cnhkmcp 2.3.0__py3-none-any.whl → 2.3.2__py3-none-any.whl

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852471944247400.json

@@ -0,0 +1,22 @@
+{
+    "template": "divide(vec_avg({cash_from_oper}), add(vec_avg({st_debt}), vec_avg({other_st_liab}), vec_avg({cash_paid_for_int_debt})))",
+    "idea": "**Concept**: Fundamental Solvency Essence\n- **Sample Fields Used**: cash_from_oper, st_debt, other_st_liab, cash_paid_for_int_debt\n- **Definition**: Operating cash flow coverage of all short-term obligations including interest, measuring pure liquidity adequacy without refinancing dependency\n- **Why This Feature**: Essential liquidity metric focusing on operational self-sufficiency; removes equity market and long-term financing noise\n- **Logical Meaning**: Can the business fund its immediate obligations from operations alone?\n- **is filling nan necessary**: Ensure all liability components are captured; missing values may understate obligations\n- **Directionality**: Values >2.0 indicate operational self-sufficiency; <1.0 indicates dependency on external financing\n- **Boundary Conditions**: Values approaching zero indicate immediate liquidity crisis; extremely high values indicate inefficient capital structure",
+    "expression_list": [
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_a_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_a_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_a_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_a_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_a_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_a_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_a_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_a_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_q_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_q_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_q_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_q_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_q_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_q_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_q_cf_act_cash_paid_for_int_debt)))",
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), add(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_cf_q_cf_act_cash_paid_for_int_debt)))"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852472483548800.json

@@ -0,0 +1,14 @@
+{
+    "template": "divide(vec_avg({cash_from_oper}), subtract(vec_avg({net_inc_avail_com_shrhldrs}), vec_avg({unrealized_gain_loss_comp_inc})))",
+    "idea": "**Concept**: Core Operating Cash Persistence\n- **Sample Fields Used**: cash_from_oper, net_inc_avail_com_shrhldrs, unrealized_gain_loss_comp_inc\n- **Definition**: Operating cash flow divided by net income adjusted for comprehensive income volatility, stripping out non-cash and non-operating distortions\n- **Why This Feature**: Distills true cash conversion efficiency by removing accounting artifacts and unrealized gains; essential measure of earnings quality\n- **Logical Meaning**: Pure cash generation capability independent of accrual accounting and mark-to-market volatility\n- **is filling nan necessary**: Unrealized gains may be zero for many firms; this is valid data\n- **Directionality**: Values consistently >1.0 indicate high-quality earnings converting to cash; <1.0 indicates aggressive revenue recognition\n- **Boundary Conditions**: Sustained values <0.5 indicate potential accounting issues; >2.0 indicates working capital optimization or deferred revenue",
+    "expression_list": [
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), subtract(vec_avg(fnd72_pit_or_is_a_is_net_inc_avail_com_shrhldrs), vec_avg(fnd72_pit_or_is_a_is_unrealized_gain_loss_comp_inc)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), subtract(vec_avg(fnd72_pit_or_is_a_is_net_inc_avail_com_shrhldrs), vec_avg(fnd72_pit_or_is_a_is_unrealized_gain_loss_comp_inc)))",
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), subtract(vec_avg(fnd72_pit_or_is_q_is_net_inc_avail_com_shrhldrs), vec_avg(fnd72_pit_or_is_a_is_unrealized_gain_loss_comp_inc)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), subtract(vec_avg(fnd72_pit_or_is_q_is_net_inc_avail_com_shrhldrs), vec_avg(fnd72_pit_or_is_a_is_unrealized_gain_loss_comp_inc)))",
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), subtract(vec_avg(fnd72_pit_or_is_q_is_net_inc_avail_com_shrhldrs), vec_avg(fnd72_pit_or_is_q_is_unrealized_gain_loss_comp_inc)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), subtract(vec_avg(fnd72_pit_or_is_q_is_net_inc_avail_com_shrhldrs), vec_avg(fnd72_pit_or_is_q_is_unrealized_gain_loss_comp_inc)))",
+        "divide(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), subtract(vec_avg(fnd72_pit_or_is_a_is_net_inc_avail_com_shrhldrs), vec_avg(fnd72_pit_or_is_q_is_unrealized_gain_loss_comp_inc)))",
+        "divide(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), subtract(vec_avg(fnd72_pit_or_is_a_is_net_inc_avail_com_shrhldrs), vec_avg(fnd72_pit_or_is_q_is_unrealized_gain_loss_comp_inc)))"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852473053891800.json

@@ -0,0 +1,22 @@
+{
+    "template": "divide(vec_avg({sales_rev_turn}), add(vec_avg({accts_rec_excl_notes_rec}), vec_avg({invtry_raw_materials}), vec_avg({invtry_in_progress})))",
+    "idea": "**Concept**: Asset Composition Efficiency\n- **Sample Fields Used**: accts_rec_excl_notes_rec, invtry_raw_materials, invtry_in_progress, sales_rev_turn\n- **Definition**: Revenue divided by sum of receivables and inventory components, measuring working capital turnover\n- **Why This Feature**: Combines multiple asset classes to assess overall working capital efficiency; low values indicate capital tied up in operations\n- **Logical Meaning**: Measures how effectively the firm converts asset investments (receivables + inventory) into revenue\n- **is filling nan necessary**: Missing inventory components should be treated as zero if not applicable to business model (e.g., service firms)\n- **Directionality**: Higher values indicate efficient asset utilization; declining trends suggest operational inefficiency\n- **Boundary Conditions**: Industry-dependent; retail typically 6-12x, manufacturing 3-6x; extreme values indicate just-in-time success or data errors",
+    "expression_list": [
+        "divide(vec_avg(fnd72_pit_or_is_a_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_a_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_a_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_a_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_q_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_a_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_a_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_a_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_a_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_a_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_a_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_q_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_q_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_a_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_a_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_q_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_a_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_a_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_q_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_a_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_q_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_a_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_q_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_a_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_a_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_a_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_q_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_q_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_q_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_a_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_q_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_q_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_a_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_q_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_q_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_q_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_q_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_q_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_q_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_q_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_a_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_q_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_q_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_a_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_q_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_q_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_q_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_a_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_a_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_q_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_a_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_q_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_q_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_q_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_a_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_q_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_a_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_q_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_a_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_a_invtry_in_progress)))",
+        "divide(vec_avg(fnd72_pit_or_is_q_sales_rev_turn), add(vec_avg(fnd72_pit_or_bs_q_bs_accts_rec_excl_notes_rec), vec_avg(fnd72_pit_or_bs_a_invtry_raw_materials), vec_avg(fnd72_pit_or_bs_a_invtry_in_progress)))"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852473617716000.json

@@ -0,0 +1,22 @@
+{
+    "template": "divide(vec_avg({st_debt}), add(vec_avg({other_st_liab}), vec_avg({def_tax_liab}), vec_avg({st_debt}), vec_avg({tot_liab_eqy})))",
+    "idea": "**Concept**: Short-Term Debt Dependency Ratio\n- **Sample Fields Used**: st_debt, other_st_liab, def_tax_liab, tot_liab_eqy\n- **Definition**: Short-term debt as proportion of total liabilities, capturing capital structure maturity profile\n- **Why This Feature**: High short-term dependency indicates refinancing risk and liquidity vulnerability; structural measure of financial risk\n- **Logical Meaning**: Decomposes liability structure to identify maturity mismatch risks in the capital stack\n- **is filling nan necessary**: Zero short-term debt is valid and should not be filled; missing total liabilities requires data validation\n- **Directionality**: Lower values indicate long-term financing security; higher values indicate reliance on rolling short-term funding\n- **Boundary Conditions**: Values >0.4 indicate dangerous short-term dependency; zero indicates conservative long-term financing",
+    "expression_list": [
+        "divide(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_a_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_a_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_a_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_a_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_q_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_q_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_q_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_a_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_q_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_q_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_q_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_q_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_q_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_a_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_a_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_q_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_q_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_a_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_q_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_tot_liab_eqy)))",
+        "divide(vec_avg(fnd72_pit_or_bs_a_bs_st_debt), add(vec_avg(fnd72_pit_or_bs_q_bs_other_st_liab), vec_avg(fnd72_pit_or_bs_a_bs_def_tax_liab), vec_avg(fnd72_pit_or_bs_a_bs_st_debt), vec_avg(fnd72_pit_or_bs_a_bs_tot_liab_eqy)))"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852474172815700.json

@@ -0,0 +1,14 @@
+{
+    "template": "quantile(divide(vec_avg({comprehensive_income}), vec_avg({net_inc_avail_com_shrhldrs})), driver=\"gaussian\", sigma=1.0)",
+    "idea": "**Concept**: Comprehensive Income Gaussian Rank\n- **Sample Fields Used**: comprehensive_income, net_inc_avail_com_shrhldrs\n- **Definition**: Cross-sectional Gaussian quantile of the ratio of comprehensive income to net income\n- **Why This Feature**: Relative positioning within universe identifies outliers in earnings quality; extreme ranks indicate unusual comprehensive income components\n- **Logical Meaning**: Relative comparison of total economic income versus reported net income across peers\n- **is filling nan necessary**: Missing comprehensive income should be excluded from ranking rather than filled\n- **Directionality**: Extreme positive ranks indicate significant OCI gains; extreme negative indicate OCI losses\n- **Boundary Conditions**: Values beyond 2 sigma indicate material divergence from typical earnings composition",
+    "expression_list": [
+        "quantile(divide(vec_avg(fnd72_pit_or_is_a_is_comprehensive_income), vec_avg(fnd72_pit_or_is_a_is_net_inc_avail_com_shrhldrs)), driver=\"gaussian\", sigma=1.0)",
+        "quantile(divide(vec_avg(fnd72_pit_or_is_a_is_comprehensive_income_per_shr), vec_avg(fnd72_pit_or_is_a_is_net_inc_avail_com_shrhldrs)), driver=\"gaussian\", sigma=1.0)",
+        "quantile(divide(vec_avg(fnd72_pit_or_is_q_is_comprehensive_income), vec_avg(fnd72_pit_or_is_a_is_net_inc_avail_com_shrhldrs)), driver=\"gaussian\", sigma=1.0)",
+        "quantile(divide(vec_avg(fnd72_pit_or_is_q_is_comprehensive_income_per_shr), vec_avg(fnd72_pit_or_is_a_is_net_inc_avail_com_shrhldrs)), driver=\"gaussian\", sigma=1.0)",
+        "quantile(divide(vec_avg(fnd72_pit_or_is_q_is_comprehensive_income), vec_avg(fnd72_pit_or_is_q_is_net_inc_avail_com_shrhldrs)), driver=\"gaussian\", sigma=1.0)",
+        "quantile(divide(vec_avg(fnd72_pit_or_is_q_is_comprehensive_income_per_shr), vec_avg(fnd72_pit_or_is_q_is_net_inc_avail_com_shrhldrs)), driver=\"gaussian\", sigma=1.0)",
+        "quantile(divide(vec_avg(fnd72_pit_or_is_a_is_comprehensive_income), vec_avg(fnd72_pit_or_is_q_is_net_inc_avail_com_shrhldrs)), driver=\"gaussian\", sigma=1.0)",
+        "quantile(divide(vec_avg(fnd72_pit_or_is_a_is_comprehensive_income_per_shr), vec_avg(fnd72_pit_or_is_q_is_net_inc_avail_com_shrhldrs)), driver=\"gaussian\", sigma=1.0)"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852474735778500.json

@@ -0,0 +1,10 @@
+{
+    "template": "subtract(ts_delta(vec_avg({chng_non_cash_work_cap}), 126), ts_delta(vec_avg({sales_rev_turn}), 126))",
+    "idea": "**Concept**: Working Capital Velocity Shift\n- **Sample Fields Used**: chng_non_cash_work_cap, sales_rev_turn\n- **Definition**: Rate of change in non-cash working capital relative to revenue growth over trailing 126 days\n- **Why This Feature**: Divergence between working capital needs and revenue growth indicates efficiency gains or deterioration in receivables/payables management\n- **Logical Meaning**: Measures whether the company is monetizing its working capital (positive divergence) or consuming cash to fund growth (negative)\n- **is filling nan necessary**: Working capital changes can be legitimately zero; fill only true missing values\n- **Directionality**: Positive divergence (working capital decreasing while revenue growing) indicates efficiency; negative suggests cash consumption\n- **Boundary Conditions**: Extreme values indicate one-time working capital events or seasonal distortions",
+    "expression_list": [
+        "subtract(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_chng_non_cash_work_cap), 126), ts_delta(vec_avg(fnd72_pit_or_is_a_sales_rev_turn), 126))",
+        "subtract(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_chng_non_cash_work_cap), 126), ts_delta(vec_avg(fnd72_pit_or_is_q_sales_rev_turn), 126))",
+        "subtract(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_chng_non_cash_work_cap), 126), ts_delta(vec_avg(fnd72_pit_or_is_a_sales_rev_turn), 126))",
+        "subtract(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_chng_non_cash_work_cap), 126), ts_delta(vec_avg(fnd72_pit_or_is_q_sales_rev_turn), 126))"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852475315478500.json

@@ -0,0 +1,14 @@
+{
+    "template": "subtract(vec_avg({eff_int_rate}), divide(vec_avg({int_expense}), ts_mean(vec_avg({tot_asset}), 126)))",
+    "idea": "**Concept**: Effective Interest Rate Spread\n- **Sample Fields Used**: eff_int_rate, int_expense, tot_asset\n- **Definition**: Effective interest rate minus implied rate (interest expense/average total assets), measuring debt cost efficiency\n- **Why This Feature**: Positive spread indicates efficient debt management vs industry; negative suggests high-cost borrowing or inefficient capital structure\n- **Logical Meaning**: Relative positioning of the firm's debt costs versus its asset scale and stated effective rates\n- **is filling nan necessary**: Ensure both rate and expense data align temporally before calculation\n- **Directionality**: Negative values indicate favorable borrowing costs relative to asset base; positive suggests expensive leverage\n- **Boundary Conditions**: Extreme deviations (>5%) indicate measurement errors or non-standard debt instruments",
+    "expression_list": [
+        "subtract(vec_avg(fnd72_pit_or_is_a_eff_int_rate), divide(vec_avg(fnd72_pit_or_is_a_is_int_expense), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 126)))",
+        "subtract(vec_avg(fnd72_pit_or_is_a_eff_int_rate), divide(vec_avg(fnd72_pit_or_is_a_is_int_expense), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 126)))",
+        "subtract(vec_avg(fnd72_pit_or_is_a_eff_int_rate), divide(vec_avg(fnd72_pit_or_is_q_is_int_expense), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 126)))",
+        "subtract(vec_avg(fnd72_pit_or_is_a_eff_int_rate), divide(vec_avg(fnd72_pit_or_is_q_is_int_expense), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 126)))",
+        "subtract(vec_avg(fnd72_pit_or_is_q_eff_int_rate), divide(vec_avg(fnd72_pit_or_is_q_is_int_expense), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 126)))",
+        "subtract(vec_avg(fnd72_pit_or_is_q_eff_int_rate), divide(vec_avg(fnd72_pit_or_is_q_is_int_expense), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 126)))",
+        "subtract(vec_avg(fnd72_pit_or_is_q_eff_int_rate), divide(vec_avg(fnd72_pit_or_is_a_is_int_expense), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 126)))",
+        "subtract(vec_avg(fnd72_pit_or_is_q_eff_int_rate), divide(vec_avg(fnd72_pit_or_is_a_is_int_expense), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 126)))"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852475912897000.json

@@ -0,0 +1,8 @@
+{
+    "template": "ts_corr(vec_avg({cash_from_oper}), ts_delay(vec_avg({cash_from_oper}), 252), 504)",
+    "idea": "**Concept**: Operating Cash Flow Persistence\n- **Sample Fields Used**: cash_from_oper\n- **Definition**: Rolling 504-day autocorrelation of operating cash flow levels\n- **Why This Feature**: Persistent operating cash flows indicate sustainable business models; erratic cash generation suggests cyclicality or working capital management issues\n- **Logical Meaning**: Captures the stability of core business cash generation independent of accounting accruals\n- **is filling nan necessary**: Quarterly gaps should be backfilled using ts_backfill to maintain continuity for time series calculations\n- **Directionality**: Higher values (closer to 1.0) indicate stable, predictable cash generation\n- **Boundary Conditions**: Negative autocorrelation suggests mean-reverting or cyclical cash flows; values >0.8 indicate exceptional stability",
+    "expression_list": [
+        "ts_corr(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), ts_delay(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), 252), 504)",
+        "ts_corr(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), ts_delay(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), 252), 504)"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852476474911100.json

@@ -0,0 +1,10 @@
+{
+    "template": "ts_sum(subtract(vec_avg({cash_from_oper}), vec_avg({cap_expend_prpty_add})), 504)",
+    "idea": "**Concept**: Cumulative Free Cash Generation\n- **Sample Fields Used**: cash_from_oper, cap_expend_prpty_add\n- **Definition**: Rolling 504-day sum of operating cash flow minus capital expenditures\n- **Why This Feature**: Cumulative free cash flow indicates long-term value creation capacity; negative accumulation signals unsustainable business model\n- **Logical Meaning**: Accumulated net cash available for shareholders after maintaining and expanding asset base\n- **is filling nan necessary**: Quarterly data requires backfilling to ensure continuous accumulation\n- **Directionality**: Positive and growing values indicate value creation; negative values indicate cash consumption\n- **Boundary Conditions**: Sustained negative accumulation over 2+ years indicates structural cash burn; extreme positive indicates cash hoarding",
+    "expression_list": [
+        "ts_sum(subtract(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add)), 504)",
+        "ts_sum(subtract(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add)), 504)",
+        "ts_sum(subtract(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add)), 504)",
+        "ts_sum(subtract(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add)), 504)"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852978914367200.json

@@ -0,0 +1,10 @@
+{
+    "template": "group_zscore(signed_power(divide(ts_delta(winsorize(ts_backfill(vec_avg({cap_expend_prpty_add}), 63), std=4), 252), ts_mean(winsorize(ts_backfill(vec_avg({tot_asset}), 63), std=4), 252)), 1.5), sector)",
+    "idea": "遵循清洗→非线性变换→组中性化逻辑链。首先对向量型资本开支与总资产数据进行63日回溯填充与4倍标准差截尾(winsorize)，消除极端脏点；计算252日同比投资加速度后，使用signed_power(1.5)放大尾部极端值（高扩张或高收缩）但保留方向符号，强化战略拐点的辨识度；最后通过sector层面的group_zscore剥离行业固有资本密集度差异，使得科技业与公用事业的投资加速信号具备可比性，适用于跨行业选股。",
+    "expression_list": [
+        "group_zscore(signed_power(divide(ts_delta(winsorize(ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 63), std=4), 252), ts_mean(winsorize(ts_backfill(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 63), std=4), 252)), 1.5), sector)",
+        "group_zscore(signed_power(divide(ts_delta(winsorize(ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 63), std=4), 252), ts_mean(winsorize(ts_backfill(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 63), std=4), 252)), 1.5), sector)",
+        "group_zscore(signed_power(divide(ts_delta(winsorize(ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 63), std=4), 252), ts_mean(winsorize(ts_backfill(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 63), std=4), 252)), 1.5), sector)",
+        "group_zscore(signed_power(divide(ts_delta(winsorize(ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 63), std=4), 252), ts_mean(winsorize(ts_backfill(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 63), std=4), 252)), 1.5), sector)"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852979426164800.json

@@ -0,0 +1,10 @@
+{
+    "template": "trade_when(or(greater(ts_rank(divide(ts_delta(vec_avg({cap_expend_prpty_add}), 252), ts_mean(vec_avg({tot_asset}), 252)), 504), 0.85), less(ts_rank(divide(ts_delta(vec_avg({cap_expend_prpty_add}), 252), ts_mean(vec_avg({tot_asset}), 252)), 504), 0.15)), group_zscore(divide(ts_delta(vec_avg({cap_expend_prpty_add}), 252), ts_mean(vec_avg({tot_asset}), 252)), industry), 0)",
+    "idea": "采用时序分位过滤与截面中性化结合。先计算投资加速度指标，再通过ts_rank观察其在过去504日（两年）历史区间中的相对位置；仅当指标处于历史极端高位（>85%分位，激进扩张）或极端低位（<15%分位，持续收缩）时，才保留经industry组zscore标准化后的信号，中间温和区间视为噪音赋值为0。该结构有效识别管理层投资行为的周期性拐点，避免在常态波动中频繁换手。",
+    "expression_list": [
+        "trade_when(or(greater(ts_rank(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), 504), 0.85), less(ts_rank(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), 504), 0.15)), group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), industry), 0)",
+        "trade_when(or(greater(ts_rank(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), 504), 0.85), less(ts_rank(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), 504), 0.15)), group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), industry), 0)",
+        "trade_when(or(greater(ts_rank(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), 504), 0.85), less(ts_rank(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), 504), 0.15)), group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), industry), 0)",
+        "trade_when(or(greater(ts_rank(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), 504), 0.85), less(ts_rank(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), 504), 0.15)), group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), industry), 0)"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852979945511100.json

@@ -0,0 +1,10 @@
+{
+    "template": "add(right_tail(group_zscore(divide(ts_delta(vec_avg({cap_expend_prpty_add}), 252), ts_mean(vec_avg({tot_asset}), 252)), industry), 1.5), left_tail(group_zscore(divide(ts_delta(vec_avg({cap_expend_prpty_add}), 252), ts_mean(vec_avg({tot_asset}), 252)), industry), -1.5))",
+    "idea": "基于边界条件逻辑构建双尾事件驱动模板。对投资加速度进行行业中性化后，使用right_tail(1.5)隔离右尾（高于1.5倍标准差的激进扩张），使用left_tail(-1.5)隔离左尾（低于-1.5倍标准差的战略收缩），两者相加后中间区间自动变为NaN。该模板严格对应原始逻辑中“>50%资产规模的激进增长”与“持续负值”的边界条件，仅在极端制度转换时建仓，适用于捕捉战略层面的资产轻重模式切换。",
+    "expression_list": [
+        "add(right_tail(group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), industry), 1.5), left_tail(group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), industry), -1.5))",
+        "add(right_tail(group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), industry), 1.5), left_tail(group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), industry), -1.5))",
+        "add(right_tail(group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), industry), 1.5), left_tail(group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), industry), -1.5))",
+        "add(right_tail(group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), industry), 1.5), left_tail(group_zscore(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), industry), -1.5))"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852980480251500.json

@@ -0,0 +1,10 @@
+{
+    "template": "divide(ts_regression(divide(vec_avg({cap_expend_prpty_add}), vec_avg({tot_asset})), ts_step(), 252, rettype=2), add(1, ts_std_dev(divide(vec_avg({cap_expend_prpty_add}), vec_avg({tot_asset})), 252)))",
+    "idea": "将简单差分升级为趋势质量度量。使用ts_regression(..., rettype=2)计算过去252日资本开支占比（capex/assets）时间序列的线性斜率，捕捉持续性投资趋势而非一次性脉冲；分母引入(1+波动率)惩罚项，对投资节奏混乱（高ts_std_dev）的公司降权，形成“趋势/噪音”信息比。该模板识别的是管理层执行战略时的坚定程度与资金配置的稳定性，而非单纯的投资规模变化。",
+    "expression_list": [
+        "divide(ts_regression(divide(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), vec_avg(fnd72_pit_or_bs_a_bs_tot_asset)), ts_step(), 252, rettype=2), add(1, ts_std_dev(divide(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), vec_avg(fnd72_pit_or_bs_a_bs_tot_asset)), 252)))",
+        "divide(ts_regression(divide(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), vec_avg(fnd72_pit_or_bs_q_bs_tot_asset)), ts_step(), 252, rettype=2), add(1, ts_std_dev(divide(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), vec_avg(fnd72_pit_or_bs_q_bs_tot_asset)), 252)))",
+        "divide(ts_regression(divide(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), vec_avg(fnd72_pit_or_bs_a_bs_tot_asset)), ts_step(), 252, rettype=2), add(1, ts_std_dev(divide(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), vec_avg(fnd72_pit_or_bs_a_bs_tot_asset)), 252)))",
+        "divide(ts_regression(divide(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), vec_avg(fnd72_pit_or_bs_q_bs_tot_asset)), ts_step(), 252, rettype=2), add(1, ts_std_dev(divide(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), vec_avg(fnd72_pit_or_bs_q_bs_tot_asset)), 252)))"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769852981007315500.json

@@ -0,0 +1,10 @@
+{
+    "template": "group_zscore(ts_decay_linear(add(multiply(divide(ts_delta(vec_avg({cap_expend_prpty_add}), 63), ts_mean(vec_avg({tot_asset}), 63)), 0.5), multiply(divide(ts_delta(vec_avg({cap_expend_prpty_add}), 126), ts_mean(vec_avg({tot_asset}), 126)), 0.3), multiply(divide(ts_delta(vec_avg({cap_expend_prpty_add}), 252), ts_mean(vec_avg({tot_asset}), 252)), 0.2)), 10), sector)",
+    "idea": "构建多horizons加权复合动量。分别计算63日（短期）、126日（中期）、252日（长期）的投资加速度，按0.5:0.3:0.2权重加权合成，兼顾即时变化与长期趋势；通过ts_decay_linear(10)进行10日线性衰减平滑，降低单日异常波动的干扰；最后经sector组zscore中性化。该结构通过时间维度上的加权衰减避免过度反应，适用于捕捉不同时间尺度的投资拐点与平滑的动量延续。",
+    "expression_list": [
+        "group_zscore(ts_decay_linear(add(multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 63), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 63)), 0.5), multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 126), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 126)), 0.3), multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), 0.2)), 10), sector)",
+        "group_zscore(ts_decay_linear(add(multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 63), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 63)), 0.5), multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 126), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 126)), 0.3), multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), 0.2)), 10), sector)",
+        "group_zscore(ts_decay_linear(add(multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 63), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 63)), 0.5), multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 126), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 126)), 0.3), multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_a_bs_tot_asset), 252)), 0.2)), 10), sector)",
+        "group_zscore(ts_decay_linear(add(multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 63), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 63)), 0.5), multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 126), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 126)), 0.3), multiply(divide(ts_delta(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add), 252), ts_mean(vec_avg(fnd72_pit_or_bs_q_bs_tot_asset), 252)), 0.2)), 10), sector)"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769854621979784200.json

@@ -0,0 +1,10 @@
+{
+    "template": "group_zscore(winsorize(ts_sum(subtract(ts_backfill(vec_avg({cash_from_oper}),63), ts_backfill(vec_avg({cap_expend_prpty_add}),63)), 504), std=4), industry)",
+    "idea": "对季度经营现金流与资本支出进行63日回溯填充以确保数据连续性，经4倍标准差Winsorize处理极端会计异常值后，计算504日滚动累计自由现金流，最后进行行业内Z-score标准化。此举消除了不同行业资本密集度差异（如重工业 vs 服务业）带来的不可比性，使得现金创造能力的衡量具备跨行业可比性，同时极端值处理防止了单次大额资产处置或并购对累积值的扭曲。",
+    "expression_list": [
+        "group_zscore(winsorize(ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add),63)), 504), std=4), industry)",
+        "group_zscore(winsorize(ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add),63)), 504), std=4), industry)",
+        "group_zscore(winsorize(ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add),63)), 504), std=4), industry)",
+        "group_zscore(winsorize(ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add),63)), 504), std=4), industry)"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769854622483457900.json

@@ -0,0 +1,10 @@
+{
+    "template": "multiply(ts_sum(subtract(vec_avg({cash_from_oper}), vec_avg({cap_expend_prpty_add})), 504), divide(1, add(1, vec_stddev(subtract({cash_from_oper}, {cap_expend_prpty_add})))))",
+    "idea": "在累计自由现金流总量的基础上，引入向量标准差衡量现金流生成的稳定性（波动性惩罚）。通过将504日累计值除以(1+现金流波动率)，对盈利质量差、现金流大起大落的公司进行降权，筛选出具有'持续、稳定'现金创造能力的标的，而非依赖一次性非经常性损益的伪高现金流公司。",
+    "expression_list": [
+        "multiply(ts_sum(subtract(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add)), 504), divide(1, add(1, vec_stddev(subtract(fnd72_pit_or_cf_a_cf_cash_from_oper, fnd72_pit_or_cf_a_cf_cap_expend_prpty_add)))))",
+        "multiply(ts_sum(subtract(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add)), 504), divide(1, add(1, vec_stddev(subtract(fnd72_pit_or_cf_q_cf_cash_from_oper, fnd72_pit_or_cf_a_cf_cap_expend_prpty_add)))))",
+        "multiply(ts_sum(subtract(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add)), 504), divide(1, add(1, vec_stddev(subtract(fnd72_pit_or_cf_a_cf_cash_from_oper, fnd72_pit_or_cf_q_cf_cap_expend_prpty_add)))))",
+        "multiply(ts_sum(subtract(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add)), 504), divide(1, add(1, vec_stddev(subtract(fnd72_pit_or_cf_q_cf_cash_from_oper, fnd72_pit_or_cf_q_cf_cap_expend_prpty_add)))))"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769854623010559800.json

@@ -0,0 +1,10 @@
+{
+    "template": "ts_zscore(subtract(ts_sum(subtract(ts_backfill(vec_avg({cash_from_oper}),63), ts_backfill(vec_avg({cap_expend_prpty_add}),63)), 126), ts_sum(subtract(ts_backfill(vec_avg({cash_from_oper}),63), ts_backfill(vec_avg({cap_expend_prpty_add}),63)), 504)), 504)",
+    "idea": "计算最近126日(半年)与最近504日(两年)累计自由现金流的差值，并进行504日时序Z-score标准化。该信号捕捉的是现金创造能力的'加速度'或'边际变化'而非绝对水平：当短期累计显著优于长期累计时，表明公司现金流生成进入改善通道；反之则提示业务恶化拐点，适用于识别财务趋势的动态反转。",
+    "expression_list": [
+        "ts_zscore(subtract(ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add),63)), 126), ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add),63)), 504)), 504)",
+        "ts_zscore(subtract(ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add),63)), 126), ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add),63)), 504)), 504)",
+        "ts_zscore(subtract(ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add),63)), 126), ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add),63)), 504)), 504)",
+        "ts_zscore(subtract(ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add),63)), 126), ts_sum(subtract(ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper),63), ts_backfill(vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add),63)), 504)), 504)"
+    ]
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769854623572902300.json

@@ -0,0 +1,5 @@
+{
+    "template": "trade_when(greater(group_rank(ts_mean({volume},21), industry), 0.2), divide(group_zscore(ts_sum(subtract(ts_backfill(vec_avg({cash_from_oper}),63), ts_backfill(vec_avg({cap_expend_prpty_add}),63)), 504), industry), ts_backfill({market_cap},63)), 0)",
+    "idea": "首先通过行业内成交量排名过滤掉流动性最差的后20%股票以控制交易成本与冲击成本；然后将经行业中性化处理的累计自由现金流除以市值，得到行业相对自由现金流收益率(FCF Yield)。该模板识别的是那些在同业中既具备强劲现金创造能力、当前估值又相对低估（市值分母）且具备足够流动性的标的，结合了质量与价值的双重逻辑。",
+    "expression_list": []
+}

cnhkmcp/untracked/APP/trailSomeAlphas/skills/brain-feature-implementation/data/fundamental72_GLB_delay1/fundamental72_GLB_1_idea_1769854624091016000.json

@@ -0,0 +1,10 @@
+{
+    "template": "ts_decay_exp_window(signed_power(group_scale(subtract(ts_sum(subtract(vec_avg({cash_from_oper}), vec_avg({cap_expend_prpty_add})), 252), ts_sum(subtract(vec_avg({cash_from_oper}), vec_avg({cap_expend_prpty_add})), 504)), industry), 2), 126)",
+    "idea": "比较最近252日(一年)与504日(两年)累计自由现金流的差异以识别近期改善趋势，通过group_scale进行行业内[0,1]线性缩放后，使用signed_power(幂次=2)非线性放大尾部信号（极好与极差的标的），最后应用126日指数衰减加权平滑。该结构强调'近期趋势优于远期'的动量特征，同时通过指数衰减降低因子换手率，并突出那些现金创造能力处于行业极端分位点的公司。",
+    "expression_list": [
+        "ts_decay_exp_window(signed_power(group_scale(subtract(ts_sum(subtract(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add)), 252), ts_sum(subtract(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add)), 504)), industry), 2), 126)",
+        "ts_decay_exp_window(signed_power(group_scale(subtract(ts_sum(subtract(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add)), 252), ts_sum(subtract(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_a_cf_cap_expend_prpty_add)), 504)), industry), 2), 126)",
+        "ts_decay_exp_window(signed_power(group_scale(subtract(ts_sum(subtract(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add)), 252), ts_sum(subtract(vec_avg(fnd72_pit_or_cf_a_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add)), 504)), industry), 2), 126)",
+        "ts_decay_exp_window(signed_power(group_scale(subtract(ts_sum(subtract(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add)), 252), ts_sum(subtract(vec_avg(fnd72_pit_or_cf_q_cf_cash_from_oper), vec_avg(fnd72_pit_or_cf_q_cf_cap_expend_prpty_add)), 504)), industry), 2), 126)"
+    ]
+}