@piagent/platform 1.6.0 → 1.7.1

package/evals/architecture-conformance-v1/runner.mjs

@@ -0,0 +1,77 @@
+#!/usr/bin/env node
+import fs from "node:fs";
+import path from "node:path";
+import { fileURLToPath } from "node:url";
+
+import { inspectArchitecture } from "../../scripts/check-architecture.mjs";
+
+const laneRoot = path.dirname(fileURLToPath(import.meta.url));
+const lane = JSON.parse(fs.readFileSync(path.join(laneRoot, "lane.json"), "utf8"));
+const argumentsList = process.argv.slice(2);
+const outputIndex = argumentsList.indexOf("--output");
+const outputArgument = outputIndex >= 0 ? argumentsList[outputIndex + 1] : undefined;
+if (outputIndex >= 0 && !outputArgument) throw new Error("--output requires a path");
+const outputValueIndex = outputIndex >= 0 ? outputIndex + 1 : -1;
+const unknown = argumentsList.filter((_, index) => index !== outputIndex && index !== outputValueIndex);
+if (unknown.length > 0) throw new Error(`Unknown option: ${unknown[0]}`);
+const requiredGates = {
+  architectureCheckPasses: true,
+  sourceCoverageComplete: true,
+  dependencyBoundariesPass: true,
+  lineBudgetsPass: true,
+  providerCalls: 0,
+  modelTokens: 0
+};
+if (lane?.schemaVersion !== 1 || lane.id !== "architecture-conformance-v1"
+  || lane.providerRequired !== false || lane.modelTokensExpected !== 0
+  || lane.checker !== "scripts/check-architecture.mjs"
+  || lane.configuration !== "architecture/layers.json"
+  || JSON.stringify(lane.releaseGates) !== JSON.stringify(requiredGates)
+  || typeof lane.claimBoundary !== "string" || lane.claimBoundary.length === 0) {
+  throw new Error("Unsupported architecture conformance lane configuration");
+}
+
+let inspection;
+let failure = null;
+try {
+  inspection = inspectArchitecture();
+} catch (error) {
+  failure = error instanceof Error ? error.message : String(error);
+  inspection = { ok: false, errors: [failure], files: 0, layers: {} };
+}
+const errors = Array.isArray(inspection.errors) ? inspection.errors : ["invalid-architecture-result"];
+const gates = {
+  architectureCheckPasses: inspection.ok === true && errors.length === 0,
+  sourceCoverageComplete: errors.every((error) => !String(error).includes("source file is not assigned")),
+  dependencyBoundariesPass: errors.every((error) => !String(error).includes(" cannot import ")),
+  lineBudgetsPass: errors.every((error) => !String(error).includes(" lines exceeds ")),
+  providerCalls: 0,
+  modelTokens: 0
+};
+const passed = failure === null
+  && Object.entries(gates).every(([name, value]) => name === "providerCalls" || name === "modelTokens" ? value === 0 : value === true);
+const report = {
+  schemaVersion: 1,
+  laneId: lane.id,
+  evidenceClass: "provider-free-architecture-conformance",
+  passed,
+  provider: { required: false, used: false, calls: 0, modelTokens: 0 },
+  architecture: {
+    filesChecked: Number.isSafeInteger(inspection.files) ? inspection.files : 0,
+    layersChecked: Object.keys(inspection.layers ?? {}).length,
+    errors: errors.map(String)
+  },
+  gates,
+  claimBoundary: lane.claimBoundary
+};
+
+if (outputArgument) {
+  const outputPath = path.resolve(outputArgument);
+  fs.mkdirSync(path.dirname(outputPath), { recursive: true, mode: 0o700 });
+  fs.writeFileSync(outputPath, `${JSON.stringify(report, null, 2)}\n`, { mode: 0o600 });
+}
+if (passed) process.stdout.write(`${JSON.stringify(report, null, 2)}\n`);
+else {
+  process.stderr.write(`${JSON.stringify(report, null, 2)}\n`);
+  process.exitCode = 1;
+}

package/evals/fixtures/approved-host-contracts.invalid.json

@@ -0,0 +1,43 @@
+{
+  "version": "approved-host-contracts-v1",
+  "projectId": "cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc",
+  "operatorRequestDigest": "operator-request-v1:aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+  "verifierDigest": "dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd",
+  "backend": {
+    "imageId": "sha256:bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb",
+    "dockerSocket": "/fixture-only/engine.sock",
+    "timeoutMs": 30001
+  },
+  "contracts": [
+    {
+      "criterionId": "identity-contract",
+      "criterionHash": "eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee",
+      "sourcePath": "src/identity.js",
+      "exportName": "identity",
+      "maxAttempts": 2,
+      "checks": [
+        {
+          "id": "identity",
+          "cases": [
+            {
+              "id": "positive-number",
+              "args": [
+                {
+                  "type": "number",
+                  "value": 7
+                }
+              ],
+              "expected": {
+                "outcome": "return",
+                "value": {
+                  "type": "number",
+                  "value": 7
+                }
+              }
+            }
+          ]
+        }
+      ]
+    }
+  ]
+}

package/evals/fixtures/approved-host-contracts.valid.json

@@ -0,0 +1,43 @@
+{
+  "version": "approved-host-contracts-v1",
+  "projectId": "cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc",
+  "operatorRequestDigest": "operator-request-v1:aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+  "verifierDigest": "dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd",
+  "backend": {
+    "imageId": "sha256:bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb",
+    "dockerSocket": "/fixture-only/engine.sock",
+    "timeoutMs": 1000
+  },
+  "contracts": [
+    {
+      "criterionId": "identity-contract",
+      "criterionHash": "eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee",
+      "sourcePath": "src/identity.js",
+      "exportName": "identity",
+      "maxAttempts": 2,
+      "checks": [
+        {
+          "id": "identity",
+          "cases": [
+            {
+              "id": "positive-number",
+              "args": [
+                {
+                  "type": "number",
+                  "value": 7
+                }
+              ],
+              "expected": {
+                "outcome": "return",
+                "value": {
+                  "type": "number",
+                  "value": 7
+                }
+              }
+            }
+          ]
+        }
+      ]
+    }
+  ]
+}

package/evals/fixtures/benchmark-assurance-evidence.invalid.json

@@ -29,7 +29,7 @@
   },
   "disjointness": {
     "taxonomyDigest": "d42458d04861a60488a5f98532936db71d43dcccc0298d75b09c2255f2922cb1",
-    "publicExposureDigest": "629c1e9f835464b5e3db1c6daac67d8f2661c691ac10201b904e4ed8b377dbef",
+    "publicExposureDigest": "80bed5d7ffd39dd9ebbc3207d5e6f09e9e56b9b254fd55814e6225677cb727f5",
     "reportDigest": "5555555555555555555555555555555555555555555555555555555555555555",
     "familyDisjoint": true,
     "repositoryDisjoint": false,

package/evals/fixtures/benchmark-assurance-evidence.valid.json

@@ -29,7 +29,7 @@
   },
   "disjointness": {
     "taxonomyDigest": "d42458d04861a60488a5f98532936db71d43dcccc0298d75b09c2255f2922cb1",
-    "publicExposureDigest": "629c1e9f835464b5e3db1c6daac67d8f2661c691ac10201b904e4ed8b377dbef",
+    "publicExposureDigest": "80bed5d7ffd39dd9ebbc3207d5e6f09e9e56b9b254fd55814e6225677cb727f5",
     "reportDigest": "5555555555555555555555555555555555555555555555555555555555555555",
     "familyDisjoint": true,
     "repositoryDisjoint": true,

package/evals/fixtures/benchmark-attempt-outcome.invalid.json

@@ -0,0 +1,42 @@
+{
+  "schemaVersion": 3,
+  "contractId": "benchmark-attempt-outcome-v3",
+  "attemptId": "golden-exit-zero-error",
+  "scenarioKind": "source-change",
+  "transportStatus": "completed",
+  "operationStatus": "completed",
+  "taskStatus": "completed",
+  "semanticStatus": "pass",
+  "gradeStatus": "pass",
+  "runValidity": "valid",
+  "usageStatus": "exact",
+  "failureClass": "none",
+  "countsTowardQuality": true,
+  "countsTowardUsage": true,
+  "evidence": {
+    "providerStarted": true,
+    "processExitCode": 0,
+    "threadIdPresent": true,
+    "usageReported": true,
+    "terminalAgentMessage": true,
+    "requiredOutputEvidencePresent": true,
+    "errorEvents": 1,
+    "turnFailedEvents": 0,
+    "itemErrorEvents": 0,
+    "failedCommandEvents": 0,
+    "fileChangeCount": 1,
+    "mutationExpected": true,
+    "refusal": null
+  },
+  "usage": {
+    "providerInput": 1200,
+    "cacheRead": 800,
+    "cacheWrite": 0,
+    "output": 200,
+    "reasoning": 50,
+    "fresh": 600,
+    "totalTraffic": 1400,
+    "billedCost": null,
+    "billedCostStatus": "unavailable"
+  }
+}

package/evals/fixtures/benchmark-attempt-outcome.valid.json

@@ -0,0 +1,42 @@
+{
+  "schemaVersion": 3,
+  "contractId": "benchmark-attempt-outcome-v3",
+  "attemptId": "golden-source-change",
+  "scenarioKind": "source-change",
+  "transportStatus": "completed",
+  "operationStatus": "completed",
+  "taskStatus": "completed",
+  "semanticStatus": "pass",
+  "gradeStatus": "pass",
+  "runValidity": "valid",
+  "usageStatus": "exact",
+  "failureClass": "none",
+  "countsTowardQuality": true,
+  "countsTowardUsage": true,
+  "evidence": {
+    "providerStarted": true,
+    "processExitCode": 0,
+    "threadIdPresent": true,
+    "usageReported": true,
+    "terminalAgentMessage": true,
+    "requiredOutputEvidencePresent": true,
+    "errorEvents": 0,
+    "turnFailedEvents": 0,
+    "itemErrorEvents": 0,
+    "failedCommandEvents": 0,
+    "fileChangeCount": 1,
+    "mutationExpected": true,
+    "refusal": null
+  },
+  "usage": {
+    "providerInput": 1200,
+    "cacheRead": 800,
+    "cacheWrite": 0,
+    "output": 200,
+    "reasoning": 50,
+    "fresh": 600,
+    "totalTraffic": 1400,
+    "billedCost": null,
+    "billedCostStatus": "unavailable"
+  }
+}

package/evals/fixtures/benchmark-independent-verification-plan.invalid.json

@@ -0,0 +1,7 @@
+{
+  "schemaVersion": 1,
+  "kind": "benchmark-independent-verification-plan-v1",
+  "suiteDigest": "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+  "verifierDigest": "bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb",
+  "scenarios": []
+}

package/evals/fixtures/benchmark-independent-verification-plan.valid.json

@@ -0,0 +1,21 @@
+{
+  "schemaVersion": 1,
+  "kind": "benchmark-independent-verification-plan-v1",
+  "suiteDigest": "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+  "verifierDigest": "bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb",
+  "scenarios": [{
+    "scenarioId": "identity",
+    "plans": [{
+      "schemaVersion": 1,
+      "operatorRequestDigest": "operator-request-v1:cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc",
+      "backend": { "imageId": "sha256:dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd", "dockerSocket": "/fixture-only/engine.sock", "timeoutMs": 1000 },
+      "contracts": [{
+        "criterionId": "identity-contract",
+        "criterionHash": "eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee",
+        "sourcePath": "src/identity.js", "exportName": "identity", "maxAttempts": 2,
+        "checks": [{ "id": "identity", "cases": [{ "id": "one", "args": [{ "type": "number", "value": 1 }],
+          "expected": { "outcome": "return", "value": { "type": "number", "value": 1 } } }] }]
+      }]
+    }]
+  }]
+}

package/evals/fixtures/codex-relative-efficiency-v1.invalid.json

@@ -0,0 +1,4 @@
+{
+  "schemaVersion": 2,
+  "passed": true
+}

package/evals/fixtures/codex-relative-efficiency-v1.valid.json

@@ -0,0 +1,94 @@
+{
+  "schemaVersion": 1,
+  "policy": {
+    "schemaVersion": 1,
+    "id": "codex-relative-efficiency-v1",
+    "baselineSurface": "codex-cli",
+    "candidateSurface": "piagent",
+    "estimand": "fixed-workload-family-geometric-mean",
+    "outcomeConditioning": "none",
+    "attemptPolicy": "accepted-plus-exact-provider-started-failed-attempts",
+    "maximumTotalTokenTrafficRatio": 0.7,
+    "maximumTotalTokenTrafficRatioUpper95": 0.7,
+    "maximumApiEquivalentCostRatio": 0.7,
+    "maximumApiEquivalentCostRatioUpper95": 0.7,
+    "maximumSubagentSessionsPerAttempt": 1,
+    "maximumSubagentTrafficShare": 0.05
+  },
+  "required": true,
+  "claim": "Piagent uses at least 30% less total provider traffic and API-equivalent cost than Codex CLI on the same work.",
+  "billedOrSubscriptionCost": false,
+  "costBoundary": "API-equivalent text-token pricing; not OAuth subscription debiting",
+  "checks": { "same-workload": true, "quality-noninferior": true },
+  "failures": [],
+  "passed": true,
+  "families": [
+    {
+      "scenarioId": "backend-change",
+      "expectedAttempts": 3,
+      "pairedAttempts": 3,
+      "exact": true,
+      "issues": [],
+      "baselineTokenTraffic": 300000,
+      "candidateTokenTraffic": 180000,
+      "totalTokenTrafficRatio": 0.6,
+      "baselineApiEquivalentCostUsd": 0.3,
+      "candidateApiEquivalentCostUsd": 0.18,
+      "apiEquivalentCostRatio": 0.6,
+      "candidateSubagentSessions": 0,
+      "candidateSubagentTraffic": 0,
+      "maximumSubagentSessionsPerAttempt": 0,
+      "candidateAttempts": 3,
+      "subagentEvidenceComplete": true
+    },
+    {
+      "scenarioId": "frontend-change",
+      "expectedAttempts": 3,
+      "pairedAttempts": 3,
+      "exact": true,
+      "issues": [],
+      "baselineTokenTraffic": 300000,
+      "candidateTokenTraffic": 180000,
+      "totalTokenTrafficRatio": 0.6,
+      "baselineApiEquivalentCostUsd": 0.3,
+      "candidateApiEquivalentCostUsd": 0.18,
+      "apiEquivalentCostRatio": 0.6,
+      "candidateSubagentSessions": 0,
+      "candidateSubagentTraffic": 0,
+      "maximumSubagentSessionsPerAttempt": 0,
+      "candidateAttempts": 3,
+      "subagentEvidenceComplete": true
+    }
+  ],
+  "totalTokenTraffic": {
+    "complete": true,
+    "expectedScenarioFamilies": 2,
+    "completeScenarioFamilies": 2,
+    "ratio": 0.6,
+    "ratioConfidence95": { "lower": 0.6, "upper": 0.6, "sampleUnit": "scenario-family", "scenarioCount": 2 },
+    "ratioConfidence95Raw": { "lower": 0.6, "upper": 0.6, "sampleUnit": "scenario-family", "scenarioCount": 2 },
+    "aggregateBaselineTokens": 600000,
+    "aggregateCandidateTokens": 360000,
+    "aggregateRatio": 0.6,
+    "deltaPercent": -40
+  },
+  "apiEquivalentCost": {
+    "complete": true,
+    "expectedScenarioFamilies": 2,
+    "completeScenarioFamilies": 2,
+    "ratio": 0.6,
+    "ratioConfidence95": { "lower": 0.6, "upper": 0.6, "sampleUnit": "scenario-family", "scenarioCount": 2 },
+    "ratioConfidence95Raw": { "lower": 0.6, "upper": 0.6, "sampleUnit": "scenario-family", "scenarioCount": 2 },
+    "aggregateBaselineUsd": 0.6,
+    "aggregateCandidateUsd": 0.36,
+    "aggregateRatio": 0.6,
+    "deltaPercent": -40
+  },
+  "subagents": {
+    "candidateAttempts": 6,
+    "sessions": 0,
+    "tokenTraffic": 0,
+    "trafficShare": 0,
+    "maximumObservedSessionsPerAttempt": 0
+  }
+}

package/evals/fixtures/composite-criterion.invalid.json

@@ -0,0 +1,10 @@
+{
+  "route": "composite",
+  "compositeContractVersion": "composite-criterion-v1",
+  "criterionIndex": -1,
+  "criterionId": "fixture",
+  "criterionText": "x",
+  "criterionHash": "2d711642b726b04401627ca9fbac32f5c8530fb1903cc4db02258717921a4881",
+  "coverage": [],
+  "facts": []
+}

package/evals/fixtures/composite-criterion.valid.json

@@ -0,0 +1,33 @@
+{
+  "route": "composite",
+  "compositeContractVersion": "composite-criterion-v1",
+  "criterionIndex": 0,
+  "criterionId": "fixture",
+  "criterionText": "x",
+  "criterionHash": "2d711642b726b04401627ca9fbac32f5c8530fb1903cc4db02258717921a4881",
+  "coverage": [
+    {
+      "id": "whole",
+      "startByte": 0,
+      "endByte": 1,
+      "textHash": "2d711642b726b04401627ca9fbac32f5c8530fb1903cc4db02258717921a4881",
+      "roles": ["behavior"],
+      "requiredFactIds": ["code"]
+    }
+  ],
+  "facts": [
+    {
+      "id": "code",
+      "kind": "bounded-code-checks",
+      "specDigest": "aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+      "producerId": "producer",
+      "producerDigest": "bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb",
+      "ruleDigest": "cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc",
+      "stage": "content",
+      "parameters": {
+        "codePlanDigest": "dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd",
+        "backendProfileDigest": "eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee"
+      }
+    }
+  ]
+}

package/evals/fixtures/contract-family-library.invalid.json

@@ -0,0 +1,42 @@
+{
+  "schemaVersion": 1,
+  "families": [
+    {
+      "id": "literal-identity",
+      "version": 1,
+      "domain": "pure-function",
+      "description": "Return the declared bounded argument unchanged.",
+      "parameters": {
+        "call": "export",
+        "argument": "value"
+      },
+      "template": {
+        "exportName": {
+          "$parameter": "call"
+        },
+        "checks": [
+          {
+            "id": "identity",
+            "cases": [
+              {
+                "id": "literal",
+                "args": [
+                  {
+                    "$parameter": "argument"
+                  }
+                ],
+                "expected": {
+                  "outcome": "return",
+                  "value": {
+                    "$parameter": "argument"
+                  }
+                }
+              }
+            ]
+          }
+        ],
+        "source": "export const unreviewed = true;"
+      }
+    }
+  ]
+}

package/evals/fixtures/contract-family-library.valid.json

@@ -0,0 +1,41 @@
+{
+  "schemaVersion": 1,
+  "families": [
+    {
+      "id": "literal-identity",
+      "version": 1,
+      "domain": "pure-function",
+      "description": "Return the declared bounded argument unchanged.",
+      "parameters": {
+        "call": "export",
+        "argument": "value"
+      },
+      "template": {
+        "exportName": {
+          "$parameter": "call"
+        },
+        "checks": [
+          {
+            "id": "identity",
+            "cases": [
+              {
+                "id": "literal",
+                "args": [
+                  {
+                    "$parameter": "argument"
+                  }
+                ],
+                "expected": {
+                  "outcome": "return",
+                  "value": {
+                    "$parameter": "argument"
+                  }
+                }
+              }
+            ]
+          }
+        ]
+      }
+    }
+  ]
+}

package/evals/fixtures/contract-selection-recipe.invalid.json

@@ -0,0 +1,29 @@
+{
+  "schemaVersion": 1,
+  "backend": {
+    "imageId": "sha256:bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb",
+    "dockerSocket": "/fixture-only/engine.sock",
+    "timeoutMs": 1000
+  },
+  "selections": [
+    {
+      "criterion": {
+        "text": "Return the declared bounded argument unchanged.",
+        "obligation": "requested-behavior"
+      },
+      "family": {
+        "id": "literal-identity",
+        "version": 1
+      },
+      "parameters": {
+        "call": "identity",
+        "argument": {
+          "type": "number",
+          "value": 7
+        }
+      },
+      "sourcePath": "../escape.js",
+      "maxAttempts": 2
+    }
+  ]
+}

package/evals/fixtures/contract-selection-recipe.valid.json

@@ -0,0 +1,29 @@
+{
+  "schemaVersion": 1,
+  "backend": {
+    "imageId": "sha256:bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb",
+    "dockerSocket": "/fixture-only/engine.sock",
+    "timeoutMs": 1000
+  },
+  "selections": [
+    {
+      "criterion": {
+        "text": "Return the declared bounded argument unchanged.",
+        "obligation": "requested-behavior"
+      },
+      "family": {
+        "id": "literal-identity",
+        "version": 1
+      },
+      "parameters": {
+        "call": "identity",
+        "argument": {
+          "type": "number",
+          "value": 7
+        }
+      },
+      "sourcePath": "src/identity.js",
+      "maxAttempts": 2
+    }
+  ]
+}

package/evals/fixtures/helper-request.invalid.json

@@ -1 +1 @@
-{"schemaVersion":1,"policyVersion":"role-policy-v1","role":"reviewer","objectiveHash":"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa","objectiveText":"Review.","taskId":"task-1","taskRunId":"task-1-run-1","sessionHash":"bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb","authority":"read-only","readScope":["**"],"writeScope":["src/**"],"allowedTools":["read","edit"],"model":null,"contextBudget":6000,"ceilings":{"timeSeconds":600,"calls":40,"retries":0},"outputSchema":"review-v1","stoppingRule":"Stop.","approvalRestrictions":[],"parentAuthorityDigest":"cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc","singleWriterOwnership":null,"deduplicationKey":"dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd"}
+{"schemaVersion":2,"policyVersion":"role-policy-v1","role":"reviewer","objectiveHash":"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa","objectiveText":"Review.","taskId":"task-1","taskRunId":"task-1-run-1","sessionHash":"bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb","authority":"read-only","readScope":["**"],"writeScope":["src/**"],"allowedTools":["read","edit"],"model":null,"contextBudget":6000,"ceilings":{"timeSeconds":600,"calls":40,"retries":0},"outputSchema":"review-v1","stoppingRule":"Stop.","approvalRestrictions":[],"parentAuthorityDigest":"cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc","singleWriterOwnership":null,"deduplicationKey":"dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd","contextTransfer":{"mode":"isolated-minimal","inheritParentHistory":false,"maxSeedTokens":2048,"estimatedSeedTokens":101}}

package/evals/fixtures/helper-request.valid.json

@@ -1 +1 @@
-{"schemaVersion":1,"policyVersion":"role-policy-v1","role":"scout","objectiveHash":"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa","objectiveText":"Map the bounded source entry points.","taskId":"task-1","taskRunId":"task-1-run-1","sessionHash":"bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb","authority":"read-only","readScope":["src/**"],"writeScope":[],"allowedTools":["read","grep","find","ls"],"model":null,"contextBudget":6000,"ceilings":{"timeSeconds":600,"calls":40,"retries":0},"outputSchema":"scout-result-v1","stoppingRule":"Stop when bounded evidence is sufficient.","approvalRestrictions":["no-external-write"],"parentAuthorityDigest":"cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc","singleWriterOwnership":null,"deduplicationKey":"dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd"}
+{"schemaVersion":2,"policyVersion":"role-policy-v1","role":"scout","objectiveHash":"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa","objectiveText":"Map the bounded source entry points.","taskId":"task-1","taskRunId":"task-1-run-1","sessionHash":"bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb","authority":"read-only","readScope":["src/**"],"writeScope":[],"allowedTools":["read","grep","find","ls"],"model":null,"contextBudget":6000,"ceilings":{"timeSeconds":600,"calls":40,"retries":0},"outputSchema":"scout-result-v1","stoppingRule":"Stop when bounded evidence is sufficient.","approvalRestrictions":["no-external-write"],"parentAuthorityDigest":"cccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc","singleWriterOwnership":null,"deduplicationKey":"dddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddddd","contextTransfer":{"mode":"isolated-minimal","inheritParentHistory":false,"maxSeedTokens":2048,"estimatedSeedTokens":125}}

package/evals/fixtures/host-contract-plan.invalid.json

@@ -0,0 +1,42 @@
+{
+  "schemaVersion": 1,
+  "operatorRequestDigest": "operator-request-v1:aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa",
+  "backend": {
+    "imageId": "sha256:bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb",
+    "dockerSocket": "/fixture-only/engine.sock",
+    "timeoutMs": 1000
+  },
+  "contracts": [
+    {
+      "criterionId": "identity-contract",
+      "criterionHash": "eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee",
+      "sourcePath": "src/identity.js",
+      "exportName": "identity",
+      "maxAttempts": 2,
+      "checks": [
+        {
+          "id": "identity",
+          "cases": [
+            {
+              "id": "positive-number",
+              "args": [
+                {
+                  "type": "number",
+                  "value": 7
+                }
+              ],
+              "expected": {
+                "outcome": "return",
+                "value": {
+                  "type": "number",
+                  "value": 7
+                }
+              }
+            }
+          ]
+        }
+      ]
+    }
+  ],
+  "approved": true
+}