ci: refresh schema bundle after parity anchors

Author: brycewang-stanford

schemas/agent_cards.json

@@ -4860,7 +4860,7 @@
       "Z is relevant for W given A, X (bridge first stage)"
     ],
     "category": "causal",
-    "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor).",
+    "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor). Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
     "example": "sp.proximal(df, y=\"y\", treat=\"d\", proxy_z=[\"z\"], proxy_w=[\"w\"])",
     "failure_modes": [
       {
@@ -4893,7 +4893,7 @@
     ],
     "reference": "Tchetgen Tchetgen et al. (2020); Miao, Geng & Tchetgen Tchetgen (2018)",
     "signature": {
-      "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor).",
+      "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor). Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
       "name": "proximal",
       "parameters": {
         "properties": {
@@ -4957,9 +4957,10 @@
     "stability": "stable",
     "typical_n_min": 1000,
     "validation_notes": [
+      "tests/reference_parity/test_proximal_parity.py",
       "API/unit contract evidence: tests/test_proximal.py"
     ],
-    "validation_status": "api_stable"
+    "validation_status": "validated"
   },
   {
     "alternatives": [
@@ -13694,7 +13695,7 @@
       "A bridge function exists (completeness conditions hold)."
     ],
     "category": "causal",
-    "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge.",
+    "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
     "example": "sp.fortified_pci(df, y=\"y\", treat=\"d\", proxy_z=[\"z\"], proxy_w=[\"w\"])",
     "failure_modes": [
       {
@@ -13712,7 +13713,7 @@
     ],
     "reference": "Yu, Shi & Tchetgen Tchetgen (2025). arXiv:2506.13152.",
     "signature": {
-      "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge.",
+      "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
       "name": "fortified_pci",
       "parameters": {
         "properties": {
@@ -13763,9 +13764,10 @@
     "stability": "stable",
     "typical_n_min": 500,
     "validation_notes": [
+      "tests/reference_parity/test_proximal_parity.py",
       "API/unit contract evidence: tests/test_proximal_frontiers.py"
     ],
-    "validation_status": "api_stable"
+    "validation_status": "validated"
   },
   {
     "alternatives": [
@@ -13777,7 +13779,7 @@
       "A bridge function exists (completeness conditions hold)."
     ],
     "category": "causal",
-    "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system.",
+    "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
     "example": "sp.bidirectional_pci(df, y=\"y\", treat=\"d\", proxy_z=[\"z\"], proxy_w=[\"w\"])",
     "failure_modes": [
       {
@@ -13795,7 +13797,7 @@
     ],
     "reference": "Min, Zhang & Luo (2025). arXiv:2507.13965.",
     "signature": {
-      "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system.",
+      "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
       "name": "bidirectional_pci",
       "parameters": {
         "properties": {
@@ -13846,9 +13848,10 @@
     "stability": "stable",
     "typical_n_min": 500,
     "validation_notes": [
+      "tests/reference_parity/test_proximal_parity.py",
       "API/unit contract evidence: tests/test_proximal_frontiers.py"
     ],
-    "validation_status": "api_stable"
+    "validation_status": "validated"
   },
   {
     "alternatives": [

schemas/functions.json

@@ -2595,7 +2595,7 @@
     }
   },
   {
-    "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor).",
+    "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor). Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
     "name": "proximal",
     "parameters": {
       "properties": {
@@ -7104,7 +7104,7 @@
     }
   },
   {
-    "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge.",
+    "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
     "name": "fortified_pci",
     "parameters": {
       "properties": {
@@ -7153,7 +7153,7 @@
     }
   },
   {
-    "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system.",
+    "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
     "name": "bidirectional_pci",
     "parameters": {
       "properties": {

schemas/tools.json

@@ -1510,7 +1510,7 @@
     "name": "bibtex"
   },
   {
-    "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system. Assumptions: The proxies are valid negative controls (relevant to the confounder, excluded from the causal channel); A bridge function exists (completeness conditions hold). Pre-conditions: Treatment-inducing and outcome-inducing proxy variables (negative controls) for the unobserved confounder. Failure modes: Proxies are weak or invalid -- the bridge function is poorly identified -> Test proxy relevance, select stronger proxies, or fall back to sensitivity analysis. Alternatives: sp.select_pci_proxies, sp.dml. Typical minimum N: 500.",
+    "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact). Assumptions: The proxies are valid negative controls (relevant to the confounder, excluded from the causal channel); A bridge function exists (completeness conditions hold). Pre-conditions: Treatment-inducing and outcome-inducing proxy variables (negative controls) for the unobserved confounder. Failure modes: Proxies are weak or invalid -- the bridge function is poorly identified -> Test proxy relevance, select stronger proxies, or fall back to sensitivity analysis. Alternatives: sp.select_pci_proxies, sp.dml. Typical minimum N: 500.",
     "input_schema": {
       "properties": {
         "covariates": {
@@ -7357,7 +7357,7 @@
     "name": "forest_diagnostics"
   },
   {
-    "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge. Assumptions: The proxies are valid negative controls (relevant to the confounder, excluded from the causal channel); A bridge function exists (completeness conditions hold). Pre-conditions: Treatment-inducing and outcome-inducing proxy variables (negative controls) for the unobserved confounder. Failure modes: Proxies are weak or invalid -- the bridge function is poorly identified -> Test proxy relevance, select stronger proxies, or fall back to sensitivity analysis. Alternatives: sp.select_pci_proxies, sp.dml. Typical minimum N: 500.",
+    "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact). Assumptions: The proxies are valid negative controls (relevant to the confounder, excluded from the causal channel); A bridge function exists (completeness conditions hold). Pre-conditions: Treatment-inducing and outcome-inducing proxy variables (negative controls) for the unobserved confounder. Failure modes: Proxies are weak or invalid -- the bridge function is poorly identified -> Test proxy relevance, select stronger proxies, or fall back to sensitivity analysis. Alternatives: sp.select_pci_proxies, sp.dml. Typical minimum N: 500.",
     "input_schema": {
       "properties": {
         "covariates": {
@@ -14651,7 +14651,7 @@
     "name": "propensity_score"
   },
   {
-    "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor). Assumptions: Existence of an outcome bridge function h(w, a, x) that recovers E[Y(a) | U, X]; Z and W are conditionally independent given U and (A, X); Z perp Y | U, A, X (exclusion on Z). Pre-conditions: at least one treatment-side proxy Z (independent of outcome given U, X); at least one outcome-side proxy W (independent of treatment given U, X); proxy_z and proxy_w measure the same unmeasured confounder U from different angles. Failure modes: First-stage (Z -> W) too weak -> Try richer Z or more proxies; without first-stage strength the bridge is underidentified; Proxies collapse to nearly-constant -> Proxy variation insufficient -- redesign measurement or fall back to sensitivity (sp.sensemakr); Estimate highly sensitive to bridge specification -> Report multiple bridge families; compare with sp.negative_control_outcome / _exposure. Alternatives: sp.negative_control_outcome, sp.negative_control_exposure, sp.double_negative_control, sp.iv. Typical minimum...",
+    "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor). Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact). Assumptions: Existence of an outcome bridge function h(w, a, x) that recovers E[Y(a) | U, X]; Z and W are conditionally independent given U and (A, X); Z perp Y | U, A, X (exclusion on Z). Pre-conditions: at least one treatment-side proxy Z (independent of outcome given U, X); at least one outcome-side proxy W (independent of treatment given U, X); proxy_z and proxy_w measure the same unmeasured confounder U from different angles. Failure modes: First-stage (Z -> W) too weak -> Try richer Z or more proxies; without first-stage strength the bridge is underidentified; Proxies collapse to nearly-constant -> Proxy variation insufficient -- redesign measurement or fall back to sensitivity (sp.sensemakr); Estimate highly sensitive to bridge specification -> Report multiple bridge families; compare with sp.negative_control_outcome / _exposure. Alternatives: sp.neg...",
     "input_schema": {
       "properties": {
         "bridge": {

src/statspai/schemas/agent_cards.json

@@ -4860,7 +4860,7 @@
       "Z is relevant for W given A, X (bridge first stage)"
     ],
     "category": "causal",
-    "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor).",
+    "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor). Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
     "example": "sp.proximal(df, y=\"y\", treat=\"d\", proxy_z=[\"z\"], proxy_w=[\"w\"])",
     "failure_modes": [
       {
@@ -4893,7 +4893,7 @@
     ],
     "reference": "Tchetgen Tchetgen et al. (2020); Miao, Geng & Tchetgen Tchetgen (2018)",
     "signature": {
-      "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor).",
+      "description": "Proximal Causal Inference via linear 2SLS on the outcome bridge. Identifies ATE with unmeasured confounding using two proxy variables: a treatment-side Z (instrument for W) and an outcome-side W (endogenous bridge regressor). Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
       "name": "proximal",
       "parameters": {
         "properties": {
@@ -4957,9 +4957,10 @@
     "stability": "stable",
     "typical_n_min": 1000,
     "validation_notes": [
+      "tests/reference_parity/test_proximal_parity.py",
       "API/unit contract evidence: tests/test_proximal.py"
     ],
-    "validation_status": "api_stable"
+    "validation_status": "validated"
   },
   {
     "alternatives": [
@@ -13694,7 +13695,7 @@
       "A bridge function exists (completeness conditions hold)."
     ],
     "category": "causal",
-    "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge.",
+    "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
     "example": "sp.fortified_pci(df, y=\"y\", treat=\"d\", proxy_z=[\"z\"], proxy_w=[\"w\"])",
     "failure_modes": [
       {
@@ -13712,7 +13713,7 @@
     ],
     "reference": "Yu, Shi & Tchetgen Tchetgen (2025). arXiv:2506.13152.",
     "signature": {
-      "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge.",
+      "description": "Fortified proximal causal inference (Yu, Shi & Tchetgen Tchetgen 2025). Adds a bridge-function stability constraint that gives robust ATT under mild misspecification of the outcome/treatment bridge. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
       "name": "fortified_pci",
       "parameters": {
         "properties": {
@@ -13763,9 +13764,10 @@
     "stability": "stable",
     "typical_n_min": 500,
     "validation_notes": [
+      "tests/reference_parity/test_proximal_parity.py",
       "API/unit contract evidence: tests/test_proximal_frontiers.py"
     ],
-    "validation_status": "api_stable"
+    "validation_status": "validated"
   },
   {
     "alternatives": [
@@ -13777,7 +13779,7 @@
       "A bridge function exists (completeness conditions hold)."
     ],
     "category": "causal",
-    "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system.",
+    "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
     "example": "sp.bidirectional_pci(df, y=\"y\", treat=\"d\", proxy_z=[\"z\"], proxy_w=[\"w\"])",
     "failure_modes": [
       {
@@ -13795,7 +13797,7 @@
     ],
     "reference": "Min, Zhang & Luo (2025). arXiv:2507.13965.",
     "signature": {
-      "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system.",
+      "description": "Bidirectional proximal causal inference (Min, Zhang & Luo 2025). Solves for both outcome and treatment bridges simultaneously in a single two-way regression system. Validation: validated evidence tier (known-truth, reference, external-parity, or Monte Carlo artifact).",
       "name": "bidirectional_pci",
       "parameters": {
         "properties": {
@@ -13846,9 +13848,10 @@
     "stability": "stable",
     "typical_n_min": 500,
     "validation_notes": [
+      "tests/reference_parity/test_proximal_parity.py",
       "API/unit contract evidence: tests/test_proximal_frontiers.py"
     ],
-    "validation_status": "api_stable"
+    "validation_status": "validated"
   },
   {
     "alternatives": [