npm - @openpond/evals - Versions diffs - 0.11.21 → 0.12.1 - Mend

@openpond/evals 0.11.21 → 0.12.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (311) hide show

package/dist/types/human-review/comparison.d.ts CHANGED Viewed

@@ -2,47 +2,47 @@ import { z } from 'zod';
 import { type ExperimentCaseIdentity } from '../experiments.js';
 import { ReviewReleaseSchema, type HumanReview, type HumanAttemptRefSchema } from './contracts.js';
 export declare const HumanComparisonRuleSchema: z.ZodDiscriminatedUnion<[z.ZodObject<{
+    feedbackKey: z.ZodString;
+    criterionId: z.ZodString;
     kind: z.ZodLiteral<"normalized_score">;
     minimum: z.ZodNumber;
     maximum: z.ZodNumber;
     passThreshold: z.ZodOptional<z.ZodNumber>;
+}, z.core.$strict>, z.ZodObject<{
     feedbackKey: z.ZodString;
     criterionId: z.ZodString;
-}, z.core.$strict>, z.ZodObject<{
     kind: z.ZodLiteral<"boolean">;
+}, z.core.$strict>, z.ZodObject<{
     feedbackKey: z.ZodString;
     criterionId: z.ZodString;
-}, z.core.$strict>, z.ZodObject<{
     kind: z.ZodLiteral<"category">;
+}, z.core.$strict>, z.ZodObject<{
     feedbackKey: z.ZodString;
     criterionId: z.ZodString;
-}, z.core.$strict>, z.ZodObject<{
     kind: z.ZodLiteral<"preference_score">;
     tieScore: z.ZodNullable<z.ZodNumber>;
-    feedbackKey: z.ZodString;
-    criterionId: z.ZodString;
 }, z.core.$strict>], "kind">;
 export declare const HumanComparisonSelectionsSchema: z.ZodObject<{
     rules: z.ZodArray<z.ZodDiscriminatedUnion<[z.ZodObject<{
+        feedbackKey: z.ZodString;
+        criterionId: z.ZodString;
         kind: z.ZodLiteral<"normalized_score">;
         minimum: z.ZodNumber;
         maximum: z.ZodNumber;
         passThreshold: z.ZodOptional<z.ZodNumber>;
+    }, z.core.$strict>, z.ZodObject<{
         feedbackKey: z.ZodString;
         criterionId: z.ZodString;
-    }, z.core.$strict>, z.ZodObject<{
         kind: z.ZodLiteral<"boolean">;
+    }, z.core.$strict>, z.ZodObject<{
         feedbackKey: z.ZodString;
         criterionId: z.ZodString;
-    }, z.core.$strict>, z.ZodObject<{
         kind: z.ZodLiteral<"category">;
+    }, z.core.$strict>, z.ZodObject<{
         feedbackKey: z.ZodString;
         criterionId: z.ZodString;
-    }, z.core.$strict>, z.ZodObject<{
         kind: z.ZodLiteral<"preference_score">;
         tieScore: z.ZodNullable<z.ZodNumber>;
-        feedbackKey: z.ZodString;
-        criterionId: z.ZodString;
     }, z.core.$strict>], "kind">>;
     baseline: z.ZodObject<{
         execution: z.ZodObject<{
@@ -178,10 +178,10 @@ declare const Evidence: z.ZodObject<{
                         }, z.core.$strict>;
                     }, z.core.$strict>>;
                     split: z.ZodEnum<{
+                        frozen_eval: "frozen_eval";
+                        test: "test";
                         train: "train";
                         validation: "validation";
-                        test: "test";
-                        frozen_eval: "frozen_eval";
                     }>;
                     population: z.ZodArray<z.ZodObject<{
                         taskId: z.ZodString;
@@ -198,8 +198,8 @@ declare const Evidence: z.ZodObject<{
                         configurationHash: z.ZodString;
                         output: z.ZodEnum<{
                             boolean: "boolean";
-                            score: "score";
                             category: "category";
+                            score: "score";
                         }>;
                         categories: z.ZodArray<z.ZodString>;
                     }, z.core.$strict>>;
@@ -208,17 +208,17 @@ declare const Evidence: z.ZodObject<{
                         graderId: z.ZodString;
                         fields: z.ZodArray<z.ZodObject<{
                             destination: z.ZodEnum<{
-                                output: "output";
-                                input: "input";
-                                expectedOutput: "expectedOutput";
                                 evaluatorContext: "evaluatorContext";
+                                expectedOutput: "expectedOutput";
+                                input: "input";
+                                output: "output";
                             }>;
                             path: z.ZodArray<z.ZodString>;
                             source: z.ZodEnum<{
-                                output: "output";
-                                input: "input";
-                                expectedOutput: "expectedOutput";
                                 evaluatorContext: "evaluatorContext";
+                                expectedOutput: "expectedOutput";
+                                input: "input";
+                                output: "output";
                             }>;
                             sourcePath: z.ZodArray<z.ZodString>;
                         }, z.core.$strict>>;
@@ -275,9 +275,9 @@ declare const Evidence: z.ZodObject<{
                 protocol: z.ZodLiteral<"openpond.evaluation-execution.v1">;
                 targetKind: z.ZodEnum<{
                     agent: "agent";
-                    model: "model";
                     fixture: "fixture";
                     harness: "harness";
+                    model: "model";
                 }>;
             }, z.core.$strict>>;
         }, z.core.$strict>>;
@@ -311,8 +311,8 @@ declare const Evidence: z.ZodObject<{
             configurationHash: z.ZodOptional<z.ZodString>;
             output: z.ZodEnum<{
                 boolean: "boolean";
-                score: "score";
                 category: "category";
+                score: "score";
             }>;
             categories: z.ZodArray<z.ZodString>;
         }, z.core.$strict>>;
@@ -332,8 +332,8 @@ declare const Evidence: z.ZodObject<{
         }, z.core.$strict>;
         status: z.ZodEnum<{
             cancelled: "cancelled";
-            failed: "failed";
             completed: "completed";
+            failed: "failed";
         }>;
         cases: z.ZodArray<z.ZodObject<{
             identity: z.ZodObject<{
@@ -343,8 +343,8 @@ declare const Evidence: z.ZodObject<{
             }, z.core.$strict>;
             status: z.ZodEnum<{
                 cancelled: "cancelled";
-                failed: "failed";
                 completed: "completed";
+                failed: "failed";
                 unavailable: "unavailable";
             }>;
             output: z.ZodNullable<z.ZodUnknown>;
@@ -360,8 +360,8 @@ declare const Evidence: z.ZodObject<{
                     revision: z.ZodUnion<readonly [z.ZodNumber, z.ZodString]>;
                 }, z.core.$strict>;
                 status: z.ZodEnum<{
-                    pending: "pending";
                     failed: "failed";
+                    pending: "pending";
                     scored: "scored";
                     unavailable: "unavailable";
                 }>;
@@ -401,25 +401,25 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
     projectId: z.ZodString;
     selections: z.ZodObject<{
         rules: z.ZodArray<z.ZodDiscriminatedUnion<[z.ZodObject<{
+            feedbackKey: z.ZodString;
+            criterionId: z.ZodString;
             kind: z.ZodLiteral<"normalized_score">;
             minimum: z.ZodNumber;
             maximum: z.ZodNumber;
             passThreshold: z.ZodOptional<z.ZodNumber>;
+        }, z.core.$strict>, z.ZodObject<{
             feedbackKey: z.ZodString;
             criterionId: z.ZodString;
-        }, z.core.$strict>, z.ZodObject<{
             kind: z.ZodLiteral<"boolean">;
+        }, z.core.$strict>, z.ZodObject<{
             feedbackKey: z.ZodString;
             criterionId: z.ZodString;
-        }, z.core.$strict>, z.ZodObject<{
             kind: z.ZodLiteral<"category">;
+        }, z.core.$strict>, z.ZodObject<{
             feedbackKey: z.ZodString;
             criterionId: z.ZodString;
-        }, z.core.$strict>, z.ZodObject<{
             kind: z.ZodLiteral<"preference_score">;
             tieScore: z.ZodNullable<z.ZodNumber>;
-            feedbackKey: z.ZodString;
-            criterionId: z.ZodString;
         }, z.core.$strict>], "kind">>;
         baseline: z.ZodObject<{
             execution: z.ZodObject<{
@@ -506,17 +506,17 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                     pairwise: "pairwise";
                 }>;
                 origin: z.ZodEnum<{
-                    owner_attested_local: "owner_attested_local";
+                    hosted_retained: "hosted_retained";
                     native_local: "native_local";
+                    owner_attested_local: "owner_attested_local";
                     recorded_evidence: "recorded_evidence";
-                    hosted_retained: "hosted_retained";
                 }>;
                 status: z.ZodEnum<{
-                    submitted: "submitted";
                     accepted: "accepted";
-                    rejected: "rejected";
                     cancelled: "cancelled";
                     pending: "pending";
+                    rejected: "rejected";
+                    submitted: "submitted";
                 }>;
                 grader: z.ZodObject<{
                     id: z.ZodString;
@@ -529,8 +529,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                     abstain: z.ZodBoolean;
                     note: z.ZodString;
                     strength: z.ZodOptional<z.ZodEnum<{
-                        slight: "slight";
                         moderate: "moderate";
+                        slight: "slight";
                         strong: "strong";
                     }>>;
                 }, z.core.$strict>>;
@@ -570,17 +570,17 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                     pairwise: "pairwise";
                 }>;
                 origin: z.ZodEnum<{
-                    owner_attested_local: "owner_attested_local";
+                    hosted_retained: "hosted_retained";
                     native_local: "native_local";
+                    owner_attested_local: "owner_attested_local";
                     recorded_evidence: "recorded_evidence";
-                    hosted_retained: "hosted_retained";
                 }>;
                 status: z.ZodEnum<{
-                    submitted: "submitted";
                     accepted: "accepted";
-                    rejected: "rejected";
                     cancelled: "cancelled";
                     pending: "pending";
+                    rejected: "rejected";
+                    submitted: "submitted";
                 }>;
                 grader: z.ZodObject<{
                     id: z.ZodString;
@@ -593,8 +593,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                     abstain: z.ZodBoolean;
                     note: z.ZodString;
                     strength: z.ZodOptional<z.ZodEnum<{
-                        slight: "slight";
                         moderate: "moderate";
+                        slight: "slight";
                         strong: "strong";
                     }>>;
                 }, z.core.$strict>>;
@@ -696,10 +696,10 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                             }, z.core.$strict>;
                         }, z.core.$strict>>;
                         split: z.ZodEnum<{
+                            frozen_eval: "frozen_eval";
+                            test: "test";
                             train: "train";
                             validation: "validation";
-                            test: "test";
-                            frozen_eval: "frozen_eval";
                         }>;
                         population: z.ZodArray<z.ZodObject<{
                             taskId: z.ZodString;
@@ -716,8 +716,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                             configurationHash: z.ZodString;
                             output: z.ZodEnum<{
                                 boolean: "boolean";
-                                score: "score";
                                 category: "category";
+                                score: "score";
                             }>;
                             categories: z.ZodArray<z.ZodString>;
                         }, z.core.$strict>>;
@@ -726,17 +726,17 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                             graderId: z.ZodString;
                             fields: z.ZodArray<z.ZodObject<{
                                 destination: z.ZodEnum<{
-                                    output: "output";
-                                    input: "input";
-                                    expectedOutput: "expectedOutput";
                                     evaluatorContext: "evaluatorContext";
+                                    expectedOutput: "expectedOutput";
+                                    input: "input";
+                                    output: "output";
                                 }>;
                                 path: z.ZodArray<z.ZodString>;
                                 source: z.ZodEnum<{
-                                    output: "output";
-                                    input: "input";
-                                    expectedOutput: "expectedOutput";
                                     evaluatorContext: "evaluatorContext";
+                                    expectedOutput: "expectedOutput";
+                                    input: "input";
+                                    output: "output";
                                 }>;
                                 sourcePath: z.ZodArray<z.ZodString>;
                             }, z.core.$strict>>;
@@ -793,9 +793,9 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                     protocol: z.ZodLiteral<"openpond.evaluation-execution.v1">;
                     targetKind: z.ZodEnum<{
                         agent: "agent";
-                        model: "model";
                         fixture: "fixture";
                         harness: "harness";
+                        model: "model";
                     }>;
                 }, z.core.$strict>>;
             }, z.core.$strict>>;
@@ -829,8 +829,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                 configurationHash: z.ZodOptional<z.ZodString>;
                 output: z.ZodEnum<{
                     boolean: "boolean";
-                    score: "score";
                     category: "category";
+                    score: "score";
                 }>;
                 categories: z.ZodArray<z.ZodString>;
             }, z.core.$strict>>;
@@ -850,8 +850,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
             }, z.core.$strict>;
             status: z.ZodEnum<{
                 cancelled: "cancelled";
-                failed: "failed";
                 completed: "completed";
+                failed: "failed";
             }>;
             cases: z.ZodArray<z.ZodObject<{
                 identity: z.ZodObject<{
@@ -861,8 +861,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                 }, z.core.$strict>;
                 status: z.ZodEnum<{
                     cancelled: "cancelled";
-                    failed: "failed";
                     completed: "completed";
+                    failed: "failed";
                     unavailable: "unavailable";
                 }>;
                 output: z.ZodNullable<z.ZodUnknown>;
@@ -878,8 +878,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                         revision: z.ZodUnion<readonly [z.ZodNumber, z.ZodString]>;
                     }, z.core.$strict>;
                     status: z.ZodEnum<{
-                        pending: "pending";
                         failed: "failed";
+                        pending: "pending";
                         scored: "scored";
                         unavailable: "unavailable";
                     }>;
@@ -1007,10 +1007,10 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                             }, z.core.$strict>;
                         }, z.core.$strict>>;
                         split: z.ZodEnum<{
+                            frozen_eval: "frozen_eval";
+                            test: "test";
                             train: "train";
                             validation: "validation";
-                            test: "test";
-                            frozen_eval: "frozen_eval";
                         }>;
                         population: z.ZodArray<z.ZodObject<{
                             taskId: z.ZodString;
@@ -1027,8 +1027,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                             configurationHash: z.ZodString;
                             output: z.ZodEnum<{
                                 boolean: "boolean";
-                                score: "score";
                                 category: "category";
+                                score: "score";
                             }>;
                             categories: z.ZodArray<z.ZodString>;
                         }, z.core.$strict>>;
@@ -1037,17 +1037,17 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                             graderId: z.ZodString;
                             fields: z.ZodArray<z.ZodObject<{
                                 destination: z.ZodEnum<{
-                                    output: "output";
-                                    input: "input";
-                                    expectedOutput: "expectedOutput";
                                     evaluatorContext: "evaluatorContext";
+                                    expectedOutput: "expectedOutput";
+                                    input: "input";
+                                    output: "output";
                                 }>;
                                 path: z.ZodArray<z.ZodString>;
                                 source: z.ZodEnum<{
-                                    output: "output";
-                                    input: "input";
-                                    expectedOutput: "expectedOutput";
                                     evaluatorContext: "evaluatorContext";
+                                    expectedOutput: "expectedOutput";
+                                    input: "input";
+                                    output: "output";
                                 }>;
                                 sourcePath: z.ZodArray<z.ZodString>;
                             }, z.core.$strict>>;
@@ -1104,9 +1104,9 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                     protocol: z.ZodLiteral<"openpond.evaluation-execution.v1">;
                     targetKind: z.ZodEnum<{
                         agent: "agent";
-                        model: "model";
                         fixture: "fixture";
                         harness: "harness";
+                        model: "model";
                     }>;
                 }, z.core.$strict>>;
             }, z.core.$strict>>;
@@ -1140,8 +1140,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                 configurationHash: z.ZodOptional<z.ZodString>;
                 output: z.ZodEnum<{
                     boolean: "boolean";
-                    score: "score";
                     category: "category";
+                    score: "score";
                 }>;
                 categories: z.ZodArray<z.ZodString>;
             }, z.core.$strict>>;
@@ -1161,8 +1161,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
             }, z.core.$strict>;
             status: z.ZodEnum<{
                 cancelled: "cancelled";
-                failed: "failed";
                 completed: "completed";
+                failed: "failed";
             }>;
             cases: z.ZodArray<z.ZodObject<{
                 identity: z.ZodObject<{
@@ -1172,8 +1172,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                 }, z.core.$strict>;
                 status: z.ZodEnum<{
                     cancelled: "cancelled";
-                    failed: "failed";
                     completed: "completed";
+                    failed: "failed";
                     unavailable: "unavailable";
                 }>;
                 output: z.ZodNullable<z.ZodUnknown>;
@@ -1189,8 +1189,8 @@ export declare const HumanComparisonProjectionSchema: z.ZodObject<{
                         revision: z.ZodUnion<readonly [z.ZodNumber, z.ZodString]>;
                     }, z.core.$strict>;
                     status: z.ZodEnum<{
-                        pending: "pending";
                         failed: "failed";
+                        pending: "pending";
                         scored: "scored";
                         unavailable: "unavailable";
                     }>;
@@ -1250,25 +1250,25 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
     projectId: string;
     selections: {
         rules: ({
+            feedbackKey: string;
+            criterionId: string;
             kind: "normalized_score";
             minimum: number;
             maximum: number;
-            feedbackKey: string;
-            criterionId: string;
             passThreshold?: number | undefined;
         } | {
-            kind: "boolean";
             feedbackKey: string;
             criterionId: string;
+            kind: "boolean";
         } | {
-            kind: "category";
             feedbackKey: string;
             criterionId: string;
+            kind: "category";
         } | {
-            kind: "preference_score";
-            tieScore: number | null;
             feedbackKey: string;
             criterionId: string;
+            kind: "preference_score";
+            tieScore: number | null;
         })[];
         baseline: {
             execution: {
@@ -1351,8 +1351,8 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                 } | null;
                 attemptIds: string[];
                 mode: "individual" | "pairwise";
-                origin: "owner_attested_local" | "native_local" | "recorded_evidence" | "hosted_retained";
-                status: "submitted" | "accepted" | "rejected" | "cancelled" | "pending";
+                origin: "hosted_retained" | "native_local" | "owner_attested_local" | "recorded_evidence";
+                status: "accepted" | "cancelled" | "pending" | "rejected" | "submitted";
                 grader: {
                     id: string;
                     revision: number;
@@ -1363,7 +1363,7 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                     value: string | number | boolean | null;
                     abstain: boolean;
                     note: string;
-                    strength?: "slight" | "moderate" | "strong" | undefined;
+                    strength?: "moderate" | "slight" | "strong" | undefined;
                 }[];
             }[];
             contentHash: string;
@@ -1397,8 +1397,8 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                 } | null;
                 attemptIds: string[];
                 mode: "individual" | "pairwise";
-                origin: "owner_attested_local" | "native_local" | "recorded_evidence" | "hosted_retained";
-                status: "submitted" | "accepted" | "rejected" | "cancelled" | "pending";
+                origin: "hosted_retained" | "native_local" | "owner_attested_local" | "recorded_evidence";
+                status: "accepted" | "cancelled" | "pending" | "rejected" | "submitted";
                 grader: {
                     id: string;
                     revision: number;
@@ -1409,7 +1409,7 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                     value: string | number | boolean | null;
                     abstain: boolean;
                     note: string;
-                    strength?: "slight" | "moderate" | "strong" | undefined;
+                    strength?: "moderate" | "slight" | "strong" | undefined;
                 }[];
             }[];
             contentHash: string;
@@ -1480,7 +1480,35 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                         };
                         declaredProfileCatalogHash: string | null;
                         protectedProfileClosureHash: string;
-                        split: "train" | "validation" | "test" | "frozen_eval";
+                        recordedOrigin?: {
+                            execution: {
+                                id: string;
+                                contentHash: string;
+                            };
+                            originalEvidence: {
+                                id: string;
+                                contentHash: string;
+                            };
+                            members: {
+                                caseId: string;
+                                seed: string;
+                                fixtureId: string;
+                                sourceId: string;
+                                snapshotHash: string;
+                                boundaryRevisionHash: string;
+                            }[];
+                        } | undefined;
+                        gradingSource?: {
+                            pass: {
+                                id: string;
+                                contentHash: string;
+                            };
+                            evidence: {
+                                id: string;
+                                contentHash: string;
+                            };
+                        } | undefined;
+                        split: "frozen_eval" | "test" | "train" | "validation";
                         population: {
                             taskId: string;
                             seed: string;
@@ -1494,16 +1522,16 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                             };
                             feedbackKey: string;
                             configurationHash: string;
-                            output: "boolean" | "score" | "category";
+                            output: "boolean" | "category" | "score";
                             categories: string[];
                         }[];
                         fieldMappingsHash: string;
                         fieldMappings: {
                             graderId: string;
                             fields: {
-                                destination: "output" | "input" | "expectedOutput" | "evaluatorContext";
+                                destination: "evaluatorContext" | "expectedOutput" | "input" | "output";
                                 path: string[];
-                                source: "output" | "input" | "expectedOutput" | "evaluatorContext";
+                                source: "evaluatorContext" | "expectedOutput" | "input" | "output";
                                 sourcePath: string[];
                             }[];
                         }[];
@@ -1514,34 +1542,6 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                             requireComplete: boolean;
                         };
                         contentHash: string;
-                        recordedOrigin?: {
-                            execution: {
-                                id: string;
-                                contentHash: string;
-                            };
-                            originalEvidence: {
-                                id: string;
-                                contentHash: string;
-                            };
-                            members: {
-                                caseId: string;
-                                seed: string;
-                                fixtureId: string;
-                                sourceId: string;
-                                snapshotHash: string;
-                                boundaryRevisionHash: string;
-                            }[];
-                        } | undefined;
-                        gradingSource?: {
-                            pass: {
-                                id: string;
-                                contentHash: string;
-                            };
-                            evidence: {
-                                id: string;
-                                contentHash: string;
-                            };
-                        } | undefined;
                     } | undefined;
                     sourceCandidate?: {
                         candidateId: string;
@@ -1579,31 +1579,13 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                 kind: "fixture";
                 configurationHash: string;
             };
-            evaluators: {
-                release: {
-                    id: string;
-                    contentHash: string;
-                    revision: string | number;
-                };
-                feedbackKey: string;
-                output: "boolean" | "score" | "category";
-                categories: string[];
-                configurationHash?: string | undefined;
-            }[];
-            population: {
-                caseId: string;
-                seed: string;
-                fixtureId: string | null;
-            }[];
-            createdAt: string;
-            contentHash: string;
             execution?: {
                 packageHash: string;
                 runtimeTargetHash: string;
                 metricPolicyHash: string;
                 compatibility?: {
                     protocol: "openpond.evaluation-execution.v1";
-                    targetKind: "agent" | "model" | "fixture" | "harness";
+                    targetKind: "agent" | "fixture" | "harness" | "model";
                 } | undefined;
             } | undefined;
             lineage?: {
@@ -1626,6 +1608,24 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                     rulesHash: string;
                 } | undefined;
             } | undefined;
+            evaluators: {
+                release: {
+                    id: string;
+                    contentHash: string;
+                    revision: string | number;
+                };
+                feedbackKey: string;
+                configurationHash?: string | undefined;
+                output: "boolean" | "category" | "score";
+                categories: string[];
+            }[];
+            population: {
+                caseId: string;
+                seed: string;
+                fixtureId: string | null;
+            }[];
+            createdAt: string;
+            contentHash: string;
         };
         result: {
             schemaVersion: "openpond.experimentResult.v1";
@@ -1633,14 +1633,14 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                 id: string;
                 contentHash: string;
             };
-            status: "cancelled" | "failed" | "completed";
+            status: "cancelled" | "completed" | "failed";
             cases: {
                 identity: {
                     caseId: string;
                     seed: string;
                     fixtureId: string | null;
                 };
-                status: "cancelled" | "failed" | "completed" | "unavailable";
+                status: "cancelled" | "completed" | "failed" | "unavailable";
                 output: unknown;
                 error: {
                     code: string;
@@ -1653,8 +1653,9 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                         contentHash: string;
                         revision: string | number;
                     };
-                    status: "pending" | "failed" | "scored" | "unavailable";
+                    status: "failed" | "pending" | "scored" | "unavailable";
                     value: string | number | boolean | null;
+                    passed?: boolean | null | undefined;
                     reasoning: string | null;
                     evidenceRefs: {
                         id: string;
@@ -1662,7 +1663,6 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                         mediaType: string | null;
                         sizeBytes: number | null;
                     }[];
-                    passed?: boolean | null | undefined;
                 }[];
                 usage: {
                     inputTokens: number | null;
@@ -1749,7 +1749,35 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                         };
                         declaredProfileCatalogHash: string | null;
                         protectedProfileClosureHash: string;
-                        split: "train" | "validation" | "test" | "frozen_eval";
+                        recordedOrigin?: {
+                            execution: {
+                                id: string;
+                                contentHash: string;
+                            };
+                            originalEvidence: {
+                                id: string;
+                                contentHash: string;
+                            };
+                            members: {
+                                caseId: string;
+                                seed: string;
+                                fixtureId: string;
+                                sourceId: string;
+                                snapshotHash: string;
+                                boundaryRevisionHash: string;
+                            }[];
+                        } | undefined;
+                        gradingSource?: {
+                            pass: {
+                                id: string;
+                                contentHash: string;
+                            };
+                            evidence: {
+                                id: string;
+                                contentHash: string;
+                            };
+                        } | undefined;
+                        split: "frozen_eval" | "test" | "train" | "validation";
                         population: {
                             taskId: string;
                             seed: string;
@@ -1763,16 +1791,16 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                             };
                             feedbackKey: string;
                             configurationHash: string;
-                            output: "boolean" | "score" | "category";
+                            output: "boolean" | "category" | "score";
                             categories: string[];
                         }[];
                         fieldMappingsHash: string;
                         fieldMappings: {
                             graderId: string;
                             fields: {
-                                destination: "output" | "input" | "expectedOutput" | "evaluatorContext";
+                                destination: "evaluatorContext" | "expectedOutput" | "input" | "output";
                                 path: string[];
-                                source: "output" | "input" | "expectedOutput" | "evaluatorContext";
+                                source: "evaluatorContext" | "expectedOutput" | "input" | "output";
                                 sourcePath: string[];
                             }[];
                         }[];
@@ -1783,34 +1811,6 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                             requireComplete: boolean;
                         };
                         contentHash: string;
-                        recordedOrigin?: {
-                            execution: {
-                                id: string;
-                                contentHash: string;
-                            };
-                            originalEvidence: {
-                                id: string;
-                                contentHash: string;
-                            };
-                            members: {
-                                caseId: string;
-                                seed: string;
-                                fixtureId: string;
-                                sourceId: string;
-                                snapshotHash: string;
-                                boundaryRevisionHash: string;
-                            }[];
-                        } | undefined;
-                        gradingSource?: {
-                            pass: {
-                                id: string;
-                                contentHash: string;
-                            };
-                            evidence: {
-                                id: string;
-                                contentHash: string;
-                            };
-                        } | undefined;
                     } | undefined;
                     sourceCandidate?: {
                         candidateId: string;
@@ -1848,31 +1848,13 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                 kind: "fixture";
                 configurationHash: string;
             };
-            evaluators: {
-                release: {
-                    id: string;
-                    contentHash: string;
-                    revision: string | number;
-                };
-                feedbackKey: string;
-                output: "boolean" | "score" | "category";
-                categories: string[];
-                configurationHash?: string | undefined;
-            }[];
-            population: {
-                caseId: string;
-                seed: string;
-                fixtureId: string | null;
-            }[];
-            createdAt: string;
-            contentHash: string;
             execution?: {
                 packageHash: string;
                 runtimeTargetHash: string;
                 metricPolicyHash: string;
                 compatibility?: {
                     protocol: "openpond.evaluation-execution.v1";
-                    targetKind: "agent" | "model" | "fixture" | "harness";
+                    targetKind: "agent" | "fixture" | "harness" | "model";
                 } | undefined;
             } | undefined;
             lineage?: {
@@ -1895,6 +1877,24 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                     rulesHash: string;
                 } | undefined;
             } | undefined;
+            evaluators: {
+                release: {
+                    id: string;
+                    contentHash: string;
+                    revision: string | number;
+                };
+                feedbackKey: string;
+                configurationHash?: string | undefined;
+                output: "boolean" | "category" | "score";
+                categories: string[];
+            }[];
+            population: {
+                caseId: string;
+                seed: string;
+                fixtureId: string | null;
+            }[];
+            createdAt: string;
+            contentHash: string;
         };
         result: {
             schemaVersion: "openpond.experimentResult.v1";
@@ -1902,14 +1902,14 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                 id: string;
                 contentHash: string;
             };
-            status: "cancelled" | "failed" | "completed";
+            status: "cancelled" | "completed" | "failed";
             cases: {
                 identity: {
                     caseId: string;
                     seed: string;
                     fixtureId: string | null;
                 };
-                status: "cancelled" | "failed" | "completed" | "unavailable";
+                status: "cancelled" | "completed" | "failed" | "unavailable";
                 output: unknown;
                 error: {
                     code: string;
@@ -1922,8 +1922,9 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                         contentHash: string;
                         revision: string | number;
                     };
-                    status: "pending" | "failed" | "scored" | "unavailable";
+                    status: "failed" | "pending" | "scored" | "unavailable";
                     value: string | number | boolean | null;
+                    passed?: boolean | null | undefined;
                     reasoning: string | null;
                     evidenceRefs: {
                         id: string;
@@ -1931,7 +1932,6 @@ export declare function verifyHumanComparisonProjection(raw: unknown, originals?
                         mediaType: string | null;
                         sizeBytes: number | null;
                     }[];
-                    passed?: boolean | null | undefined;
                 }[];
                 usage: {
                     inputTokens: number | null;