{
  "schema": "agent-rerun-receipt.v1",
  "record": {
    "id": "native-claude-blueprint-prose-scope",
    "taskId": "blueprint-prose-scope",
    "taskTitle": "Prose PRD: CSV import with exclusions and acceptance criteria",
    "agent": "Claude Code",
    "requestedModel": "claude-sonnet-5-5",
    "resolvedModel": "claude-sonnet-5-5",
    "wallSeconds": 8.677,
    "inputTokens": 42355,
    "outputTokens": 753,
    "totalTokens": 43108,
    "costUsd": null,
    "energyWh": null,
    "assertionsPassed": 10,
    "assertionsTotal": 10,
    "observedNativeModel": "claude-sonnet-5-5",
    "modelIdentityStatus": "MATCHED_NATIVE_RECEIPT",
    "modelIdentityEvidence": "native exporter modelUsage canonical model",
    "providerResponseModel": null,
    "reasoningEffort": null,
    "family": "blueprint",
    "status": "PASS",
    "tier": "native subscription",
    "diagnosticOnly": false,
    "comparisonEligible": null,
    "cohort": "original-model",
    "resultNote": null,
    "limitations": [
      "Auth uses existing native subscription/prepaid account; billing decrement unavailable.",
      "Native models differ: this is a product pilot, not a same-model comparison.",
      "Synthetic authored fixture; no statistical ranking.",
      "Planning structure rubric only; exact task count and exclusion text checks require semantic review. Not code execution quality.",
      "Custom pilot receipt has not been admitted through BenchmarkResult to JSON/DuckDB/Neon; no leaderboard promotion."
    ],
    "evidenceFile": "native-claude-blueprint-prose-scope.json"
  },
  "sourceRecord": {
    "id": "native-claude-blueprint-prose-scope",
    "family": "blueprint-planning",
    "taskId": "blueprint-prose-scope",
    "taskTitle": "Prose PRD: CSV import with exclusions and acceptance criteria",
    "agent": "Claude Code",
    "requestedModel": "claude-sonnet-5-5",
    "resolvedModel": "claude-sonnet-5-5",
    "status": "PASS",
    "wallSeconds": 8.677,
    "inputTokens": 42355,
    "outputTokens": 753,
    "totalTokens": 43108,
    "costUsd": null,
    "energyWh": null,
    "assertionsPassed": 10,
    "assertionsTotal": 10,
    "behaviorPassed": 8,
    "behaviorTotal": 11,
    "scopePassed": true,
    "changedPaths": [],
    "addedPaths": [
      "plan.json"
    ],
    "processExitCode": 0,
    "oracleExitCode": 1,
    "evidencePaths": [
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/command.json",
        "sha256": "70d5ba29e3c0b203fc1d88af52c9b0ea4f39b4d4d12474cc77194ceec052ce93"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/freeze.json",
        "sha256": "4f3f3a10b4539f8b512a7c9ab24e8f52630a8c0914aaf94c8015ec35617f3fc2"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/oracle.stderr.log",
        "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/oracle.stdout.log",
        "sha256": "7ebc1ed28e0f6c46d94c7977f2200fa2c905ac76c58ec88c00a1061161184d52"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/prompt.txt",
        "sha256": "6fa70e7b4440527c999b6d415c3115c484748276488d5013ba8d2b71a2b1ab46"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/sandbox.sb",
        "sha256": "6b1c843c9d77219c6f4c3f1d3f17a5cd17693c224fccd5aad76669ff50e14ab4"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/semantic-checks.json",
        "sha256": "60291ca416605988322a996c8b5abc0e3f7d78d7d6b92a8340c6df80879c50ad"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/stderr.log",
        "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/stdout.log",
        "sha256": "3324f473764041f007f151156b54019a31767cd886796642e65a31c203a3fcf3"
      },
      {
        "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/workspace/plan.json",
        "sha256": "9e5a6eb6886c181195cc9e9b1e30910408403a3f62ee502e6e3e8514de3d0462"
      }
    ],
    "limitations": [
      "Auth uses existing native subscription/prepaid account; billing decrement unavailable.",
      "Native models differ: this is a product pilot, not a same-model comparison.",
      "Synthetic authored fixture; no statistical ranking.",
      "Planning structure rubric only; exact task count and exclusion text checks require semantic review. Not code execution quality.",
      "Custom pilot receipt has not been admitted through BenchmarkResult to JSON/DuckDB/Neon; no leaderboard promotion."
    ],
    "cachedInputTokens": 31762,
    "reportedApiEquivalentUsd": 0.053070200000000005,
    "observedNativeModel": "claude-sonnet-5-5",
    "modelIdentityEvidence": "native exporter modelUsage canonical model",
    "threeStoreAdmission": "NOT_VERIFIED",
    "admitted": false,
    "tier": "native subscription",
    "modelIdentityStatus": "MATCHED_NATIVE_RECEIPT",
    "acceptanceStatus": "PASS",
    "rawStructuralChecksPassed": 8,
    "rawStructuralChecksTotal": 11,
    "semanticChecksPassed": 10,
    "semanticChecksTotal": 10,
    "semanticChecksUnknown": 0,
    "planningSemanticReview": "Independent requirement/dependency/exclusion rubric; raw exact-count/keyword scores retained as diagnostic only."
  },
  "sourceFile": "artifacts/agent_comparison_rerun_20261016/native/reviewed-summary.json",
  "sourceFileSha256": "d470470c759ba7357132831170430dcd2876add878a4fb767db2fe55d9163a70",
  "files": [
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/command.json",
      "sha256": "70d5ba29e3c0b203fc1d88af52c9b0ea4f39b4d4d12474cc77194ceec052ce93",
      "bytes": 1559
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/freeze.json",
      "sha256": "4f3f3a10b4539f8b512a7c9ab24e8f52630a8c0914aaf94c8015ec35617f3fc2",
      "bytes": 430,
      "content": "{\n  \"frozenAt\": \"2026-10-16T03:59:48.195465+00:00\",\n  \"syntheticInput\": true,\n  \"promptSha256\": \"6fa70e7b4440527c999b6d415c3115c484748276488d5013ba8d2b71a2b1ab46\",\n  \"fixtureContractSha256\": \"6c7c9b41a18ecf622b6f03ca745acad60fe5c261300ef2294e5b6aa95214b21c\",\n  \"fixtureHashes\": {},\n  \"requestedModel\": \"claude-sonnet-5-5\",\n  \"taskId\": \"blueprint-prose-scope\",\n  \"workingDirectory\": \"<temporary-workspace>\"\n}\n",
      "contentRedacted": true,
      "publishedContentSha256": "fb95551c56ab54d5108e63b0f0bddbe9eaea9ca9c3508250de7aae47baff9ad0"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/oracle.stderr.log",
      "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
      "bytes": 0,
      "content": "",
      "contentRedacted": false,
      "publishedContentSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/oracle.stdout.log",
      "sha256": "7ebc1ed28e0f6c46d94c7977f2200fa2c905ac76c58ec88c00a1061161184d52",
      "bytes": 689,
      "content": "[\n  {\n    \"name\": \"covers:csv+parser\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"covers:email+valid\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"covers:summary+count\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"covers:test+csv\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"covers:test+duplicate\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"excludes:salesforce\",\n    \"passed\": false\n  },\n  {\n    \"name\": \"excludes:cryptocurrency\",\n    \"passed\": false\n  },\n  {\n    \"name\": \"exact_task_count\",\n    \"passed\": false\n  },\n  {\n    \"name\": \"unique_nonempty_ids\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"no_dangling_dependencies\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"acyclic\",\n    \"passed\": true\n  }\n]\n",
      "contentRedacted": false,
      "publishedContentSha256": "7ebc1ed28e0f6c46d94c7977f2200fa2c905ac76c58ec88c00a1061161184d52"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/prompt.txt",
      "sha256": "6fa70e7b4440527c999b6d415c3115c484748276488d5013ba8d2b71a2b1ab46",
      "bytes": 1006,
      "content": "Convert the provided PRD into a work breakdown plan. Return only JSON with a tasks array. Each task has task_id (T-001 etc in document order), name, description, and deps (task_id strings). Include delivery and acceptance work. Honor exclusions; risks are context, not deliverables. Keep explicit dependencies and make a valid acyclic dependency graph. Do not implement code or predict measurements.\n\n# TEST Benchmark Contact Import\n\n## Scope\nImplement a CSV parser that reads UTF-8 contact files with name and email columns.\n\nBuild an email validator that rejects rows without an at sign or with a blank name.\n\nCreate an import summary reporting accepted and rejected row counts.\n\n## Acceptance Criteria\n- Test empty CSV files and quoted commas.\n- Test that duplicate email rows are reported consistently.\n\n## Non-goals\n- Build a Salesforce synchronization connector.\n- Implement cryptocurrency billing.\n\n## Risks\n- Malformed CSV input may contain unclosed quotes.\n\n\nWrite only the JSON plan to plan.json.",
      "contentRedacted": false,
      "publishedContentSha256": "6fa70e7b4440527c999b6d415c3115c484748276488d5013ba8d2b71a2b1ab46"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/sandbox.sb",
      "sha256": "6b1c843c9d77219c6f4c3f1d3f17a5cd17693c224fccd5aad76669ff50e14ab4",
      "bytes": 753
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/semantic-checks.json",
      "sha256": "60291ca416605988322a996c8b5abc0e3f7d78d7d6b92a8340c6df80879c50ad",
      "bytes": 642,
      "content": "[\n  {\n    \"name\": \"covers:csv+parser\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"covers:email+valid\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"covers:summary+count\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"covers:test+csv\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"covers:test+duplicate\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"does_not_deliver:salesforce\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"does_not_deliver:cryptocurrency\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"unique_nonempty_ids\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"no_dangling_dependencies\",\n    \"passed\": true\n  },\n  {\n    \"name\": \"acyclic\",\n    \"passed\": true\n  }\n]\n",
      "contentRedacted": false,
      "publishedContentSha256": "60291ca416605988322a996c8b5abc0e3f7d78d7d6b92a8340c6df80879c50ad"
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/stderr.log",
      "sha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
      "bytes": 0
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/stdout.log",
      "sha256": "3324f473764041f007f151156b54019a31767cd886796642e65a31c203a3fcf3",
      "bytes": 14387
    },
    {
      "path": "artifacts/agent_comparison_rerun_20261016/native/claude-blueprint-prose-scope/workspace/plan.json",
      "sha256": "9e5a6eb6886c181195cc9e9b1e30910408403a3f62ee502e6e3e8514de3d0462",
      "bytes": 1683,
      "content": "{\n  \"tasks\": [\n    {\n      \"task_id\": \"T-001\",\n      \"name\": \"Implement UTF-8 CSV parser\",\n      \"description\": \"Implement a CSV parser that reads UTF-8 contact files with name and email columns, including handling of quoted fields containing commas and malformed input such as unclosed quotes.\",\n      \"deps\": []\n    },\n    {\n      \"task_id\": \"T-002\",\n      \"name\": \"Build email validator\",\n      \"description\": \"Build a validator that rejects rows without an at sign in the email or with a blank name.\",\n      \"deps\": [\"T-001\"]\n    },\n    {\n      \"task_id\": \"T-003\",\n      \"name\": \"Create import summary\",\n      \"description\": \"Create an import summary reporting accepted and rejected row counts, including consistent reporting of duplicate email rows.\",\n      \"deps\": [\"T-001\", \"T-002\"]\n    },\n    {\n      \"task_id\": \"T-004\",\n      \"name\": \"Test empty CSV files and quoted commas\",\n      \"description\": \"Write and run tests covering empty CSV files and quoted fields containing commas (acceptance criterion).\",\n      \"deps\": [\"T-001\", \"T-003\"]\n    },\n    {\n      \"task_id\": \"T-005\",\n      \"name\": \"Test duplicate email reporting consistency\",\n      \"description\": \"Write and run tests verifying that duplicate email rows are reported consistently in the import summary (acceptance criterion).\",\n      \"deps\": [\"T-002\", \"T-003\"]\n    },\n    {\n      \"task_id\": \"T-006\",\n      \"name\": \"Package and deliver contact import\",\n      \"description\": \"Package the parser, validator, and import summary with passing tests for delivery, and confirm acceptance criteria are met. Salesforce synchronization and cryptocurrency billing are excluded.\",\n      \"deps\": [\"T-004\", \"T-005\"]\n    }\n  ]\n}\n",
      "contentRedacted": false,
      "publishedContentSha256": "9e5a6eb6886c181195cc9e9b1e30910408403a3f62ee502e6e3e8514de3d0462"
    }
  ],
  "disclosure": "Synthetic inputs and observed execution. Original-byte hashes retained; selected source and oracle outputs reproduced. Private provider transcripts remain local."
}