
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "getPa1WeightOverrides_" - function_name: getPa1WeightOverrides_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "object" return_shape: object
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "recordPa1FeedbackEntry_" - function_name: recordPa1FeedbackEntry_
min_arity: 2 min_arity: 2
max_arity: 2 max_arity: 2
return_shape: "void" return_shape: void
sheet_key: "pa1_feedback" sheet_key: pa1_feedback
status: "active" status: active
- function_name: "getSellPassAccuracyRate_" - function_name: getSellPassAccuracyRate_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "number" return_shape: number
sheet_key: "pa1_feedback" sheet_key: pa1_feedback
status: "active" status: active
- function_name: "evaluatePa1FeedbackBatch_" - function_name: evaluatePa1FeedbackBatch_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "pa1_feedback" sheet_key: pa1_feedback
status: "active" status: active
- function_name: "evaluatePa1FeedbackBatch_" - function_name: evaluatePa1FeedbackBatch_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "data_feed" sheet_key: data_feed
status: "active" status: active
- function_name: "evaluatePa1FeedbackBatch_" - function_name: evaluatePa1FeedbackBatch_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "adjustPaeWeights_" - function_name: adjustPaeWeights_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "updateEvaluationDashboard_" - function_name: updateEvaluationDashboard_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "evaluation_dashboard" sheet_key: evaluation_dashboard
status: "active" status: active
- function_name: "updateEvaluationDashboard_" - function_name: updateEvaluationDashboard_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "daily_history" sheet_key: daily_history
status: "active" status: active
- function_name: "updateEvaluationDashboard_" - function_name: updateEvaluationDashboard_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "macro" sheet_key: macro
status: "active" status: active
- function_name: "getAlphaHistorySummary_" - function_name: getAlphaHistorySummary_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "object" return_shape: object
sheet_key: "alpha_history" sheet_key: alpha_history
status: "active" status: active
- function_name: "auditYamlGasCoverage_" - function_name: auditYamlGasCoverage_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "object" return_shape: object
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "calcTradeQualityScorer_" - function_name: calcTradeQualityScorer_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "object" return_shape: object
sheet_key: "trade_quality_history" sheet_key: trade_quality_history
status: "active" status: active
- function_name: "calcTradeQualityScorer_" - function_name: calcTradeQualityScorer_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "object" return_shape: object
sheet_key: "data_feed" sheet_key: data_feed
status: "active" status: active
- function_name: "calcTradeQualityScorer_" - function_name: calcTradeQualityScorer_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "object" return_shape: object
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "calcPatternBlacklistAuto_" - function_name: calcPatternBlacklistAuto_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "object" return_shape: object
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "calcAlphaFeedbackLoop_" - function_name: calcAlphaFeedbackLoop_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "object" return_shape: object
sheet_key: "monthly_history" sheet_key: monthly_history
status: "active" status: active
- function_name: "calcAlphaFeedbackLoop_" - function_name: calcAlphaFeedbackLoop_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "object" return_shape: object
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "getAlphaFeedbackJson_" - function_name: getAlphaFeedbackJson_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "object" return_shape: object
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "_writeRebalanceSheet_" - function_name: _writeRebalanceSheet_
min_arity: 4 min_arity: 4
max_arity: 4 max_arity: 4
return_shape: "void" return_shape: void
sheet_key: "rebalance" sheet_key: rebalance
status: "active" status: active
- function_name: "readSettingsTab_" - function_name: readSettingsTab_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "object" return_shape: object
sheet_key: "settings" sheet_key: settings
status: "active" status: active
- function_name: "readPerformanceSheet_" - function_name: readPerformanceSheet_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "object" return_shape: object
sheet_key: "performance" sheet_key: performance
status: "active" status: active
- function_name: "readExistingEpsRevision_" - function_name: readExistingEpsRevision_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "object" return_shape: object
sheet_key: "core_satellite" sheet_key: core_satellite
status: "active" status: active
- function_name: "calcFcBudget_" - function_name: calcFcBudget_
min_arity: 2 min_arity: 2
max_arity: 2 max_arity: 2
return_shape: "number" return_shape: number
sheet_key: "performance" sheet_key: performance
status: "active" status: active
- function_name: "readAccountSnapshotMap_" - function_name: readAccountSnapshotMap_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "object" return_shape: object
sheet_key: "account_snapshot" sheet_key: account_snapshot
status: "active" status: active
- function_name: "initAccountSnapshotTemplate_" - function_name: initAccountSnapshotTemplate_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "universe" sheet_key: universe
status: "active" status: active
- function_name: "runCoreSatelliteBatch" - function_name: runCoreSatelliteBatch
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "core_satellite" sheet_key: core_satellite
status: "active" status: active
- function_name: "buildDataFeedMap_" - function_name: buildDataFeedMap_
min_arity: 1 min_arity: 1
max_arity: 1 max_arity: 1
return_shape: "object" return_shape: object
sheet_key: "data_feed" sheet_key: data_feed
status: "active" status: active
- function_name: "updatePa1WeightsManual_" - function_name: updatePa1WeightsManual_
min_arity: 0 min_arity: 0
max_arity: 0 max_arity: 0
return_shape: "void" return_shape: void
sheet_key: "settings" sheet_key: settings
status: "active" status: active
spec/operating_cadence.yaml
+52 -42
View File
@@ -1,58 +1,68 @@
meta:
has_code_implementation: true
code_path:
- spec\operating_cadence.yaml
schema_version: operating_cadence.v2 schema_version: operating_cadence.v2
updated_at: '2026-06-10T23:29:00+09:00' updated_at: '2026-06-10T23:29:00+09:00'
goal: > goal: '주말 리밸런싱과 매월 1/11/21 중간점검 cadence를 release DAG input으로 연결한다. rebalance_required,
주말 리밸런싱과 매월 1/11/21 중간점검 cadence를 release DAG input으로 연결한다. mid_check_required는 final_decision_packet에 반드시 포함된다.
rebalance_required, mid_check_required는 final_decision_packet에 반드시 포함된다.
timezone: Asia/Seoul
'
timezone: Asia/Seoul
cadence: cadence:
weekly_rebalance: weekly_rebalance:
days: [Saturday, Sunday] days:
trigger_field: rebalance_required # final packet에 삽입되는 boolean 필드 - Saturday
- Sunday
trigger_field: rebalance_required
required_sections: required_sections:
- portfolio_rebalance_playbook - portfolio_rebalance_playbook
- ticker_action_matrix - ticker_action_matrix
- sell_priority_table # 매도 후보 2개 이상이면 반드시 포함 - sell_priority_table
- cash_and_risk_budget_snapshot - cash_and_risk_budget_snapshot
interim_check: interim_check:
dates: [1, 11, 21] dates:
trigger_field: mid_check_required # final packet에 삽입되는 boolean 필드 - 1
- 11
- 21
trigger_field: mid_check_required
required_sections: required_sections:
- engine_health_card - engine_health_card
- data_missing_and_harness_update_list - data_missing_and_harness_update_list
- shadow_ledger_review - shadow_ledger_review
- calibration_drift_report - calibration_drift_report
release_dag_integration: release_dag_integration:
input_node: build_operating_cadence_signal_v1 input_node: build_operating_cadence_signal_v1
output_fields: output_fields:
- name: rebalance_required - name: rebalance_required
type: bool type: bool
description: 현재 일자가 Saturday 또는 Sunday이면 true description: 현재 일자가 Saturday 또는 Sunday이면 true
- name: mid_check_required - name: mid_check_required
type: bool type: bool
description: 현재 일자의 day가 1, 11, 21 중 하나이면 true description: 현재 일자의 day가 1, 11, 21 중 하나이면 true
- name: cadence_label - name: cadence_label
type: str type: str
enum: [WEEKEND_REBALANCE, MID_MONTH_CHECK, NORMAL] enum:
- WEEKEND_REBALANCE
- MID_MONTH_CHECK
- NORMAL
downstream_nodes: downstream_nodes:
- build_final_context # context pack에 cadence 정보 삽입 - build_final_context
- build_final_decision_packet # packet에 rebalance_required 삽입 - build_final_decision_packet
rules: rules:
- id: RULE_WEEKEND_REBALANCE_CHECK - id: RULE_WEEKEND_REBALANCE_CHECK
condition: current_day in [Saturday, Sunday] condition: current_day in [Saturday, Sunday]
action: Enforce rebalance review playbook. sell priority table must appear. action: Enforce rebalance review playbook. sell priority table must appear.
packet_field: rebalance_required packet_field: rebalance_required
- id: RULE_INTERIM_CHECK - id: RULE_INTERIM_CHECK
condition: current_date.day in [1, 11, 21] condition: current_date.day in [1, 11, 21]
action: Enforce interim quality check playbook. shadow ledger review mandatory. action: Enforce interim quality check playbook. shadow ledger review mandatory.
packet_field: mid_check_required packet_field: mid_check_required
- id: RULE_CADENCE_IN_PACKET - id: RULE_CADENCE_IN_PACKET
condition: always condition: always
action: > action: 'rebalance_required와 mid_check_required는 final_decision_packet_active에 항상
rebalance_required와 mid_check_required는 final_decision_packet_active에 포함된다. 값이 없으면 DATA_MISSING — 하네스 업데이트 필요.
항상 포함된다. 값이 없으면 DATA_MISSING — 하네스 업데이트 필요.
'
owner: quant_architect owner: quant_architect
lifecycle_state: active lifecycle_state: active
spec/ownership_map.yaml
+283 -133
View File
@@ -1,136 +1,286 @@
meta: meta:
title: "은퇴자산포트폴리오 — 문서 소유권 맵" title: 은퇴자산포트폴리오 — 문서 소유권 맵
version: "2026-05-15-F10_fragmentation_guard" version: 2026-05-15-F10_fragmentation_guard
role: "governance" role: governance
purpose: "파일별 책임 범위를 고정해 문서 파편화와 규칙 중복을 방지한다." purpose: 파일별 책임 범위를 고정해 문서 파편화와 규칙 중복을 방지한다.
has_code_implementation: true
code_path:
- spec\ownership_map.yaml
ownership_map: ownership_map:
"spec/00_execution_contract.yaml": spec/00_execution_contract.yaml:
owns: ["master_prohibitions", "hard_stops", "capture_read_ledger", "order_validation_contract"] owns:
must_not_own: ["종목 점수", "시장국면 세부 계산", "계좌별 세제 라우팅"] - master_prohibitions
"spec/risk/aggregate_risk.yaml": - hard_stops
owns: ["Total_Heat", "포트폴리오 하드스톱", "상관충격", "통합 리스크 엔진"] - capture_read_ledger
must_not_own: ["종목 진입 점수", "HTS 출력 스키마"] - order_validation_contract
"spec/risk/circuit_breakers.yaml": must_not_own:
owns: ["서킷브레이커", "거래비용 통제", "집중도 브레이크", "섹터 급락 대응"] - 종목 점수
must_not_own: ["종목 진입 점수", "HTS 출력 스키마"] - 시장국면 세부 계산
"spec/risk/market_risk_cash.yaml": - 계좌별 세제 라우팅
owns: ["MRS 현금 비중", "VIX/US10Y 위험 현금 룰"] spec/risk/aggregate_risk.yaml:
must_not_own: ["종목 진입 점수", "HTS 출력 스키마"] owns:
"spec/risk/risk_control.yaml": - Total_Heat
owns: ["risk_control compatibility index"] - 포트폴리오 하드스톱
must_not_own: ["새 임계값"] - 상관충격
"spec/risk/portfolio_exposure.yaml": - 통합 리스크 엔진
owns: ["cash_floor", "중복노출", "목표 버킷", "ETF 단계 감축", "현금 버퍼"] must_not_own:
must_not_own: ["개별 종목 점수", "JSON Schema"] - 종목 진입 점수
"spec/risk/quality_control.yaml": - HTS 출력 스키마
owns: ["리스크 품질관리 요약", "금지 규칙 참조"] spec/risk/circuit_breakers.yaml:
must_not_own: ["새 임계값"] owns:
"spec/strategy/sector_model.yaml": - 서킷브레이커
owns: ["sector_model", "CSCS", "등급 alias"] - 거래비용 통제
must_not_own: ["계좌 라우팅", "현금 하한"] - 집중도 브레이크
"spec/strategy/entry_gates.yaml": - 섹터 급락 대응
owns: ["entry_timing_guardrails", "daily_leader_scan", "anti_climax_buy_gate", "staged_entry", "pullback_reentry"] must_not_own:
must_not_own: ["Total_Heat", "capture_read_ledger"] - 종목 진입 점수
"spec/strategy/stock_model.yaml": - HTS 출력 스키마
owns: ["stock_model", "core_satellite_rule"] spec/risk/market_risk_cash.yaml:
must_not_own: ["시장 전체 현금 비중"] owns:
"spec/strategy/rebalancing_trigger.yaml": - MRS 현금 비중
owns: ["rebalancing_trigger"] - VIX/US10Y 위험 현금 룰
must_not_own: ["손절/익절 세부 실행"] must_not_own:
"spec/02_data_contract.yaml": - 종목 진입 점수
owns: ["source_priority", "data_rule", "data_completeness_gate", "xlsx_analysis_protocol"] - HTS 출력 스키마
must_not_own: ["매수/매도 결론"] spec/risk/risk_control.yaml:
"spec/12_field_dictionary.yaml": owns:
owns: ["canonical field names", "field aliases", "field units", "field missing policy"] - risk_control compatibility index
must_not_own: ["투자 판단 임계값", "매수/매도 결론"] must_not_own:
"spec/13_formula_registry.yaml": - 새 임계값
owns: ["executable formulas", "formula inputs", "formula outputs", "formula missing policy"] spec/risk/portfolio_exposure.yaml:
must_not_own: ["보고서 양식", "예시 케이스"] owns:
"spec/14_raw_workbook_mapping.yaml": - cash_floor
owns: ["raw workbook sheets", "raw workbook columns", "workbook to canonical field mapping"] - 중복노출
must_not_own: ["계좌 보유수량", "계좌 현금"] - 목표 버킷
"spec/15_account_snapshot_contract.yaml": - ETF 단계 감축
owns: ["image capture account snapshot", "holding quantity", "average cost", "cash fields", "open orders"] - 현금 버퍼
must_not_own: ["시장 가격 수집", "섹터 점수"] must_not_own:
"spec/05_position_sizing.yaml": - 개별 종목 점수
owns: ["volatility_targeting", "bayesian_confidence", "kelly brake", "integer sizing"] - JSON Schema
must_not_own: ["보유수량 판독", "보고서 양식"] spec/risk/quality_control.yaml:
"spec/06_exit_policy.yaml": owns:
owns: ["exit policy compatibility index"] - 리스크 품질관리 요약
must_not_own: ["새 임계값", "신규매수 게이트"] - 금지 규칙 참조
"spec/exit/stop_loss.yaml": must_not_own:
owns: ["stop_loss"] - 새 임계값
must_not_own: ["신규매수 게이트"] spec/strategy/sector_model.yaml:
"spec/exit/take_profit.yaml": owns:
owns: ["take_profit"] - sector_model
must_not_own: ["신규매수 게이트"] - CSCS
"spec/exit/proactive_exit_radar.yaml": - 등급 alias
owns: ["proactive_exit_radar", "divergence_alert", "overhang_warning", "rotation_radar"] must_not_own:
must_not_own: ["신규매수 게이트"] - 계좌 라우팅
"spec/exit/event_response.yaml": - 현금 하한
owns: ["event_response"] spec/strategy/entry_gates.yaml:
must_not_own: ["신규매수 게이트"] owns:
"spec/exit/position_review.yaml": - entry_timing_guardrails
owns: ["position_review_cycle"] - daily_leader_scan
must_not_own: ["신규매수 게이트"] - anti_climax_buy_gate
"spec/07_output_schema.yaml": - staged_entry
owns: ["recommendation_grade", "HTS table columns", "human output contract"] - pullback_reentry
must_not_own: ["투자 판단 임계값"] must_not_own:
"schemas/output_schema.json": - Total_Heat
owns: ["machine-readable final output schema"] - capture_read_ledger
must_not_own: ["투자 규칙 수치"] spec/strategy/stock_model.yaml:
owns:
# ── 호환 인덱스 (redirect-only, 실제 규칙은 canonical_split_files 참조) ── - stock_model
# 2026-06-22 WBS-7.11: spec/03_risk_policy.yaml, spec/04_strategy_rules.yaml은 - core_satellite_rule
# role: deprecated_redirect(영구 유지가 아닌 완전 폐기 대상)였으며 활성 참조 0건을 must_not_own:
# 확인 후 실삭제했다. 캐노니컬 split 파일들은 영향 없이 그대로 유지된다. - 시장 전체 현금 비중
"spec/06_exit_policy.yaml": spec/strategy/rebalancing_trigger.yaml:
role: "compatibility_index" owns:
owns: ["legacy path alias for spec/exit/*.yaml"] - rebalancing_trigger
must_not_own: ["새 손절/익절 임계값"] must_not_own:
canonical_files: ["spec/exit/stop_loss.yaml", "spec/exit/take_profit.yaml", "spec/exit/proactive_exit_radar.yaml", "spec/exit/event_response.yaml", "spec/exit/position_review.yaml"] - 손절/익절 세부 실행
spec/02_data_contract.yaml:
# ── 계산 스키마 ─────────────────────────────────────────────────────────── owns:
"spec/08_scoring_rules.yaml": - source_priority
owns: ["SS001_SECTOR_MODEL_SCORE 계산 규칙", "SS001 6축 점수 임계값", "KOSDAQ 정규화 공식"] - data_rule
must_not_own: ["포트폴리오 현금 비중", "exit 규칙"] - data_completeness_gate
"spec/09_decision_flow.yaml": - xlsx_analysis_protocol
owns: ["Allowed_Action 결정 흐름", "하드게이트 우선순위 순서"] must_not_own:
must_not_own: ["점수 임계값", "포지션 사이징 공식"] - 매수/매도 결론
"spec/10_portfolio_rules.yaml": spec/12_field_dictionary.yaml:
owns: ["버킷 할당 목표", "core/satellite 비율 규칙"] owns:
must_not_own: ["개별 종목 점수", "시장국면 판정"] - canonical field names
"spec/11_market_regime.yaml": - field aliases
owns: ["MARKET_REGIME_V1 판정 로직", "MRS 계산 공식", "REGIME 6개 상태 정의"] - field units
must_not_own: ["개별 종목 점수", "포지션 사이징"] - field missing policy
must_not_own:
# ── 운영 계약서 ────────────────────────────────────────────────────────── - 투자 판단 임계값
"spec/16_data_gaps_roadmap.yaml": - 매수/매도 결론
owns: ["미구현 항목 추적", "구현 우선순위 로드맵"] spec/13_formula_registry.yaml:
must_not_own: ["확정된 계산 공식", "임계값"] owns:
"spec/17_performance_contract.yaml": - executable formulas
owns: ["performance 탭 구조", "Bayesian multiplier 계산 규칙", "fc_bucket 집계"] - formula inputs
must_not_own: ["시장국면 판정", "종목 스크리닝"] - formula outputs
"spec/18_settings_contract.yaml": - formula missing policy
owns: ["settings 탭 파라미터 정의", "total_asset_krw", "risk_budget_override", "orbit_* 키"] must_not_own:
must_not_own: ["계산 공식", "임계값 수치"] - 보고서 양식
"spec/21_harness_governance_contract.yaml": - 예시 케이스
owns: ["하네스 거버넌스 3중잠금", "실행 하드락", "배포 차단 정책"] spec/14_raw_workbook_mapping.yaml:
must_not_own: ["개별 종목 점수", "주문 가격 산출식"] owns:
"spec/strategy_execution_lock_policy.yaml": - raw workbook sheets
owns: ["전략 실행락 임계값", "점수 하네스 기반 BUY/SELL 통제 임계치"] - raw workbook columns
must_not_own: ["주문 가격 산출식", "시장국면 판정"] - workbook to canonical field mapping
must_not_own:
# ── 전략 세부 파일 ────────────────────────────────────────────────────── - 계좌 보유수량
"spec/strategy/discovery.yaml": - 계좌 현금
owns: ["종목 발굴 기준", "스크리닝 필터"] spec/15_account_snapshot_contract.yaml:
must_not_own: ["손절/익절 실행", "계좌 현금"] owns:
"spec/strategy/entry_core.yaml": - image capture account snapshot
owns: ["진입 체크리스트", "core 포지션 진입 기준"] - holding quantity
must_not_own: ["Total_Heat 계산", "capture_read_ledger"] - average cost
- cash fields
- open orders
must_not_own:
- 시장 가격 수집
- 섹터 점수
spec/05_position_sizing.yaml:
owns:
- volatility_targeting
- bayesian_confidence
- kelly brake
- integer sizing
must_not_own:
- 보유수량 판독
- 보고서 양식
spec/06_exit_policy.yaml:
role: compatibility_index
owns:
- legacy path alias for spec/exit/*.yaml
must_not_own:
- 새 손절/익절 임계값
canonical_files:
- spec/exit/stop_loss.yaml
- spec/exit/take_profit.yaml
- spec/exit/proactive_exit_radar.yaml
- spec/exit/event_response.yaml
- spec/exit/position_review.yaml
spec/exit/stop_loss.yaml:
owns:
- stop_loss
must_not_own:
- 신규매수 게이트
spec/exit/take_profit.yaml:
owns:
- take_profit
must_not_own:
- 신규매수 게이트
spec/exit/proactive_exit_radar.yaml:
owns:
- proactive_exit_radar
- divergence_alert
- overhang_warning
- rotation_radar
must_not_own:
- 신규매수 게이트
spec/exit/event_response.yaml:
owns:
- event_response
must_not_own:
- 신규매수 게이트
spec/exit/position_review.yaml:
owns:
- position_review_cycle
must_not_own:
- 신규매수 게이트
spec/07_output_schema.yaml:
owns:
- recommendation_grade
- HTS table columns
- human output contract
must_not_own:
- 투자 판단 임계값
schemas/output_schema.json:
owns:
- machine-readable final output schema
must_not_own:
- 투자 규칙 수치
spec/08_scoring_rules.yaml:
owns:
- SS001_SECTOR_MODEL_SCORE 계산 규칙
- SS001 6축 점수 임계값
- KOSDAQ 정규화 공식
must_not_own:
- 포트폴리오 현금 비중
- exit 규칙
spec/09_decision_flow.yaml:
owns:
- Allowed_Action 결정 흐름
- 하드게이트 우선순위 순서
must_not_own:
- 점수 임계값
- 포지션 사이징 공식
spec/10_portfolio_rules.yaml:
owns:
- 버킷 할당 목표
- core/satellite 비율 규칙
must_not_own:
- 개별 종목 점수
- 시장국면 판정
spec/11_market_regime.yaml:
owns:
- MARKET_REGIME_V1 판정 로직
- MRS 계산 공식
- REGIME 6개 상태 정의
must_not_own:
- 개별 종목 점수
- 포지션 사이징
spec/16_data_gaps_roadmap.yaml:
owns:
- 미구현 항목 추적
- 구현 우선순위 로드맵
must_not_own:
- 확정된 계산 공식
- 임계값
spec/17_performance_contract.yaml:
owns:
- performance 탭 구조
- Bayesian multiplier 계산 규칙
- fc_bucket 집계
must_not_own:
- 시장국면 판정
- 종목 스크리닝
spec/18_settings_contract.yaml:
owns:
- settings 탭 파라미터 정의
- total_asset_krw
- risk_budget_override
- orbit_* 키
must_not_own:
- 계산 공식
- 임계값 수치
spec/21_harness_governance_contract.yaml:
owns:
- 하네스 거버넌스 3중잠금
- 실행 하드락
- 배포 차단 정책
must_not_own:
- 개별 종목 점수
- 주문 가격 산출식
spec/strategy_execution_lock_policy.yaml:
owns:
- 전략 실행락 임계값
- 점수 하네스 기반 BUY/SELL 통제 임계치
must_not_own:
- 주문 가격 산출식
- 시장국면 판정
spec/strategy/discovery.yaml:
owns:
- 종목 발굴 기준
- 스크리닝 필터
must_not_own:
- 손절/익절 실행
- 계좌 현금
spec/strategy/entry_core.yaml:
owns:
- 진입 체크리스트
- core 포지션 진입 기준
must_not_own:
- Total_Heat 계산
- capture_read_ledger
policy: policy:
- "새 규칙 추가 전 ownership_map에서 소유 파일을 먼저 확인한다." - 새 규칙 추가 전 ownership_map에서 소유 파일을 먼저 확인한다.
- "must_not_own 영역에 규칙을 추가하면 validate_specs.py에서 실패 처리한다." - must_not_own 영역에 규칙을 추가하면 validate_specs.py에서 실패 처리한다.
- "임계값은 canonical 소유 파일에만 추가한다." - 임계값은 canonical 소유 파일에만 추가한다.
spec/profit_preservation_contract.yaml
+24 -18
View File
@@ -1,21 +1,27 @@
meta:
has_code_implementation: true
code_path:
- spec\profit_preservation_contract.yaml
schema_version: profit_preservation_contract.v2 schema_version: profit_preservation_contract.v2
goal: Define rules and metrics to preserve unrealized profits and enforce trailing stop policies. goal: Define rules and metrics to preserve unrealized profits and enforce trailing
stop policies.
metrics: metrics:
- id: profit_giveback - id: profit_giveback
description: "Ratio of profit returned from peak price since entry" description: Ratio of profit returned from peak price since entry
- id: trailing_stop - id: trailing_stop
description: "Trailing stop price level determined by ratchet stage" description: Trailing stop price level determined by ratchet stage
- id: ratchet_stage - id: ratchet_stage
description: "Lock tier based on maximum achieved profit percentage (e.g. PROFIT_LOCK_10, APEX_SUPER)" description: Lock tier based on maximum achieved profit percentage (e.g. PROFIT_LOCK_10,
- id: sell_slippage_budget APEX_SUPER)
description: "Allowed slippage threshold during execution" - id: sell_slippage_budget
description: Allowed slippage threshold during execution
rules: rules:
- id: RULE_PROFIT_GIVEBACK_GUARD - id: RULE_PROFIT_GIVEBACK_GUARD
condition: "profit_pct >= 20% and profit_giveback > 30% of peak gain" condition: profit_pct >= 20% and profit_giveback > 30% of peak gain
action: "Enforce trailing stop / lock remaining profit" action: Enforce trailing stop / lock remaining profit
- id: RULE_DRAWDOWN_GUARD_PRIORITY - id: RULE_DRAWDOWN_GUARD_PRIORITY
condition: "portfolio in value_preservation_stage" condition: portfolio in value_preservation_stage
action: "Prioritize drawdown guard over new alpha signals" action: Prioritize drawdown guard over new alpha signals
- id: RULE_VALUE_DAMAGE_LIMIT - id: RULE_VALUE_DAMAGE_LIMIT
condition: "unrealized maximum drawdown per ticker" condition: unrealized maximum drawdown per ticker
limit: "value_damage_pct_avg <= 10%" limit: value_damage_pct_avg <= 10%
spec/property_invariants.yaml
+16 -12
View File
@@ -1,15 +1,19 @@
meta:
has_code_implementation: true
code_path:
- spec\property_invariants.yaml
schema_version: property_invariants.v1 schema_version: property_invariants.v1
goal: Define investment engine metamorphic and property invariants. goal: Define investment engine metamorphic and property invariants.
invariants: invariants:
- id: INV_CASH_SHORTFALL_MONOTONICITY - id: INV_CASH_SHORTFALL_MONOTONICITY
description: "현금 부족액 증가 시 신규 매수 권한/수량은 증가할 수 없음" description: 현금 부족액 증가 시 신규 매수 권한/수량은 증가할 수 없음
formula_ref: "spec/risk/portfolio_exposure.yaml" formula_ref: spec/risk/portfolio_exposure.yaml
- id: INV_MARKET_RISK_MONOTONICITY - id: INV_MARKET_RISK_MONOTONICITY
description: "시장 위험 증가 시 position scale은 증가할 수 없음" description: 시장 위험 증가 시 position scale은 증가할 수 없음
formula_ref: "spec/risk/market_risk_cash.yaml" formula_ref: spec/risk/market_risk_cash.yaml
- id: INV_MISSING_DATA_CONFIDENCE - id: INV_MISSING_DATA_CONFIDENCE
description: "데이터 결측 추가 시 confidence는 상승할 수 없음" description: 데이터 결측 추가 시 confidence는 상승할 수 없음
formula_ref: "spec/02_data_contract.yaml" formula_ref: spec/02_data_contract.yaml
- id: INV_STALE_PRICE_ZERO_QUANTITY - id: INV_STALE_PRICE_ZERO_QUANTITY
description: "stale 가격이면 실행 주문 수량은 0" description: stale 가격이면 실행 주문 수량은 0
formula_ref: "spec/00_execution_contract.yaml" formula_ref: spec/00_execution_contract.yaml
spec/risk/circuit_breakers.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: "ko-KR" language: "ko-KR"
timezone: "Asia/Seoul" timezone: "Asia/Seoul"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["spec/strategy/action_matrix.yaml"]
migration_status: "canonical_split_active" migration_status: "canonical_split_active"
risk_control: risk_control:
spec/risk/portfolio_exposure.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: "ko-KR" language: "ko-KR"
timezone: "Asia/Seoul" timezone: "Asia/Seoul"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["spec/strategy/action_matrix.yaml"]
migration_status: "canonical_split_active" migration_status: "canonical_split_active"
authority_rule: "이 split 파일이 해당 섹션의 canonical source다. parent_file은 legacy compatibility index다." authority_rule: "이 split 파일이 해당 섹션의 canonical source다. parent_file은 legacy compatibility index다."
spec/routing_trace_v2.yaml
+30 -27
View File
@@ -1,41 +1,44 @@
meta:
has_code_implementation: true
code_path:
- spec\routing_trace_v2.yaml
formula_id: ROUTING_SERVING_DECISION_TRACE_V2 formula_id: ROUTING_SERVING_DECISION_TRACE_V2
name: 라우팅 및 서빙 결정 추적 (Routing & Serving Decision Trace) name: 라우팅 및 서빙 결정 추적 (Routing & Serving Decision Trace)
description: 라우팅부터 서빙, 결정, 수량/가격 산출까지 전체 파이프라인의 이력을 단일 Trace ID로 기록하여 LLM의 자유도를 0%로 통제합니다. description: 라우팅부터 서빙, 결정, 수량/가격 산출까지 전체 파이프라인의 이력을 단일 Trace ID로 기록하여 LLM의 자유도를 0%로
통제합니다.
rules: rules:
- id: TR001 - id: TR001
condition: "trace_id IS NULL OR step_completed == FALSE" condition: trace_id IS NULL OR step_completed == FALSE
action: "ABORT_PIPELINE" action: ABORT_PIPELINE
reason: "라우팅 파이프라인 추적 누락 또는 비정상 종료" reason: 라우팅 파이프라인 추적 누락 또는 비정상 종료
- id: TR002 - id: TR002
condition: "llm_override_detected == TRUE" condition: llm_override_detected == TRUE
action: "INVALID_LLM_OVERRIDE" action: INVALID_LLM_OVERRIDE
reason: "LLM이 하네스 결정값을 임의로 번복 또는 재생성 시도" reason: LLM이 하네스 결정값을 임의로 번복 또는 재생성 시도
output: output:
schema: routing_serving_trace_v3_json schema: routing_serving_trace_v3_json
fields: fields:
- trace_id: STRING - trace_id: STRING
- route: STRING - route: STRING
- prompt_entry: STRING - prompt_entry: STRING
- gate_path: ARRAY of STRING - gate_path: ARRAY of STRING
- final_block_reason: STRING OR NULL - final_block_reason: STRING OR NULL
- llm_serving_budget: NUMBER (Must be 0) - llm_serving_budget: NUMBER (Must be 0)
# ── 단계12: RELEASE_GATE_TRUTH (TASK-001 연동) ──────────────────────────────
routing_steps: routing_steps:
step_12: step_12:
id: 12 id: 12
name: RELEASE_GATE_TRUTH name: RELEASE_GATE_TRUTH
formula_id: RELEASE_GATE_TRUTH_V1 formula_id: RELEASE_GATE_TRUTH_V1
description: "honest_proof_score >= 70 이어야만 HTS 주문 생성 허용" description: honest_proof_score >= 70 이어야만 HTS 주문 생성 허용
inputs: inputs:
- algorithm_guidance_proof_v1.json.honest_proof_score - algorithm_guidance_proof_v1.json.honest_proof_score
- algorithm_guidance_proof_v1.json.honest_gate - algorithm_guidance_proof_v1.json.honest_gate
- pass_100_criteria_v3.json.effective_release_gate - pass_100_criteria_v3.json.effective_release_gate
output_key: effective_release_gate output_key: effective_release_gate
on_blocked: "이후 HTS 주문 생성 0건 — THEORETICAL_ONLY 렌더" on_blocked: 이후 HTS 주문 생성 0건 — THEORETICAL_ONLY 렌더
acceptance: acceptance:
- "routing_execution_log 행수 == 12" - routing_execution_log 행수 == 12
- "단계12 status가 effective_release_gate와 일치" - 단계12 status가 effective_release_gate와 일치
gs_coverage: "gas_apex_runtime_core.gs:buildRoutingExecutionLogV2_()" gs_coverage: gas_apex_runtime_core.gs:buildRoutingExecutionLogV2_()
python_tool: "tools/build_routing_execution_log_v1.py" python_tool: tools/build_routing_execution_log_v1.py
validator: "tools/validate_routing_trace_replay_v1.py" validator: tools/validate_routing_trace_replay_v1.py
spec/strategy/action_matrix.yaml
+2
View File
@@ -3,6 +3,8 @@ meta:
version: "2026-05-17-action_matrix_v1" version: "2026-05-17-action_matrix_v1"
language: "ko-KR" language: "ko-KR"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["spec/strategy/action_matrix.yaml"]
purpose: > purpose: >
gas_data_feed.gs의 Allowed_Action(매수 게이트) + calcSellDecision_(매도) + gas_data_feed.gs의 Allowed_Action(매수 게이트) + calcSellDecision_(매도) +
calcFinalDecision_(통합)이 어떤 조건에서 어떤 Final_Action을 출력하는지 calcFinalDecision_(통합)이 어떤 조건에서 어떤 Final_Action을 출력하는지
spec/strategy/discovery.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: "ko-KR" language: "ko-KR"
timezone: "Asia/Seoul" timezone: "Asia/Seoul"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["spec/strategy/action_matrix.yaml"]
migration_status: "canonical_split_active" migration_status: "canonical_split_active"
anti_late_trade_rule: anti_late_trade_rule:
spec/strategy/entry_core.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: "ko-KR" language: "ko-KR"
timezone: "Asia/Seoul" timezone: "Asia/Seoul"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["spec/strategy/action_matrix.yaml"]
migration_status: "canonical_split_active" migration_status: "canonical_split_active"
entry_timing_guardrails: entry_timing_guardrails:
spec/strategy/leader_scan.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: "ko-KR" language: "ko-KR"
timezone: "Asia/Seoul" timezone: "Asia/Seoul"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["spec/strategy/action_matrix.yaml"]
migration_status: "canonical_split_active" migration_status: "canonical_split_active"
entry_timing_guardrails: entry_timing_guardrails:
spec/strategy/rebalancing_trigger.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: "ko-KR" language: "ko-KR"
timezone: "Asia/Seoul" timezone: "Asia/Seoul"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["spec/strategy/action_matrix.yaml"]
migration_status: "canonical_split_active" migration_status: "canonical_split_active"
authority_rule: "이 split 파일이 해당 섹션의 canonical source다. parent_file은 legacy compatibility index다." authority_rule: "이 split 파일이 해당 섹션의 canonical source다. parent_file은 legacy compatibility index다."
spec/strategy/sector_model.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: "ko-KR" language: "ko-KR"
timezone: "Asia/Seoul" timezone: "Asia/Seoul"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["spec/strategy/action_matrix.yaml"]
migration_status: "canonical_split_active" migration_status: "canonical_split_active"
authority_rule: "이 split 파일이 해당 섹션의 canonical source다. parent_file은 legacy compatibility index다." authority_rule: "이 split 파일이 해당 섹션의 canonical source다. parent_file은 legacy compatibility index다."
spec/strategy/staged_entry.yaml
+2
View File
@@ -5,6 +5,8 @@ meta:
language: "ko-KR" language: "ko-KR"
timezone: "Asia/Seoul" timezone: "Asia/Seoul"
role: "canonical" role: "canonical"
has_code_implementation: true
code_path: ["spec/strategy/action_matrix.yaml"]
migration_status: "canonical_split_active" migration_status: "canonical_split_active"
entry_timing_guardrails: entry_timing_guardrails:
spec/xref_matrix.yaml
+103 -109
View File
@@ -1,145 +1,139 @@
meta: meta:
title: "은퇴자산포트폴리오 — 참조 허용 매트릭스" title: 은퇴자산포트폴리오 — 참조 허용 매트릭스
version: "2026-05-15-F10_fragmentation_guard" version: 2026-05-15-F10_fragmentation_guard
role: "governance" role: governance
purpose: "파일 간 순환 참조와 책임 침범을 방지한다." purpose: 파일 간 순환 참조와 책임 침범을 방지한다.
has_code_implementation: true
code_path:
- spec\xref_matrix.yaml
xref_matrix: xref_matrix:
"spec/08_scoring_rules.yaml": spec/08_scoring_rules.yaml:
may_reference: may_reference:
- "spec/00_execution_contract.yaml" - spec/00_execution_contract.yaml
- "spec/12_field_dictionary.yaml" - spec/12_field_dictionary.yaml
- "spec/13_formula_registry.yaml" - spec/13_formula_registry.yaml
- "spec/02_data_contract.yaml" - spec/02_data_contract.yaml
- "spec/risk/aggregate_risk.yaml" - spec/risk/aggregate_risk.yaml
- "spec/risk/market_risk_cash.yaml" - spec/risk/market_risk_cash.yaml
- "spec/risk/portfolio_exposure.yaml" - spec/risk/portfolio_exposure.yaml
- "spec/strategy/sector_model.yaml" - spec/strategy/sector_model.yaml
- "spec/07_output_schema.yaml" - spec/07_output_schema.yaml
must_not_reference: must_not_reference:
- "examples/" - examples/
- "prompts/" - prompts/
"spec/10_portfolio_rules.yaml": spec/10_portfolio_rules.yaml:
may_reference: may_reference:
- "spec/01_objective_profile.yaml" - spec/01_objective_profile.yaml
- "spec/12_field_dictionary.yaml" - spec/12_field_dictionary.yaml
- "spec/13_formula_registry.yaml" - spec/13_formula_registry.yaml
- "spec/risk/portfolio_exposure.yaml" - spec/risk/portfolio_exposure.yaml
- "spec/risk/aggregate_risk.yaml" - spec/risk/aggregate_risk.yaml
- "spec/risk/market_risk_cash.yaml" - spec/risk/market_risk_cash.yaml
must_not_reference: must_not_reference:
- "schemas/output_schema.json" - schemas/output_schema.json
"spec/07_output_schema.yaml": spec/07_output_schema.yaml:
may_reference: may_reference:
- "schemas/output_schema.json" - schemas/output_schema.json
- "RetirementAssetPortfolioReportTemplate.yaml" - RetirementAssetPortfolioReportTemplate.yaml
- "spec/00_execution_contract.yaml" - spec/00_execution_contract.yaml
- "spec/12_field_dictionary.yaml" - spec/12_field_dictionary.yaml
- "spec/13_formula_registry.yaml" - spec/13_formula_registry.yaml
must_not_reference: must_not_reference:
- "examples/" - examples/
- "prompts/" - prompts/
"schemas/output_schema.json": schemas/output_schema.json:
may_reference: [] may_reference: []
must_not_reference: must_not_reference:
- "spec/" - spec/
- "examples/" - examples/
- "prompts/" - prompts/
"examples/": examples/:
may_reference: may_reference:
- "spec/" - spec/
- "schemas/output_schema.json" - schemas/output_schema.json
must_not_reference: [] must_not_reference: []
"prompts/": prompts/:
may_reference: may_reference:
- "RetirementAssetPortfolio.yaml" - RetirementAssetPortfolio.yaml
- "spec/" - spec/
- "schemas/output_schema.json" - schemas/output_schema.json
must_not_reference: [] must_not_reference: []
spec/53_factor_conflict_matrix.yaml:
# ── 신규 계약 파일 xref ────────────────────────────────────────────────────
"spec/53_factor_conflict_matrix.yaml":
may_reference: may_reference:
- "spec/43_quant_factor_taxonomy.yaml" - spec/43_quant_factor_taxonomy.yaml
- "spec/00_execution_contract.yaml" - spec/00_execution_contract.yaml
- "spec/factor_lifecycle_registry.yaml" - spec/factor_lifecycle_registry.yaml
must_not_reference: must_not_reference:
- "examples/" - examples/
conflict_resolution_note: > conflict_resolution_note: 'spec/53 conflict_precedence_rules rank 6 = market_regime
spec/53 conflict_precedence_rules rank 6 = market_regime (포지션 스케일 조정). (포지션 스케일 조정). spec/00 P3 = cash_floor·Total_Heat·hard_stop은 어떤 alpha/strategy
spec/00 P3 = cash_floor·Total_Heat·hard_stop은 어떤 alpha/strategy 논리보다 우선. 논리보다 우선. 두 계약이 충돌하면 spec/00 P3가 우선한다 (spec/00 source_of_truth.priority=highest).
두 계약이 충돌하면 spec/00 P3가 우선한다 (spec/00 source_of_truth.priority=highest). spec/53의 rank 1(portfolio_risk_budget)이 사실상 spec/00 P3의 집행자이며, market_regime(rank
spec/53의 rank 1(portfolio_risk_budget)이 사실상 spec/00 P3의 집행자이며, 6)은 cash_floor를 완화하는 목적으로 사용될 수 없다.
market_regime(rank 6)은 cash_floor를 완화하는 목적으로 사용될 수 없다.
"spec/52_decision_trace_replay_contract.yaml": '
spec/52_decision_trace_replay_contract.yaml:
may_reference: may_reference:
- "spec/41_release_dag.yaml" - spec/41_release_dag.yaml
- "spec/40_final_decision_packet_contract.yaml" - spec/40_final_decision_packet_contract.yaml
must_not_reference: must_not_reference:
- "examples/" - examples/
spec/54_temporal_data_integrity.yaml:
"spec/54_temporal_data_integrity.yaml":
may_reference: may_reference:
- "spec/00_execution_contract.yaml" - spec/00_execution_contract.yaml
- "spec/exit/stop_loss.yaml" - spec/exit/stop_loss.yaml
must_not_reference: must_not_reference:
- "examples/" - examples/
temporal_note: > temporal_note: 'TIME_STOP 판단에 사용되는 hold_days는 반드시 entry_date 기준 실제 경과일 기반이어야 한다.
TIME_STOP 판단에 사용되는 hold_days는 반드시 entry_date 기준 실제 경과일 기반이어야 한다.
미래 날짜 hold_days 사용은 LOOKAHEAD_BIAS로 분류된다. 미래 날짜 hold_days 사용은 LOOKAHEAD_BIAS로 분류된다.
"spec/55_execution_simulator_contract.yaml": '
spec/55_execution_simulator_contract.yaml:
may_reference: may_reference:
- "spec/00_execution_contract.yaml" - spec/00_execution_contract.yaml
- "spec/13_formula_registry.yaml" - spec/13_formula_registry.yaml
- "spec/15_account_snapshot_contract.yaml" - spec/15_account_snapshot_contract.yaml
must_not_reference: must_not_reference:
- "examples/" - examples/
spec/56_renderer_copy_only_contract.yaml:
"spec/56_renderer_copy_only_contract.yaml":
may_reference: may_reference:
- "spec/00_execution_contract.yaml" - spec/00_execution_contract.yaml
- "spec/40_final_decision_packet_contract.yaml" - spec/40_final_decision_packet_contract.yaml
must_not_reference: must_not_reference:
- "examples/" - examples/
spec/57_shadow_promotion_scorecard.yaml:
"spec/57_shadow_promotion_scorecard.yaml":
may_reference: may_reference:
- "spec/44_live_replay_separation.yaml" - spec/44_live_replay_separation.yaml
- "spec/43_quant_factor_taxonomy.yaml" - spec/43_quant_factor_taxonomy.yaml
must_not_reference: must_not_reference:
- "examples/" - examples/
spec/58_llm_determinism_contract.yaml:
"spec/58_llm_determinism_contract.yaml":
may_reference: may_reference:
- "spec/46_low_capability_execution_pack.yaml" - spec/46_low_capability_execution_pack.yaml
- "spec/00_execution_contract.yaml" - spec/00_execution_contract.yaml
must_not_reference: must_not_reference:
- "examples/" - examples/
# ── 충돌 해소 우선순위 선언 ────────────────────────────────────────────────
conflict_resolution_precedence: conflict_resolution_precedence:
canonical_authority: "spec/00_execution_contract.yaml (source_of_truth.priority=highest)" canonical_authority: spec/00_execution_contract.yaml (source_of_truth.priority=highest)
rules: rules:
- id: "XREF_CR001" - id: XREF_CR001
description: > description: 'spec/53 rank 6 market_regime vs spec/00 P3 cash_floor: spec/00 P3가
spec/53 rank 6 market_regime vs spec/00 P3 cash_floor: 절대 우선한다. market_regime은 position_scale만 조정하며 cash_floor·Total_Heat·hard_stop을
spec/00 P3가 절대 우선한다. market_regime은 position_scale만 조정하며 완화하는 목적으로 사용 금지.
cash_floor·Total_Heat·hard_stop을 완화하는 목적으로 사용 금지.
winner: "spec/00_execution_contract.yaml:P3_no_risk_block_override"
loser: "spec/53_factor_conflict_matrix.yaml:rank=6_market_regime"
- id: "XREF_CR002" '
description: > winner: spec/00_execution_contract.yaml:P3_no_risk_block_override
spec/54 temporal integrity vs spec/00 P4 intraday_lock: loser: spec/53_factor_conflict_matrix.yaml:rank=6_market_regime
spec/00 P4가 상위 계약. spec/54는 backfill/lookahead를 차단하며, - id: XREF_CR002
spec/00 P4는 장중 신규 매수·전량 매도를 차단한다. 두 계약은 서로 보완적. description: 'spec/54 temporal integrity vs spec/00 P4 intraday_lock: spec/00
winner: "spec/00_execution_contract.yaml:P4_no_intraday_speculation" P4가 상위 계약. spec/54는 backfill/lookahead를 차단하며, spec/00 P4는 장중 신규 매수·전량 매도를 차단한다.
supplementary: "spec/54_temporal_data_integrity.yaml:NO_LOOKAHEAD" 두 계약은 서로 보완적.
'
winner: spec/00_execution_contract.yaml:P4_no_intraday_speculation
supplementary: spec/54_temporal_data_integrity.yaml:NO_LOOKAHEAD
policy: policy:
- "must_not_reference 위반은 검증 실패." - must_not_reference 위반은 검증 실패.
- "may_reference 외 참조는 경고 대상. critical 파일은 실패 대상." - may_reference 외 참조는 경고 대상. critical 파일은 실패 대상.
- "schemas/output_schema.json은 어떤 spec도 참조하지 않는다." - schemas/output_schema.json은 어떤 spec도 참조하지 않는다.
- "conflict_resolution_precedence는 spec 충돌 시 판단 순서를 명시한다. spec/00이 항상 최상위." - conflict_resolution_precedence는 spec 충돌 시 판단 순서를 명시한다. spec/00이 항상 최상위.
tests/parity/test_execution_decision_parity_v1.py
+191
View File
@@ -0,0 +1,191 @@
"""
Parity test for execution_decision_v1.py against GAS source.
F05: Exit/sell action decision logic.
Tests calc_exit_sell_action() with core priorities and edge cases.
Source: src/gas_adapter_parts/gdf_01_price_metrics.gs:calcExitSellAction_
"""
import pytest
from formulas.execution_decision_v1 import calc_exit_sell_action
class TestExitSellActionPriorities:
"""Test exit/sell action priority hierarchy."""
def test_hold_default(self):
"""Default HOLD when no signals trigger."""
result = calc_exit_sell_action({
"close": 100,
"profitPct": 5,
"rwPartial": 0,
})
assert result["action"] == "HOLD"
def test_priority_1_stop_action(self):
"""Priority 1: STOP_OR_TIME_EXIT_READY → EXIT_100."""
result = calc_exit_sell_action({
"close": 100,
"stopPrice": 95,
"timingAction": "STOP_OR_TIME_EXIT_READY",
})
assert result["action"] == "EXIT_100"
assert result["ratio_pct"] == 100
def test_priority_1_strong_rw(self):
"""Priority 1: rwPartial >= 4 → EXIT_100."""
result = calc_exit_sell_action({
"close": 100,
"stopPrice": 95,
"rwPartial": 4,
})
assert result["action"] == "EXIT_100"
def test_priority_5_profit_50(self):
"""Priority 5: profitPct >= 50 → PROFIT_TRIM_50."""
result = calc_exit_sell_action({
"close": 100,
"profitPct": 55,
"tp2Price": 155,
"atr20": 2,
})
assert result["action"] == "PROFIT_TRIM_50"
assert result["ratio_pct"] == 50
def test_priority_5_take_profit_tier1(self):
"""Priority 5: profitPct >= 10 → TAKE_PROFIT_TIER1."""
result = calc_exit_sell_action({
"close": 100,
"profitPct": 15,
"tp1Price": 115,
"atr20": 2,
})
assert result["action"] == "TAKE_PROFIT_TIER1"
assert result["ratio_pct"] == 25
def test_priority_4_trailing_stop_breach(self):
"""Priority 4: close <= trailingStop → TRAILING_STOP_BREACH."""
result = calc_exit_sell_action({
"close": 95,
"trailingStop": 98,
"atr20": 2,
})
assert result["action"] == "TRAILING_STOP_BREACH"
assert result["ratio_pct"] == 70
def test_priority_4_rw_medium(self):
"""Priority 4: rwPartial >= 2 → TRIM_50."""
result = calc_exit_sell_action({
"close": 100,
"atr20": 2,
"rwPartial": 2,
})
assert result["action"] == "TRIM_50"
assert result["ratio_pct"] == 50
def test_priority_6_time_stop_near(self):
"""Priority 6: daysToTimeStop <= 7 → TIME_TRIM_50."""
result = calc_exit_sell_action({
"close": 100,
"atr20": 2,
"daysToTimeStop": 5,
"profitPct": 0,
"rwPartial": 0,
})
assert result["action"] == "TIME_TRIM_50"
assert result["ratio_pct"] == 50
def test_price_source_tier2(self):
"""When tp2Price available, use it for PROFIT_TRIM_50."""
result = calc_exit_sell_action({
"close": 100,
"profitPct": 55,
"tp2Price": 155,
"atr20": 2,
})
assert result["price_source"] == "TP2_PRICE"
assert result["price_basis"] == "TAKE_PROFIT_TIER2_PRICE"
def test_price_fallback_to_close_protect(self):
"""When tp2Price absent, use closeProtectLimit."""
result = calc_exit_sell_action({
"close": 100,
"profitPct": 55,
"atr20": 2,
})
assert result["price_source"] == "CLOSE_PROFIT_PROTECT"
assert result["price_basis"] == "PRIOR_CLOSE_X_0.998"
def test_validation_confirmed(self):
"""Validation = SIGNAL_CONFIRMED when price valid."""
result = calc_exit_sell_action({
"close": 100,
"stopPrice": 95,
"timingAction": "STOP_OR_TIME_EXIT_READY",
})
assert result["validation"] == "SIGNAL_CONFIRMED"
def test_validation_hold_no_action(self):
"""Validation = NO_SELL_ACTION when HOLD."""
result = calc_exit_sell_action({
"close": 100,
"profitPct": 5,
})
assert result["validation"] == "NO_SELL_ACTION"
def test_rw_early_warning_trim_33(self):
"""Priority 4b: rwPartial >= 1 + timingExitScore >= 30 → TRIM_33."""
result = calc_exit_sell_action({
"close": 100,
"atr20": 2,
"rwPartial": 1,
"timingExitScore": 35,
})
assert result["action"] == "TRIM_33"
assert result["ratio_pct"] == 33
def test_rw_signal_only_trim_25(self):
"""Priority 4c: rwPartial >= 1 only → TRIM_25."""
result = calc_exit_sell_action({
"close": 100,
"atr20": 2,
"rwPartial": 1,
"timingExitScore": 0,
})
assert result["action"] == "TRIM_25"
assert result["ratio_pct"] == 25
def test_profit_trim_35(self):
"""Priority 5: profitPct >= 30 → PROFIT_TRIM_35."""
result = calc_exit_sell_action({
"close": 100,
"profitPct": 35,
"tp2Price": 135,
"atr20": 2,
})
assert result["action"] == "PROFIT_TRIM_35"
assert result["ratio_pct"] == 35
def test_profit_trim_25(self):
"""Priority 5: profitPct >= 20 → PROFIT_TRIM_25."""
result = calc_exit_sell_action({
"close": 100,
"profitPct": 25,
"tp1Price": 125,
"atr20": 2,
})
assert result["action"] == "PROFIT_TRIM_25"
assert result["ratio_pct"] == 25
def test_time_stop_approaching(self):
"""Priority 6b: daysToTimeStop <= 14 → TIME_TRIM_25."""
result = calc_exit_sell_action({
"close": 100,
"atr20": 2,
"daysToTimeStop": 10,
"profitPct": 0,
"rwPartial": 0,
})
assert result["action"] == "TIME_TRIM_25"
assert result["ratio_pct"] == 25