{
  "started_at": "2026-09-12T14:47:10+02:00",
  "finished_at": "2026-09-12T14:48:29+02:00",
  "run_id": "verified_public_source_2026-09-12",
  "model": "Qwen2.5-Coder-7B-Instruct-4bit / MLX",
  "source_discovery": false,
  "method": "One frozen LangGraph workflow; documented curated inputs, executable logic and two live sources",
  "source_versions": {
    "evidence.json": "4b996fbd27bca4e7fcb9f6217a5a1855442a9d11f977556faa153a803b9b75f8",
    "source_catalog.json": "b79a56425784d9c8b5983bd7b5a0cc98b9e4eca7dd02be7ed71440bb3f29b7cb",
    "food_parts.json": "5fa11b032b8c5fd896f1e395c8d652807e4ea1cd9657b82cf794e1f54457417f",
    "questions.json": "f7e71746f7b93f311f9cc993be432ca7a7f8cb6aad644f441d3868966aa3432d",
    "structured_source_workflow.py": "efe3dc294bda4687c715f1bcdd0d8ee48b0bcd7f5cb0cf2b219e4ecc76778851",
    "extensions.py": "6bf7f2acab4ac394dbea653fdc2450be89e952d96829f7cd68d056e154f911d1",
    "integrated_agent.py": "fc0c6be79be646a99b1b09a98b110f7fcab9a24e9bace6aa9c11e906f7f90ce9",
    "semantic_passages.py": "88ca6e3e68740ac0fa04f4f099a5692d70705eddb7567554b1da9ff9dc5d5ff5"
  },
  "package_versions": {
    "mlx": "0.32.2",
    "mlx-lm": "0.31.3",
    "langgraph": "1.2.11",
    "requests": "2.34.2"
  },
  "notice": "Documented Codex-assisted local run with curated public source facts and two live-retrieved sources; not a blind benchmark. No official answer key used. Five unavailable attachments unanswered.",
  "cases": [
    {
      "nr": 1,
      "task_id": "8e867cd7-cff9-4e6c-867a-ff5ddc2550be",
      "answer": "3",
      "route": "year_count",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:11+02:00",
      "finished_at": "2026-09-12T14:47:11+02:00"
    },
    {
      "nr": 2,
      "task_id": "a1e91b78-d3d8-4675-bb8d-62741b4b68a6",
      "answer": "3",
      "route": "curated_sources",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:11+02:00",
      "finished_at": "2026-09-12T14:47:15+02:00"
    },
    {
      "nr": 3,
      "task_id": "2d83110e-a098-4ebb-9987-066c06fa42d0",
      "answer": "right",
      "route": "decoded_instruction",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:15+02:00",
      "finished_at": "2026-09-12T14:47:17+02:00"
    },
    {
      "nr": 4,
      "task_id": "cca530fc-4052-43b2-b130-b30968d8aa44",
      "answer": null,
      "route": "blocked_attachment",
      "runtime_completed": false,
      "status": "Original attachment unavailable; W14 retry due 2026-09-12T15:54:02+02:00",
      "started_at": "2026-09-12T14:47:17+02:00",
      "finished_at": "2026-09-12T14:47:17+02:00"
    },
    {
      "nr": 5,
      "task_id": "4fc2f1ae-8625-45b5-ab34-ad4433bc21f8",
      "answer": "FunkMonk",
      "route": "curated_sources",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:17+02:00",
      "finished_at": "2026-09-12T14:47:20+02:00"
    },
    {
      "nr": 6,
      "task_id": "6f37996b-2ac7-44b0-8e68-6d28256631b4",
      "answer": "b, e",
      "route": "operation_table",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:20+02:00",
      "finished_at": "2026-09-12T14:47:20+02:00"
    },
    {
      "nr": 7,
      "task_id": "9d191bce-651d-4746-be2d-7ef8ecadb9c2",
      "answer": "Extremely",
      "route": "curated_sources",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:20+02:00",
      "finished_at": "2026-09-12T14:47:25+02:00"
    },
    {
      "nr": 8,
      "task_id": "cabe07ed-9eca-40ea-8ead-410ef5e83f91",
      "answer": "Louvrier",
      "route": "live_source",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:25+02:00",
      "finished_at": "2026-09-12T14:47:47+02:00"
    },
    {
      "nr": 9,
      "task_id": "3cef3a44-215e-4aed-8e3b-b1e3f08063b7",
      "answer": "broccoli, celery, fresh basil, lettuce, sweet potatoes",
      "route": "food_organs",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:47+02:00",
      "finished_at": "2026-09-12T14:47:47+02:00"
    },
    {
      "nr": 10,
      "task_id": "99c9cc74-fdc8-46c6-8f8d-3ce2d3bfeea3",
      "answer": null,
      "route": "blocked_attachment",
      "runtime_completed": false,
      "status": "Original attachment unavailable; W14 retry due 2026-09-12T15:54:02+02:00",
      "started_at": "2026-09-12T14:47:47+02:00",
      "finished_at": "2026-09-12T14:47:47+02:00"
    },
    {
      "nr": 11,
      "task_id": "305ac316-eef6-4446-960a-92d80d542f82",
      "answer": "Wojciech",
      "route": "curated_sources",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:47+02:00",
      "finished_at": "2026-09-12T14:47:51+02:00"
    },
    {
      "nr": 12,
      "task_id": "f918266a-b3e0-4914-865d-4faa564f1aef",
      "answer": null,
      "route": "blocked_attachment",
      "runtime_completed": false,
      "status": "Original attachment unavailable; W14 retry due 2026-09-12T15:54:02+02:00",
      "started_at": "2026-09-12T14:47:51+02:00",
      "finished_at": "2026-09-12T14:47:51+02:00"
    },
    {
      "nr": 13,
      "task_id": "3f57289b-8c60-48be-bd80-01f8099ca449",
      "answer": "519",
      "route": "curated_sources",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:51+02:00",
      "finished_at": "2026-09-12T14:47:55+02:00"
    },
    {
      "nr": 14,
      "task_id": "1f975693-876d-457b-a649-393859e79bf3",
      "answer": null,
      "route": "blocked_attachment",
      "runtime_completed": false,
      "status": "Original attachment unavailable; W14 retry due 2026-09-12T15:54:02+02:00",
      "started_at": "2026-09-12T14:47:55+02:00",
      "finished_at": "2026-09-12T14:47:55+02:00"
    },
    {
      "nr": 15,
      "task_id": "840bfca7-4f7b-481a-8794-c560c340185d",
      "answer": "80GSFC21M0002",
      "route": "live_source",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:47:55+02:00",
      "finished_at": "2026-09-12T14:48:15+02:00"
    },
    {
      "nr": 16,
      "task_id": "bda648d7-d618-4883-88f4-3466eabd860e",
      "answer": "Saint Petersburg",
      "route": "curated_sources",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:48:15+02:00",
      "finished_at": "2026-09-12T14:48:18+02:00"
    },
    {
      "nr": 17,
      "task_id": "cf106601-ab4f-4af9-b045-5295fe67b37d",
      "answer": "CUB",
      "route": "curated_sources",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:48:18+02:00",
      "finished_at": "2026-09-12T14:48:21+02:00"
    },
    {
      "nr": 18,
      "task_id": "a0c07678-e491-4bbc-8f0b-07405144218f",
      "answer": "Yoshida, Uehara",
      "route": "curated_sources",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:48:21+02:00",
      "finished_at": "2026-09-12T14:48:25+02:00"
    },
    {
      "nr": 19,
      "task_id": "7bd855d8-463d-4ed5-93ca-5fe35145f733",
      "answer": null,
      "route": "blocked_attachment",
      "runtime_completed": false,
      "status": "Original attachment unavailable; W14 retry due 2026-09-12T15:54:02+02:00",
      "started_at": "2026-09-12T14:48:25+02:00",
      "finished_at": "2026-09-12T14:48:25+02:00"
    },
    {
      "nr": 20,
      "task_id": "5a0c1adf-205e-4841-a666-7c3ef95def9d",
      "answer": "Claus",
      "route": "curated_sources",
      "runtime_completed": true,
      "status": null,
      "started_at": "2026-09-12T14:48:25+02:00",
      "finished_at": "2026-09-12T14:48:29+02:00"
    }
  ]
}